Suprarenal Inferior Vena Cava Filters: A 20-Year Single-Center Experience

Transcription

1 Suprarenal Inferior Vena Cava Filters: A 20-Year Single-Center Experience 1 Sanjeeva P. Kalva, MD, Chrysanthi Chlapoutaki, MD, Stephan Wicky, MD, Alan J. Greenfield, MD, Arthur C. Waltman, MD, and Christos A. Athanasoulis, MD PURPOSE: To assess the clinical safety and efficacy of suprarenal inferior vena cava (IVC) filters during long-term follow-up. MATERIALS AND METHODS: In this retrospective study, the authors collected the following data about patients who underwent suprarenal IVC filter placement at their institution between 1988 and 2007: demographics, clinical presentation, indications for filter placement, reasons for placing the filter in the suprarenal IVC, type of filter, frequency of pulmonary embolism (PE) after filter placement, and filter-related problems during follow-up. RESULTS: Seventy patients (32 male and 38 female patients; mean age, 60 years) had suprarenal IVC filters. Sixty-two patients presented with symptoms of venous thromboembolism (VTE) and eight had incidental asymptomatic VTE at imaging. Indications for filter placement were as follows: contraindication to anticoagulation n 48), ( complications and/or failure of anticoagulation n ( 12), added protection n ( 8), and prophylaxis n ( 2). Suprarenal placement was chosen due to IVC thrombus n ( 41), intrinsic and/or extrinsic narrowing of the infrarenal IVC n ( 9), renal and/or gonadal vein thrombus n ( 3), congenital IVC anomalies n ( 6), pelvic mass n ( 5), pregnancy (n 3), and other reasons (n 3). The following filters were used: Greenfield n ( 29), Simon Nitinol n ( 5), Vena-Tech (n 3), TrapEase (n 22), OptEase (n 3), Tulip (n 6), Bird s Nest n ( 1), and Recovery n ( 1). During follow-up (mean, 573 days 953), postfilter PE was suspected in 10 patients; eight patients underwent computed tomography (CT), one of whom had PE at CT. None developed new symptoms of caval thrombosis. Abdominal CT (performed in 30 patient at a mean of 543 days 768) showed thrombus in the filter in three patients, fracture in one patient, and penetration of the IVC wall in two patients. CONCLUSIONS: Suprarenal filters are safe and effective in preventing PE. The placement of IVC filters above the renal veins does not carry an added risk of complications. J Vasc Interv Radiol 2008; 19: Abbreviations: DVT deep venous thrombosis, IVC inferior vena cava, PE pulmonary embolism, VTE venous thromboembolism IMPLANTATION of inferior vena tion (1 5). Currently available IVC fil- are approved for placement in theflow). Indications for suprarenal IVC (above the level of renal vein in- cava (IVC) filters is as an alternativeters therapeutic option in patients with ve-infrarenanous thromboembolism (VTE) who however, it may be necessary to im-sic narrowing of the infrarenal IVC IVC. In certain patients, placement include intrinsic or extrin- otherwise cannot receive anticoagulaplant the IVC filter in the suprarenalthat precludes placement of an IVC filter in the infrarenal IVC 6), ( thrombus in the gonadal or renal veins 7 9), ( and anatomic variations of the IVC or From the Department of Radiology, Massachusetts Disclaimer: The abstract has been presented at the the renal veins. Compared with the General Hospital, Harvard Medical School, GRB- Radiological Society of North America Annual infrarenal IVC, the suprarenal IVC is 290, 55 Fruit St, Boston, MA Received November 2, 2007; revision received March 26, 2008; of the data in this paper has been published from Meeting held at Chicago, IL during Nov Some larger in diameter but shorter in accepted March 30, Address correspondence our hospital in various original papers; however, length. Its diameter is variable, often to S.P.K.; skalva@partners.org this paper included more long term follow-up clinical and imaging data whenever available than is right heart pressures. These factors in- 1 Current address: Hopital Armand-Trousseau-La reported in prior papers. influenced by cardiac pulsations and Roche-Guyon, Groupement Hospitalier Universitaire Est, Paris, France. fluence the choice of the filter and the SIR, 2008 risk of incomplete protection (10), filter migration (11), penetration (12), S.W. received honoraria for speaking engagements from Cordis Corporation. DOI: /j.jvir and device fracture. In addition, 1041

2 1042 Suprarenal Inferior Vena Cava Filters July 2008 JVIR thrombus propagation, renal vein thrombosis, and renal failure (13) have been reported with suprarenal IVC filter placement. Original reports by Greenfield et al (6,14) with 148 suprarenal Greenfield IVC filters showed no statistically significant difference in safety and efficacy of suprarenal IVC filters compared with infrarenal IVC filters. A recent study with four different designs of suprarenal IVC filters in 22 patients showed a higher rate of filter migration (15). Athanasoulis et al (16) reported on 44 patients who had filter implantation either in the suprarenal IVC or adjacent to the renal vein inflow and found equal efficacy and safety of these filters compared to traditional infrarenal filters. However, the above studies included only Greenfield-type filters (original 24 F, stainless steel, titanium; Boston Scientific, Natick, Massachusetts), LGM (I and II; B. Braun Medical Inc, Bethlehem, Pennsylvania), and Bird s Nest filters (Cook Medical Inc, Bloomington, Indiana) and the follow-up data were short-term. In this study, we evaluated the safety and efficacy of suprarenal IVC filters of various designs that were placed during a 20- year period in our hospital. MATERIALS AND METHODS This is an institutional review board approved, retrospective study of patients who underwent implantation of an IVC filter above the renal veins at our institution between 1988 and The institutional review board waived the requirement to obtain informed consent from the subjects of this study. The study also complied with the Health Insurance Portability and Accountability Act. Patients Patients who had a suprarenal IVC filter placed at our hospital were identified by using our interventional radiology database system (Hi-IQ System, Health & Inventory Information for Quality, ConexSys, a subsidiary of Society of Interventional Radiology, Woonsocket, Rhode Island) and hospital information system database (Folio views, 4.2, Open Market Inc, Burlington, Massachusetts). During the 20-year period of this review, 4,027 IVC filters were placed Table 1 Underlying Medical Diseases before Filter Placement No. of Patients Underlying Diseases (n 70) Malignancy 39 (56) Perioperative state 9 (13) Congestive heart failure 6 (8.6) Stroke 2 (2.8) Trauma 3 (4.3) Pregnancy 3 (4.3) Other* 8 (11) Note. Numbers in parentheses are percentages. * Other includes hypercoagulable states in two patients, heparin-induced thrombocytopenia in one patient, systemic lupus erythematosis with antiphospholipid antibodies in one patient, nephrolithiasis in one patient, nephrotic syndrome in one patient, pancreatitis in one patient, and leg ulcer in one patient. at our institution. Seventy patients had a suprarenal IVC filter. There were 32 male (46%) and 38 female (54%) patients. The age of the patients at filter implantation ranged from 16 to 94 years, with a mean of 60 years and median of 58 years. The male patients ranged in age from 19 to 87 years (mean age, 62 years; median age, 67 years). The female patients ranged in age from 16 to 94 years (mean age, 57 years; median age, 56 years). Clinical and Procedural Data The following preprocedural clinical details were recorded: clinical presentation at the time of filter request, underlying medical condition, imaging findings with regard to VTE at filter request, and indications for filter placement. Technical details of filter implantation including the route of venous access, findings at cavography, type of filter used, reasons for suprarenal filter placement, and complications during filter implantation were recorded. Clinical Presentation Thirty-five of the 70 patients (50%) presented with clinical symptoms and signs of deep venous thrombosis (DVT). Twenty-seven patients (38%) presented with clinical symptoms and signs of PE. The remaining eight patients (11%) had no signs or symptoms of VTE. In these eight patients, VTE was found incidentally on imaging studies obtained for some other reason and patients were referred for IVC filter placement. Table 1 lists the underlying medical conditions in all patients before filter placement. Indications for Filter Placement A contraindication to anticoagulation in the presence of VTE constituted the major indication for filter placement. Forty-eight of the 70 patients (68%) had a contraindication to anticoagulation. The contraindications included the presence of brain malignancy (n 7); other intracranial abnormality (stroke in three, intracranial bleed in three); active bleeding (gastrointestinal bleeding in four, hematuria in three); cancers with an active or with a high potential for bleeding (endometrial cancer in four, gastric cancer in two, anal cancer in one, and retroperitoneal cancer in one); perioperative state after abdominal/pelvic surgery (n 10), chest surgery (n 3), or orthopedic surgery (n 3); and others, septicemia in one, retroperitoneal tumor compressing the IVC in two, risk of fall in one). Five of the 70 patients (7.1%) had developed bleeding complications (rectus sheath hematoma in one [a pregnant patient], retroperitoneal hematoma in one, hematuria in one, subdural hematoma in one, and vaginal bleeding in one) after anticoagulation. Seven of the 70 patients (10%) had failure of anticoagulation with recurrent VTE despite use of anticoagulants. In eight patients (11%), a filter was implanted as an additional protection in addition to

3 Volume 19 Number 7 Kalva et al 1043 anticoagulation due to the presence of underlying cardiac disease or right heart strain. In two patients (2.8%), the filter was prophylactic one was placed preoperatively in a patient with ovarian cancer and the other was placed in a patient with multiorgan trauma. Filter Implantation Although a defined standard protocol has not been in place, a practice pattern described herein has been followed by almost all operators. Access to the IVC was obtained through the right common femoral vein in 48 of the 70 patients (68%), left femoral vein in seven patients (10%), right internal jugular vein in 14 patients (20%), and right basilic vein in one patient (1.4%). A pigtail catheter was placed in the IVC just above the iliac vein confluence and cavography performed. The pigtail catheter was exchanged over a wire for the filter delivery sheath, and the filter was deployed according to the manufacturer s guidelines. Repeat cavography was performed after filter implantation. Follow-up Clinical and imaging follow-up data were recorded whenever available. The concurrent use of anticoagulants after filter placement was recorded. However, the concurrent use of antiplatelet agents was not recorded. Immediate and long-term complications were reviewed. The occurrence of PE and symptomatic vena cava thrombosis after filter placement were recorded during clinical followup. Whenever postfilter PE was suspected clinically, the available imaging data including images from computed tomographic (CT) pulmonary angiography or ventilation-perfusion scintigraphy were reviewed for evidence of PE. If any patient underwent abdominal CT for any reason during follow-up, the scan was reviewed for filter migration (of 2 cm), filter fracture, migration of fractured fragments, thrombus in the filter or the IVC, and penetration of more than 3 mm through the wall of the IVC. Similarly, if any patient underwent contrast medium enhanced chest CT for any clinical reason, the CT scan was reviewed for PE. Imaging studies (ultrasonography [US] and CT venography) obtained for lower extremity DVT after filter placement were reviewed for the occurrence of new DVT. Endpoints The safety of suprarenal filters was assessed on the basis of the occurrence of symptomatic or asymptomatic problems such as vena cava thrombosis, filter migration, and device fracture during filter deployment and follow-up. The efficacy was assessed on the basis of the occurrence of fatal and nonfatal PE after filter placement. RESULTS Imaging for VTE before Filter Implantation Sixty-seven of the 70 patients (96%) were tested for VTE before filter implantation. Sixty-six of the 67 patients had VTE on imaging studies. Ten patients had both PE and DVT, 40 had DVT, and 16 had PE. One patient had no VTE at imaging. A filter was requested in this patient for prophylaxis after major trauma. Twenty-nine patients were tested for PE. CT pulmonary angiography was performed in 18 patients, conventional catheter pulmonary angiography in nine, and ventilation-perfusion scanning in two. Of the 29 patients tested for PE, 26 had PE at imaging. Fifty-nine patients were tested for DVT. Compression color Doppler US was performed in 28 patients, CT in 24, conventional venography in four, and magnetic resonance imaging in three. Of the 59 patients tested for DVT, 50 had DVT at imaging. DVT affected the left leg in 16 patients, the right leg in 13, and both legs in 13. Thrombus in the IVC was seen in six patients, including one who had additional left renal vein thrombus. One patient had left renal vein thrombus, and one additional patient had left gonadal vein thrombus. Three patients had no imaging studies prior to filter requests. One patient was on warfarin sodium (Coumadin; Bristol-Myers Squibb Co, Princeton, New Jersey) for prior PE and had clinical symptoms of recurrent PE when the filter was requested. Another patient had left renal vein thrombosis on a prior abdominal CT scan, with an underlying renal cell cancer and brain metastases. The third patient had ovarian cancer, and a filter was requested preoperatively as a prophylactic measure against PE. Findings at Cavography Cavography demonstrated thrombus in the IVC in 33 patients. Thrombus was below the renal veins in 32 patients and extended into the suprarenal IVC in one patient. One of the 33 patients also had diffuse narrowing of the infrarenal IVC, which was probably due to chronic thrombosis. Nine additional patients had thrombus in the infrarenal IVC that extended above an existing infrarenal IVC filter. Thus, overall, thrombus in the IVC was found in 42 patients. Chronic occlusion with narrowing of the IVC and retroperitoneal collateral vessels was found in three patients. One additional patient had diffuse narrowing of the infrarenal IVC with no documented collateral vessels. Two patients had renal vein thrombosis. Four patients had extrinsic compression of the infrarenal IVC. Developmental anomalies of the IVC and renal veins were found in seven patients, including a duplicated IVC in four patients, circumaortic left renal vein in two patients, and retroaortic left renal vein in one patient. Eleven patients had a normal, patent IVC. In these 11 patients, the suprarenal IVC was chosen for filter placement due to a pelvic and/or abdominal mass necessitating surgery, pregnancy, or renal/gonadal vein thrombosis. Suprarenal Placement Reasons for placing the filter above the renal veins are listed in Table 2. Acute or chronic thrombosis of the infrarenal IVC was the major reason for choosing suprarenal implantation. The second most common reason was developmental anomalies of the IVC and renal veins. Filters used in this study are listed in Table 3. The filters were successfully deployed at the intended location in all but one patient, in whom the filter migrated to the heart immediately after deployment (see the next section).

4 1044 Suprarenal Inferior Vena Cava Filters July 2008 JVIR Table 2 Reasons for Choosing Suprarenal IVC for Filter Implantation No. of Patients Reason (n 70) Thrombus in IVC 41 (58) Extrinsic compression 4 (5.7) Duplication 4 (5.7) Circumaortic renal vein 2 (2.9) Surgery* 1 (1.4) Pregnancy 3 (4.3) Pelvic mass necessitating surgery 5 (7.1) Chronic occlusion/intrinsic narrowing 5 (7.1) Technical 2 (2.9) Renal vein thrombus 2 (2.9) Gonadal vein thrombus 1 (1.4) Note. Numbers in parentheses are percentages. * Surgery was performed for an adrenal mass. One patient also had infrarenal IVC thrombus. Table 3 Filters Used in this Study Filters Used No. of Patients 12-F titanium Greenfield 23 (33) 12-F Greenfield (unspecified) 3 (4.3) 12-F stainless steel Greenfield 2 (2.9) 24-F Greenfield 1 (1.4) Vena Tech (LGM) 3 (4.3) Simon Nitinol 5 (7.1) Bird s Nest 1 (1.4) TrapEase 22 (31.4) OptEase 3 (4.3) Recovery 1 (1.4) Tulip 6 (8.6) Immediate Complications In four patients, there was a problem during deployment of the 12-F Titanium Greenfield filter. The filter legs did not open properly during deployment, and they were clumped together. This required additional maneuvers with a catheter to open the legs. During deployment of a Titanium Greenfield filter in another patient, one of the filter legs landed in the renal vein. Thrombus within the IVC migrated to the apex of the filter (Greenfield filter, unspecified) just after filter deployment in another patient. One filter (Simon Nitinol filter) migrated to the heart during deployment and was later retrieved with an endovascular approach with no further complications. Later, a suprarenal titanium Greenfield filter was placed in that patient. One Tulip filter required repositioning immediately after deployment due to incomplete opening of the filter in a patient with thrombus in the suprarenal IVC (Figure). Subsequent Interventional Procedures In one patient with IVC narrowing, a stent was placed in the IVC after filter deployment. Another patient who had presented with extensive DVT underwent thrombolysis and stent placement of the IVC and iliac veins just after filter deployment. Another patient received bilateral iliac vein stents to relieve extrinsic compression by a large endometrial cancer. One OptEase filter was successfully retrieved 6 days after placement. In another patient, a Tulip filter was successfully retrieved 79 days after placement. Clinical Follow-up Clinical follow-up was available in 61 of the 70 patients (87%). The follow-up period ranged from 2 to 3,933 days, with a mean ( standard deviation) and median follow-up of 573 days 953 and 137 days, respectively. During the follow-up, 23 of the 61 patients (38%) received anticoagulation (including heparin or low-molecularweight heparin in 10 patients and warfarin sodium in 13). Fifteen patients died during follow-up (range, 2 1,229 days; mean, 195 days 328; median, 52 days). Seven of the 61 patients (11%) died within 30 days of filter placement. An autopsy was performed in three patients. In one patient, there was no PE. One patient demonstrated multiple septic emboli in both lungs, which were seen on the CT pulmonary angiogram obtained before filter placement. Disseminated intravascular coagulation was considered to be the cause of death in this patient. Another patient had chronic PE, which were demonstrated with catheter pulmonary angiography before filter placement. There were no deaths due to PE. Late Complications During follow-up, 10 of 61 patients had symptoms and signs of PE, including one patient who had two episodes. The first episode of postfilter PE was suspected clinically after a median of 149 days (range, 1 1,147 days; mean, 340 days 403) after filter placement. Two patients had PE within the first 10 days after filter placement. Eight of the 10 patients were evaluated with CT pulmonary angiography; one had a new PE at CT. One additional patient underwent ventilation-perfusion scanning and was found to have low probability for PE. The other patient had no imaging studies. Thus, overall, the frequency of postfilter clinical PE was 11.5% (10 of 61). However, PE was confirmed with imaging only in one of the 61 patients (16%). During follow-up, none of the patients had new symptoms suggestive of IVC thrombosis or occlusion. Laboratory results of renal function were available in 65 patients after a median of 54 days (range, 1 3,656 days; mean, 543 days 947) after filter placement.

5 Volume 19 Number 7 Kalva et al 1045 Figure. Cavograms in an 80-year-old woman who was being treated with warfarin sodium for thrombus in the suprarenal IVC. The patient presented with gastrointestinal bleeding. An IVC filter was requested in view of complication from warfarin sodium therapy. (a) Image obtained through a pigtail catheter placed in the infrarenal IVC demonstrates thrombus (arrow) in the suprarenal IVC. (b) A Tulip filter was deployed; however, the legs (arrow) were deployed partially in the thrombus and the filter did not open completely. (c) The filter was retrieved and redeployed at a higher location above the thrombus. Seven of the 65 patients (11%) had worsening renal function, which occurred within a mean of 36 days 47 (range, days; median, 26 days) after filter placement. In three patients, the worsening renal function was attributed to bilateral obstructive hydronephrosis from prostate cancer (n 2) or recurrent rectal cancer (n 1). Renal failure was attributed to multiorgan failure after major hepatic resection in one patient and to septicemia with candida infection in one patient. One patient developed acute renal failure 4 weeks after filter placement, and US demonstrated patent renal veins. One patient had mild worsening of renal function, and no follow-up data were available. Imaging Follow-up Abdominal CT. After filter placement, 30 of the 70 patients (43%) underwent abdominal CT. CT was performed for various clinical reasons pertinent to the underlying medical disease and not for the IVC filter per se. CT was performed at a mean and median of 543 days 768 and 272 days, respectively, after filter placement (range, 6 2,621 days). Thrombus within the filter was noted in three filters two TrapEase and one Tulip. Thrombus was seen in the periphery of the TrapEase filter as a thin rim attached to the wall of the IVC. In the case of the Tulip filter, the thrombus was seen at the apex of the filter, without total occlusion of the IVC. There were no cases of filter migration of more than 2 cm; however, one Tulip filter migrated 1 cm superiorly. Two filters (one Recovery and one Greenfield, material unspecified) penetrated the caval wall and the filter legs were seen in the pancreas or the liver. In one patient, one of the legs of the titanium Greenfield filter was in the renal vein. Another patient with a titanium Greenfield filter developed new thrombus in the IVC without any filter thrombus. There was no extension of the thrombus into the renal veins in any patient. Chest CT. Chest CT scans obtained after filter placement were available in 17 of the 70 patients (24%). Chest CT was performed at a mean and median of 595 days 759 and 312 days, respectively, after filter placement (range, 1 2,367 days). In eight patients, CT was performed for suspected PE according to a standard thin-section ( mmthick sections) CT pulmonary angiography protocol. In the remaining nine patients, chest CT was performed for lung disease with use of 5-mm-thick sections. There was no asymptomatic PE at chest CT. As described earlier, one of the symptomatic patients had new PE at CT. Imaging for DVT. After filter placement, 13 patients underwent imaging of the lower extremities for clinically suspected PE or DVT. These imaging studies were obtained at a mean and median of 350 days 483 and 137 days, respectively, after filter placement (range, 10

6 1046 Suprarenal Inferior Vena Cava Filters July 2008 JVIR 1,606 days). Eight of the 13 patients underwent compression color Doppler US and five underwent CT venography. New DVT was seen in one patient, affecting the popliteal vein. One additional patient had thrombosis of an iliac vein stent that was placed to relieve extrinsic compression. DISCUSSION Before the introduction of IVC filters, surgical cava ligation or plication (17,18) were performed to protect patients from potentially lethal PE if these patients could not receive anticoagulants. Surgery was directed to the infrarenal IVC to avoid renal vein thrombotic occlusion. IVC filters were developed as an alternative to surgery. Therefore, they were designed for placement in the infrarenal IVC. The first-generation IVC filters were associated with a high incidence of caval thrombosis (19 21). Thus, application of filters was strictly limited to placement in the infrarenal position. An additional recommendation was to place the device with its apex near the point where the renal veins drain into the cava. Renal venous inflow would prevent propagation of thrombus above the filter in case of filter thrombosis. This practice continues today. Filter implantation in the infrarenal IVC has been the norm due to persistent concerns about suprarenal filter position. All currently available IVC filters are approved for implantation in the infrarenal IVC. However, implantation in the suprarenal IVC was described as early as 1981 by Greenfield (22). Other early reports (6,8,14,23) on suprarenal implantation of the Greenfield filter demonstrated no problems or additional complications as a result of placement above the renal veins. There have been sporadic case reports published about perforation, penetration, and migration of the suprarenal Greenfield filter (12,24,25). A more recent publication on suprarenal filters by Matchett et al (15) in 22 patients reported the use of four different types (titanium Greenfield filter, Vena Tech LGM 1& 2, Bird s Nest filter). They reported high efficacy and a 3% rate of migration with suprarenal filters. Thrombus propagation and renal dysfunction, reported in one article (13), have not been observed in other studies. Although these studies suggested the safety of suprarenal filter placement, they were confined predominantly to the Greenfield filter design. Many new devices have been approved since then, and they are used more often than the Greenfield filter due to the small size of the delivery system. The safety and efficacy of these filters have not been reported. In addition, we report CT imaging follow-up in almost half of our patients to look for asymptomatic filter-related problems. In agreement with previous reports, we found high efficacy of suprarenal IVC filters. The overall frequency of clinical PE was 11.5% and that of imaging-evident PE was 16%, which is comparable to that reported with infrarenal IVC filters (4,16,26 28). The occurrence of new, symptomatic caval thrombosis after filter placement was rare. The concurrent use of anticoagulants as clinically appropriate may also have contributed to the low occurrence of PE and symptomatic vena caval thrombosis. Other complications, such as penetration or perforation, were found in two patients at CT but were asymptomatic. Except for migration of a Simon Nitinol filter to the heart, procedural problems were predominantly due to incomplete opening of the Greenfield filter. Migration of a filter is affected by many factors, such as the diameter of the IVC at filter placement, changes in the diameter immediately after filter placement, filter design, and operator errors. The diameter of the suprarenal IVC is larger than that of the infrarenal IVC. In addition, the diameter of the IVC is related to venous return (29), blood volume (30), and respiratory cycle. These factors are important when choosing the type of filter design for implantation in the suprarenal IVC. A few filters (Greenfield, Simon Nitinol, Vena Tech, G2) are approved for an IVC of 28 mm in diameter, whereas other filters (TrapEase, OptEase, Tulip) are approved for an IVC of 30 mm. The Bird s Nest filter is approved for placement in a mega cava, up to a diameter of 40 mm. A correct assessment of IVC diameter is essential so that an appropriate filter can be chosen. Renal dysfunction, although a concern if thrombus extends into the renal vein, is uncommon after suprarenal IVC filter placement. Extension of the thrombus into the renal veins is more common in the presence of extensive thrombus in the infrarenal IVC and absence of therapy with anticoagulants. Previous studies (8,14) have shown no incidence of renal dysfunction after filter placement in the suprarenal IVC. In our study, there was no CT evidence of thrombus propagation to the renal veins after suprarenal IVC filter placement. Although seven patients developed renal dysfunction after filter placement, underlying medical conditions contributed to renal dysfunction in five patients. US showed patent renal veins in one additional patient with acute renal failure 1 month after filter placement. This study is limited because it is retrospective, based on a review of medical records. No standard protocol was followed for filter placement, as the devices and operator changed during the study period. There was no complete follow-up and no standardized follow-up imaging in all patients. However, it provides additional information about the feasibility of placing IVC filters in the cava above the renal veins in the clinical circumstances when this becomes necessary. In conclusion, we have found that if placement of IVC filters in the infrarenal cava is not feasible, placement above the renal veins may be considered. The efficacy and safety of suprarenal IVC filters is comparable to that of filters placed in the standard, below-the renal veins position. References 1. Jacobs DG, Sing RF. The role of vena caval filters in the management of venous thromboembolism. Am Surg 2003; 69: Rectenwald JE. Vena cava filters: uses and abuses. Semin Vasc Surg 2005; 18: Moores LK, Tapson VF. Vena caval filters in pulmonary embolism. Semin Vasc Med 2001; 1: Kinney TB. Update on inferior vena cava filters. J Vasc Interv Radiol 2003; 14: Weichman K, Ansell JE. Inferior vena cava filters in venous thromboembolism. Prog Cardiovasc Dis 2006; 49: Greenfield LJ, Proctor MC. Suprarenal filter placement. J Vasc Surg 1998; 28: , discussion Abujudeh H, Lim H. Emergency suprarenal inferior vena cava filter place-

7 Volume 19 Number 7 Kalva et al 1047 ment in ovarian vein thrombosis. Emerg Radiol 2004; 10: Orsini RA, Jarrell BE. Suprarenal placement of vena caval filters: indications, techniques, and results. J Vasc Surg 1984; 1: Wieland A, Mehta A, Greenfield A, Wolf M. Temporary suprarenal inferior vena cava filter for renal biopsy in a patient with renal vein thrombosis. Clin Nephrol 2004; 62: Lemmon GW, Litscher LJ. Incomplete caval protection following suprarenal caval filter placement: a case report. Angiology 2000; 51: Ganguli S, Tham JC, Komlos F, Rabkin DJ. Fracture and migration of a suprarenal inferior vena cava filter in a pregnant patient. J Vasc Interv Radiol 2006; 17: Teitelbaum GP. Perforation of the inferior vena cava by a suprarenal Greenfield filter. Radiology 1990; 174: Marcy PY, Magne N, Frenay M, Bruneton JN. Renal failure secondary to thrombotic complications of suprarenal inferior vena cava filter in cancer patients. Cardiovasc Intervent Radiol 2001; 24: Greenfield LJ, Cho KJ, Proctor MC, Sobel M, Shah S, Wingo J. Late results of suprarenal Greenfield vena cava filter placement. Arch Surg 1992; 127: Matchett WJ, Jones MP, McFarland DR, Ferris EJ. Suprarenal vena caval filter placement: follow-up of four filter types in 22 patients. J Vasc Interv Radiol 1998; 9: Athanasoulis CA, Kaufman JA, Halpern EF, Waltman AC, Geller SC, Fan CM. Inferior vena caval filters: review of a 26-year single-center clinical experience. Radiology 2000; 216: Spencer FC, Quattlebaum JK, Quattlebaum JK Jr, Sharp EH, Jude JR. Plication of the inferior vena cava for pulmonary embolism: a report of 20 cases. Ann Surg 1962; 155: Pollak EW, Sparks FC, Barker WF. Inferior vena cava interruption: indications and results with caval ligation, clips and intracaval devices in 110 cases. J Cardiovasc Surg (Torino) 1974; 15: Wingerd M, Bernhard VM, Maddison F, Towne JB. Comparison of caval filters in the management of venous thromboembolism. Arch Surg 1978; 113: Ferris EJ, McCowan TC, Carver DK, McFarland DR. Percutaneous inferior vena caval filters: follow-up of seven designs in 320 patients. Radiology 1993; 188: Menzoian JO, LoGerfo FW, Weitzman AF, Ezpeleta M, Sequeira JC. Clinical experience with the Mobin-Uddin vena cava umbrella filter. Arch Surg 1980; 115: Greenfield LJ, Peyton R, Crute S, Barnes R. Greenfield vena caval filter experience: late results in 156 patients. Arch Surg 1981; 116: Stewart JR, Peyton JW, Crute SL, Greenfield LJ. Clinical results of suprarenal placement of the Greenfield vena cava filter. Surgery 1982; 92: Kim D, Porter DH, Siegel JB, Simon M. Perforation of the inferior vena cava with aortic and vertebral penetration by a suprarenal Greenfield filter. Radiology 1989; 172: Greenfield LJ. Perforation of the inferior vena cava with aortic and vertebral penetration by a suprarenal Greenfield filter. Radiology 1990; 175: Liu WC, Do YS, Choo SW, et al. The mid-term efficacy and safety of a permanent nitinol IVC filter (TrapEase). Korean J Radiol 2005; 6: Kalva SP, Wicky S, Waltman AC, Athanasoulis CA. TrapEase vena cava filter: experience in 751 patients. J Endovasc Ther 2006; 13: Kalva SP, Athanasoulis CA, Fan CM, et al. Recovery vena cava filter: experience in 96 patients. Cardiovasc Intervent Radiol 2006; 29: Lorsomradee S, Lorsomradee S, Cromheecke S, ten Broecke PW, De Hert SG. Inferior vena cava diameter and central venous pressure correlation during cardiac surgery. J Cardiothorac Vasc Anesth 2007; 21: Chen L, Kim Y, Santucci KA. Use of ultrasound measurement of the inferior vena cava diameter as an objective tool in the assessment of children with clinical dehydration. Acad Emerg Med 2007; 14: