Safety of Inferior Vena Cava Filter Retrieval in Anticoagulated Patients

Safety of Inferior Vena Cava Filter Retrieval in Anticoagulated Patients

Original Research PULMONARY VASCULAR DISEASE Safety of Inferior Vena Cava Filter Retrieval in Anticoagulated Patients* Hanno Hoppe, MD; John A. Kaufm...

89KB Sizes 0 Downloads 7 Views

Original Research PULMONARY VASCULAR DISEASE

Safety of Inferior Vena Cava Filter Retrieval in Anticoagulated Patients* Hanno Hoppe, MD; John A. Kaufman, MD; Robert E. Barton, MD; Bryan D. Petersen, MD; Paul C. Lakin, MD; Thomas G. DeLoughery, MD; Zubin Irani, MD; Kivilcim Yavuz, MD; and Frederick S. Keller, MD

Purpose: To evaluate the safety of inferior vena cava (IVC) filter retrieval in therapeutically anticoagulated patients in comparison to prophylactically or not therapeutically anticoagulated patients with respect to retrieval-related hemorrhagic complications. Materials and methods: This was a retrospective study of 115 consecutive attempted IVC filter retrievals in 110 patients. Filter retrievals were stratified as performed in patients who were therapeutically anticoagulated (group 1), prophylactically anticoagulated (group 2), or not therapeutically anticoagulated (group 3). The collected data included anticoagulant and antiplatelet medications (type, form and duration of administration, dosage) at the time of retrieval. Phone interviews and chart review was performed for the international normalized ratio (INR), activated partial thromboplastin time, platelet count, infusion of blood products, and retrievalrelated hemorrhagic complications. Results: Group 1 included 65 attempted filter retrievals in 61 therapeutically anticoagulated patients by measured INR or dosing when receiving low-molecular-weight heparin (LMWH). Four retrievals were not successful. In patients receiving oral anticoagulation, the median INR was 2.35 (range, 2 to 8). Group 2 comprised 23 successful filter retrievals in 22 patients receiving a prophylactic dose of LMWH. Group 3 included 27 attempted filter retrievals in 27 patients not receiving therapeutic anticoagulation. Six retrievals were not successful. Five patients were receiving oral anticoagulation with a subtherapeutic INR (median, 1.49; range, 1.16 to 1.69). No anticoagulation medication was administered in 22 patients. In none of the groups were hemorrhagic complications related to the retrieval procedures identified. Conclusions: These results suggest that retrieval of vena cava filters in anticoagulated patients is safe. Interruption or reversal of anticoagulation for the retrieval of vena cava filters is not indicated. (CHEST 2007; 132:31–36) Key words: pulmonary embolism; vena cava filters; venous thrombosis Abbreviations: aPTT ⫽ activated partial thromboplastin time; DVT ⫽ deep venous thrombosis; FFP ⫽ fresh-frozen plasma; INR ⫽ international normalized ratio; IVC ⫽ inferior vena cava; LMWH ⫽ low-molecular-weight heparin; PE ⫽ pulmonary embolism; VTE ⫽ established venous thromboembolism

optimal therapy for patients with established T hevenous thromboembolism (VTE) is anticoagulation.1 When anticoagulation is contraindicated or *From the Dotter Interventional Institute (Drs. Hoppe, Kaufman, Barton, Lakin, Irani, Yavuz, and Keller) and the Division of Hematology and Medical Oncology (Dr. DeLoughery), Oregon Health and Science University; and Department of Angiography (Dr. Petersen), Portland Veterans Administration Medical Center, Portland, OR. The authors have no conflicts of interest to disclose, except for Dr. Kaufman, who is a consultant and provides research support for Cook Medical, Inc. Dr. Hoppe was supported by the Swiss National Foundation, Novartis Foundation Switzerland, and the Swiss Society of Radiology. www.chestjournal.org

ineffective, inferior vena cava (IVC) filter placement is an accepted measure to reduce the risk of pulmonary embolism (PE).2 Although data are limited, IVC filters are believed to have certain long-term risks Manuscript received December 9, 2006; revision accepted March 9, 2007. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Hanno Hoppe, MD, Dotter Interventional Institute, Oregon Health and Science University, L-605, 3181 SW Sam Jackson Park Rd, Portland, OR 97239-3098; e-mail: [email protected] DOI: 10.1378/chest.06-2897 CHEST / 132 / 1 / JULY, 2007

31

such as IVC occlusion and recurrent deep venous thrombosis (DVT).3–5 In some patients with VTE, the period of time during which a filter is necessary for protection against PE is limited.6 – 8 In these patients, anticoagulant therapy should be resumed as soon as the risk of a complication from anticoagulation has abated because the filter has no impact on the treatment of existing PE or DVT.1 Optional IVC filters can be used as permanent devices but also allow percutaneous removal. In patients with documented VTE, the current recommendations are that filter removal should not occur until the patient is adequately treated with anticoagulant medication, based on expert consensus but no data, and the risk of PE is acceptably low.8 Achievement of therapeutic anticoagulation, especially with warfarin, can take several weeks in some patients.1 Interruption of anticoagulation in these patients in order to retrieve a filter may introduce added cost and risk. Although published clinical studies6,9 –12 have shown that IVC filter retrieval is safe, the specific question of the safety of filter retrieval in therapeutically anticoagulated patients has not been previously addressed. The anticoagulant-associated risks of filter retrieval are primarily hemorrhage at the filter implantation site in the IVC and at the percutaneous venous access site.11 Expert consensus recommends that patients with VTE remain fully anticoagulated, but objective data have not been reported. Based on expert consensus, we hypothesized that IVC filter retrieval can be safely performed in therapeutically anticoagulated patients. The purpose of this retrospective study was to evaluate the safety of IVC filter retrieval in therapeutically anticoagulated patients in comparison to prophylactically or not therapeutically anticoagulated patients in terms of retrieval-related immediate or long-term hemorrhagic complications.

proximal DVT documented by an objective test such as duplex sonography, CT, contrast venography, or magnetic resonance venography. Similarly, all patients with PE had a positive objective test finding, including radionuclide ventilation/perfusion scans, pulmonary angiography, contrast-enhanced spiral CT, or gadolinium enhanced magnetic resonance angiography. The indications for filter placement were recorded according to the recommended reporting standards (Table 1).13,14 All filters were placed in an angiography suite by interventional radiologists using ultrasound-guided venous access to avoid inadvertent puncture of the carotid or femoral arteries. The filter choice was at the discretion of the interventional radiologist. A vena cavogram was obtained before and after filter placement with use of iodinated contrast material (iodixanol, 320 mgI/mL) [Visipaque; Amersham Health; Princeton, NJ]. Anticoagulation at the Time of Filter Retrieval Three groups of patients were defined. Group 1 comprised therapeutically anticoagulated patients including those receiving warfarin with a therapeutic international normalized ratio (INR) ⱖ 2, therapeutic doses of low-molecular-weight heparin (LMWH), and receiving a combination of warfarin and therapeutic LMWH. Group 2 included patients receiving prophylactic anticoagulation with LMWH. Group 3 included patients without anticoagulation and patients receiving warfarin with a subtherapeutic INR ⬍ 2.0. Patient anticoagulation status at the time of attempted IVC filter retrieval was determined by recording hematologic coagulation parameters or dosing when receiving LMWH. Type, form, duration of administration, and dosage of anticoagulation medications were recorded. The results of hematologic studies (activated partial thromboplastin time [aPTT], INR, and platelet count) obtained within 24 h prior to IVC filter retrieval were collected. An aPTT between 70 s and 110 s was considered therapeutic. A platelet count of ⬍ 50 ⫻ 109/L was defined as low. Patients receiving subcutaneous LMWH were treated according to a standardized institutional dosing protocol. In VTE patients with normal renal function, the therapeutic dose was 1 mg/kg per dose per 12 h, or 1.5 mg/kg per dose per 24 h. In patients without VTE, the prophylactic dose was 0.5 mg/kg per dose per 24 h. For the purposes of this study, patients receiving LMWH were considered anticoagulated after a minimum of 5 days of treatment. The standard of care in the hospital was not to monitor anticoagulation with plasma tests in patients treated solely with LMWH. In patients with VTE, warfarin was started as soon as possible with a therapeutic target INR ⱖ 2.0. Warfarin was usually

Materials and Methods Study Population This retrospective single institution study was approved by the Institutional Review Board and was performed according to the Health Insurance Portability and Accountability Act of 1996. One hundred fifteen consecutive attempted IVC filter retrievals in 110 consecutive patients (71 women, 39 men) between July 2001 and August 2006 were evaluated. The mean (⫾ SD) patient age was 52 ⫾ 16 years (range, 12 to 87 years). Filter Placement Informed consent was documented for all filter placements. Patients who were considered at short-term high risk for PE in whom a permanent IVC filter was not required received a retrievable vena cava filter. All patients with DVT had proven 32

Table 1—Indications for 115 IVC Filter Placements Indications Established VTE requiring interruption of anticoagulation for surgery Established VTE with other contraindication to anticoagulation Prophylaxis after trauma (no VTE) Prophylaxis in high-risk patients undergoing surgery (no VTE) Acute VTE with initial inability to achieve anticoagulation Recurrent VTE during early anticoagulation Patient with VTE initially refusing anticoagulation treatment

% (No./Total) 42.6 (49/115) 22.6 (26/115) 21.7 (25/115) 10.4 (12/115) 0.9 (1/115) 0.9 (1/115) 0.9 (1/115)

Original Research

initiated at 5 mg/d and adjusted according to patient response to titrate the INR to a therapeutic target range. IVC Filter Retrieval Informed consent was documented for all 115 filter retrievals in 110 patients. IVC filter retrieval was performed according to the instructions for use for the particular device. In general, ultrasound-guided venous access was obtained. Then, a preretrieval digital subtraction cavogram was acquired through a 5F pigtail catheter positioned inferior to the filter, injecting 50 mL of contrast medium at 25 mL/s for 2 s and filmed with digital subtraction technique in the anterior-posterior projection. Small trapped thrombi that did not extend lower than the joints of the petals and the filter legs were retrieved with the filter. If substantial trapped embolus was discovered, which was defined as filling ⬎ 20% of the filter cone, the filter was left in position and not retrieved. After filter removal, a postretrieval digital subtraction vena cavogram was performed by using a similar technique and contrast injections through the retrieval sheath positioned just inferior to the filter implantation site. The duration of filter dwell time was calculated for each patient. The route of filter retrieval and reported appearance of the IVC on the postretrieval cavogram were recorded. Infusion of Blood Products Transfusion records were reviewed for all patients within 24 h prior to the retrieval procedure and 24 h after retrieval. Infusion of fresh-frozen plasma (FFP), platelets, or other blood products were recorded. The reason for transfusion of blood products was recorded. Follow-up All patients were observed by experienced nurses for a minimum of 4 h after filter retrieval. The routine evaluation of these patients included frequent assessment (every 15 min for the first hour, every half hour subsequently) of the puncture site for bleeding or hematoma, BP, heart rate, respiratory rate, and oxygenation. Patients were questioned regarding symptoms of pain at the access site or in the abdomen. Reports of bleeding, hematoma, abdominal, back, or neck pain, new hypotension with a systolic BP change from baseline of ⬎ 25 mm Hg, or tachycardia with an increase of the resting heart rate of ⬎ 40 beats/min were collected. Follow-up was obtained from hospital records and phone interviews. Hospital records were analyzed for evidence suggestive of retrieval-related complications such as puncture site hematoma or retroperitoneal bleeding. This included transfusion requirements within 24 h, imaging or operative evidence of retroperitoneal or pericaval bleeding, and clinical or imaging evidence of access site bleeding. Telephone interviews with patients were performed to evaluate if the patient had bleeding problems after filter removal. Chart reviews and medical records alone were used on 26 patients who could not be reached by telephone. Telephone calls alone were used on 35 patients without additional chart entries. When patients had multiple filters placed, data were collected for each filter placed and retrieved. The duration of follow-up was determined from the date of each filter removal to the date of the phone interview or, when this was not feasible, the last entry in the patient’s medical record. Statistical Analysis Data were entered into a worksheet (Excel 2003; Microsoft; Redmond, WA). Descriptive and summary statistics were calcuwww.chestjournal.org

lated, and graphs were created. The age and sex distribution of patients for both groups was calculated and tested for degrees of difference using the analysis of variance test or ␹2 test, respectively (GraphPad InStat V3.0 for Windows; GraphPad; San Diego, CA). The filter implantation time and duration of follow-up for all groups were compared using the analysis of variance test; a p value ⬍ 0.05 was considered significant.

Results Group 1 comprised 61 patients (42 women, 19 men) with an average age of 57 ⫾ 15 years. Group 2 included 22 patients (15 women, 7 men) with an average age of 51 ⫾ 15 years. In group 3 were 27 patients, 14 women and 13 men (average age, 42 ⫾ 13 years). The differences in age were only between groups 1 and 3 were statistically significant (p ⬍ 0.001). The differences in sex distribution were not statistically significant. Group 1: Therapeutically Anticoagulated Patients at the Time of Filter Retrieval Group 1 included 65 attempted filter retrievals (Gu¨nther Tulip [Cook; Bloomington, IN], n ⫽ 5; Recovery [Bard Peripheral Vascular; Tempe, AZ], n ⫽ 7; OptEase [Cordis; Warren, NJ], n ⫽ 2; and G2 [Bard Peripheral Vascular], n ⫽ 2) in 61 therapeutically anticoagulated patients by measured INR or dosing when receiving LMWH (Table 2). Thirty-six filters were retrieved in patients receiving oral anticoagulation with warfarin. Four retrievals were not successful; two filters demonstrated thrombus within the filter on preretrieval cavograms and retrieval was not attempted. In two patients, filter retrieval was attempted after 40 days and 120 days, respectively, but the filter legs were incorporated within the IVC wall and could not be retrieved. Eighteen filter retrievals were performed in patients receiving a therapeutic dose of LMWH. Seven filters were retrieved in patients treated with both a therapeutic dose of LMWH and oral anticoagulation. In the patients receiving oral anticoagulation, the median INR was 2.35 (range, 2 to 8). The median duration of treatment was 14 days (range, 2 to 122 days) for oral anticoagulation and 10 days (range, 5 to 50 days) for LMWH. During 2 of the 65 procedures in group 1, patients were receiving concomitant unfractionated heparin infusions with an aPTT ⬎ 70 s. The aPTT was normal in all other patients. Six patients were receiving aspirin: 325 mg/d po (n ⫽ 4) 81 mg/d po (n ⫽ 2). The platelet count was ⬎ 50 ⫻ 109/L for 81.5% (53 of 65 retrieval procedures) [median, 295.5 ⫻ 109/L; range, 116 to 750 ⫻ 109/L] and not determinable in retrospect in 18.5% (12 of 65 procedures). Early in our experience, FFP transfusion was performed prior to filter retrieval procedures in CHEST / 132 / 1 / JULY, 2007

33

Table 2—Anticoagulation Status at the Time of Attempted IVC Filter Retrieval (n ⴝ 115) Determined by Hematologic Coagulation Parameters or Dosing When Receiving LMWH Including the Median Duration of Treatment* Median Duration of Treatment, d (Range) Type of Anticoagulation Group 1 (therapeutically anticoagulated, n ⫽ 65) Therapeutic warfarin Therapeutic LWMH Warfarin and therapeutic LWMH Group 2 (prophylactically anticoagulated, n ⫽ 23) Prophylactic LWMH Group 3 (not therapeutically anticoagulated, n ⫽ 27) Subtherapeutic warfarin No anticoagulation medication

% (No./Total)

Warfarin

LMWH

34.8 (40/115) 15.6 (18/115) 6.1 (7/115)

14 (7–122) N/A 8 (2–10)

N/A 10 (5–50) 9 (5–11)

N/A

14 (2–35)

20 (23/115) 4.4 (5/115) 19.1 (22/115)

14.5 (7–23) N/A

N/A N/A

*N/A ⫽ not applicable.

three patients with an INR ⬎ 2.5. Subsequently, 33 filters were retrieved in patients with an INR ⱖ 2.0 (including 15 patients with an INR ⬎ 2.5) without prophylactic measures. Group 2: Prophylactically Anticoagulated Patients at the Time of Filter Retrieval Group 2 included 23 successful filter retrievals (Gu¨nther tulip, n ⫽ 19; Recovery, n ⫽ 3; OptEase, n ⫽ 1) in 22 patients receiving a prophylactic dose of LMWH with a median duration of treatment of 14 days (range, 2 to 35 days). The platelet count was ⬎ 50 ⫻ 109/L for 74% (17 of 23 retrieval procedures) [median, 366 ⫻ 109/L; range, 141 to 766 ⫻ 109/L] and not determinable in retrospect in 26% (6 of 23 procedures). One filter retrieval procedure was performed in a patient who received various blood products before and after filter retrieval due to disseminated intravascular coagulation and not related to the retrieval procedure. Group 3: Not Therapeutically Anticoagulated Patients at the Time of Filter Retrieval Group 3 comprised 27 attempted filter retrievals (Gu¨nther tulip, n ⫽ 14; Recovery, n ⫽ 11; OptEase, n ⫽ 1; and G2, n ⫽ 1) in 27 patients who were not therapeutically anticoagulated. Five patients were receiving oral anticoagulation with a subtherapeutic INR (median, 1.49; range, 1.16 to 1.69) for a median duration of 14.5 days (range, 7 to 23 days). In these patients, four filters were successfully retrieved. In one patient, retrieval was attempted after 119 days, but the filter had tilted and could not be retrieved. No anticoagulation medication was administered in 22 patients at the time of filter retrieval. In four patients, filter retrieval was not attempted due to thrombus within the filter on preretrieval cavogram 34

(n ⫽ 3) or filter tilting (n ⫽ 1). In one patient, filter retrieval was attempted after 110 days using a femoral catheter in addition to the jugular approach for filter repositioning, but the filter could not be retrieved due to incorporation within the IVC wall. In the patients without anticoagulation medication, the indications for filter placement were prophylaxis without VTE in 19 patients and the risk for VTE was no longer present at the time of IVC filter retrieval. Three patients with a history of DVT received an IVC filter before surgery and were not anticoagulated afterwards because DVT was not present on ultrasound. Two patients were receiving aspirin, 325 mg/d po, and clopidogrel, 75 mg/d po, and two patients were receiving aspirin, 325 mg/d po. The platelet count was ⬎ 50 ⫻ 109/L for 81.5% (22 of 27 retrieval procedures) [median, 330.5 ⫻ 109/L; range, 64 to 633 ⫻ 109/L] and not determinable in retrospect in 18.5% (5 of 27 procedures). IVC Filter Retrieval The route of attempted filter retrieval was internal jugular vein for the Gu¨nther filter, Recovery filter, and G2 filter. The OptEase filter was retrieved using the femoral vein. The success rate for filter retrieval was 91.3% (105 of 115 patients). For all patients, the median filter indwell time on the day of attempted filter retrieval was 14 days (range, 4 to 175 days). The median filter implantation time was 16.5 days (range, 4 to 147 days) in group 1, 18 days (range, 4 to 59 days) in group 2, and 40 days (range, 6 to 175 days) in group 3. The differences in filter implantation time were significant between groups 1 and 3 and groups 2 and 3 (p ⬍ 0.001) Postretrieval cavograms demonstrated no signs of contrast extravasation. Original Research

Follow-up The median duration of follow-up after filter removal in all patients was 26 months (range, 2 to 55 months). The median follow-up duration was 29 months (range, 2 to 54 months) in group 1, 28 months (range, 3 to 53 months) in group 2, and 21 months (range, 2 to 55 months) in group 3 (p ⬎ 0.05). No minor or major bleeding complications referable to the access site, or episodes of retroperitoneal hemorrhage were documented in patients of all groups. One patient in group 1 reported onset of epigastric pain at the time of filter retrieval. The postretrieval cavogram was normal, and vital signs and serial hematocrits remained stable. A CT scan obtained immediately after the procedure showed no evidence of retroperitoneal bleeding. The pain continued, and an acute disruption of a biliary anastomosis and intraperitoneal bile leak were subsequently diagnosed, unrelated to filter retrieval. Discussion The current recommendation, based on expert consensus but no data, is that IVC filter removal in patients with VTE should not be performed until the patient is anticoagulated.8 Achievement of therapeutic anticoagulation may take up to several weeks in some patients, especially with oral anticoagulation.1,15 Subsequent interruption or reversal of anticoagulation, or conversion from warfarin to heparin in order to retrieve a filter may place the patient at risk of PE and add cost as the process of anticoagulation is resumed. The outcomes of IVC filter retrieval in patients who are anticoagulated have previously not been specifically addressed. Although patients have been described as anticoagulated at the time of filter retrieval in previous studies,6,9,11,12,16 –21 the patient’s anticoagulation status was either not reported or not reported in objective terms. The present study is the first to specifically assess the safety of IVC filter retrieval in anticoagulated patients. There is only limited information on the safety of IVC filter retrieval in anticoagulated patients in the literature. In an earlier study,10 it was reported that patients were not anticoagulated during the retrieval procedure, but full-dose anticoagulation therapy was started 2 h after IVC filter retrieval. In another study,9 anticoagulation therapy was reversed for IVC filter removal. Millward et al19 performed filter retrieval shortly after initiation of anticoagulation without complications, but coagulation parameters were not reported. The only bleeding complications observed in that study were unrelated to the filter retrieval procedure. Furthermore, the risk of recurrent PE is greatest during the first 2 weeks of www.chestjournal.org

anticoagulation, indicating that a longer duration of anticoagulation prior to filter retrieval may be warranted in patients with VTE.4,22–24 In a study by Oliva et al,20 IVC filter retrieval was performed in 27 patients without bleeding complications, of whom 12 patients were noted to be anticoagulated but specific laboratory values were not provided. Rosenthal et al21 reported uneventful retrieval of 66 IVC filters in multiple trauma patients after initiation of anticoagulation therapy, but the degree of anticoagulation was not reported. Based on these studies, the safety of filter retrieval in anticoagulated patients is difficult to assess. In the present study, 56.5% of IVC attempted filter retrieval procedures were performed in therapeutically anticoagulated patients. Bleeding at the site of filter implantation in the IVC, or the venous access site, are potential hemorrhagic complications of filter removal. Early in our experience, three patients with an INR ⬎ 2.5 received FFP for temporary reversal of anticoagulation, but this practice was abandoned due to the potential risk of transfusion of blood products and the absence of observed bleeding complications. Subsequently, 15 patients with an INR ⬎ 2.5 underwent filter retrieval without clinically evident complications. Although our study is retrospective in nature, it is consecutive and reflective of clinical practice in our institution. Our findings suggest that filter retrieval in patients considered being therapeutically anticoagulated by measured coagulation studies or LMWH dosing is safe with respect to hemorrhagic complications. There are several limitations of this study. First, the study is retrospective in nature and therefore not controlled for selection bias, detection of events, and data collection. Second, although 115 retrieval procedures were evaluated, complications that occur at a low rate could have been missed. For example, three different types of filters were used in our patients, further decreasing the potential sensitivity of the study for adverse events. Third, the longest duration of filter implantation in our series was 175 days. The risk of bleeding complications when removing filters that have been in place for extended durations cannot be determined from this study. Fourth, it was assumed that all patients on LMWH were compliant and therefore therapeutically anticoagulated. Lastly, none of our patients were excessively anticoagulated. Reversal of anticoagulation in patients with extremely elevated INR or aPTT was not addressed in our study. In conclusion, this is the first study to specifically address the question of safety of filter retrieval in therapeutically anticoagulated patients. Our results suggest that retrieval of optional IVC filters in anticoagulated patients is safe. Based on these reCHEST / 132 / 1 / JULY, 2007

35

sults, we hypothesize that routine interruption or reversal of anticoagulation for retrieval of IVC filters may not be necessary. References 1 Buller HR, Agnelli G, Hull RD, et al. Antithrombotic therapy for venous thromboembolic disease: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004; 126:401S– 428S 2 Streiff MB. Vena caval filters: a comprehensive review. Blood 2000; 95:3669 –3677 3 Athanasoulis CA. Complications of vena cava filters. Radiology 1993; 188:614 – 615 4 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 5 Prepic Study Group. Eight-year follow-up of patients with permanent vena cava filters in the prevention of pulmonary embolism: the PREPIC (Prevention du Risque d’Embolie Pulmonaire par Interruption Cave) randomized study. Circulation 2005; 112:416 – 422 6 Terhaar OA, Lyon SM, Given MF, et al. Extended interval for retrieval of Gunther Tulip filters. J Vasc Interv Radiol 2004; 15:1257–1262 7 Kinney TB. Update on inferior vena cava filters. J Vasc Interv Radiol 2003; 14:425– 440 8 Kaufman J, Kinney T, Streiff M, et al. Guidelines for the use of retrievable and convertible vena cava filters: report from the society of interventional radiology multidisciplinary consensus conference. J Vasc Interv Radiol 2006; 17:449 – 459 9 Asch MR. Initial experience in humans with a new retrievable inferior vena cava filter. Radiology 2002; 225:835– 844 10 Wicky S, Doenz F, Meuwly J-Y, et al. Clinical experience with retrievable Gu¨nther Tulip vena cava filters. J Endovasc Ther 2003; 10:994 –1000 11 Neuerburg JM, Gunther RW, Vorwerk D, et al. Results of a multicenter study of the retrievable Tulip Vena Cava filter: early clinical experience. Cardiovasc Intervent Radiol 1997; 20:10 –16 12 Grande WJ, Trerotola SO, Reilly PM, et al. Experience with

36

13

14

15

16 17

18 19

20 21 22 23

24

the recovery filter as a retrievable inferior vena cava filter. J Vasc Interv Radiol 2005; 16:1189 –1193 Greenfield LJ, Rutherford RB. Recommended reporting standards for vena caval filter placement and patient followup: Vena Caval Filter Consensus Conference. J Vasc Interv Radiol 1999; 10:1013–1019 Millward SF, Grassi CJ, Kinney TB, et al. Reporting standards for inferior vena caval filter placement and patient follow-up: supplement for temporary and retrievable/optional filters. J Vasc Interv Radiol 2005; 16:441– 443 Ansell J, Hirsh J, Poller L, et al. The pharmacology and management of the vitamin K antagonists: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004; 126:204S–233S Binkert CA, Sasadeusz K, Stavropoulos SW. Retrievability of the recovery vena cava filter after dwell times longer than 180 days. J Vasc Interv Radiol 2006; 17:299 –302 De Gregorio MA, Gimeno MJ, Madariaga B, 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 Hoff WS, Hoey BA, Wainwright GA, et al. Early experience with retrievable inferior vena cava filters in high-risk trauma patients. J Am Coll Surg 2004; 199:869 – 874 Millward SF, Oliva VL, Bell SD, et al. Gu¨nther Tulip retrievable vena cava filter: results from the registry of the Canadian Interventional Radiology Association. J Vasc Interv Radiol 2001; 12:1053–1058 Oliva VL, Szatmari F, Giroux MF, et al. The Jonas study: evaluation of the retrievability of the Cordis OptEase inferior vena cava filter. J Vasc Interv Radiol 2005; 16:1439 –1445 Rosenthal D, Wellons ED, Lai KM, et al. Retrievable inferior vena cava filters: initial clinical results. Ann Vasc Surg 2006; 20:157–165 Carson J, Kelley M, Duff A, et al. The clinical course of pulmonary embolism. N Engl J Med 1992; 326:1240 –1245 Douketis J, Foster G, Crowther M, et al. Clinical risk factors and timing of recurrent venous thromboembolism during the initial 3 months of anticoagulant therapy. Arch Intern Med 2000; 160:3431–3436 Eichinger S, Weltermann A, Minar E, et al. Symptomatic pulmonary embolism and the risk of recurrent venous thromboembolism. Arch Intern Med 2004; 164:92–96

Original Research