- Email: [email protected]
Bedside inferior vena cava filter placement by intravascular ultrasound in critically ill patients is safe and effective for an extended time
From the Southern Association for Vascular Surgery
Bedside inferior vena cava filter placement by intravascular ultrasound in critically ill patients is safe and effective for an extended time Roan J. Glocker, MD, Oluwafunmi Awonuga, MD, Zdenek Novak, MD, Benjamin J. Pearce, MD, Mark Patterson, MD, Thomas C. Matthews, MD, William D. Jordan, MD, and Marc A. Passman, MD, Birmingham, Ala Background: Bedside inferior vena cava filter (IVCF) placement by intravascular ultrasound (IVUS) guidance has previously been shown to be a safe and effective technique, especially for critically ill patients, with initial experience of a prospectively implemented algorithm. The purpose of this study was to evaluate the effectiveness of IVUS-guided filter placement in critically ill patients with experience now extending out 5 years from implementation. Methods: All patients undergoing bedside IVUS-guided IVCF placement from 2008 to 2012 were identified. Records were reviewed on the basis of IVCF reporting standards. Outcomes data including technical success, complications, and mortality were analyzed at 30 days. Results: During the 5-year period, 398 patients underwent attempted bedside IVCF placement by IVUS. Technical feasibility was possible in 396 cases (99.5%); two bedside procedures were aborted because of inadequate IVUS visualization. Overall technical success was achieved in 393 of 396 (99.2%), with
malpositioned IVCF in three cases. An optional IVCF was used in 372 (93.9%) and a permanent IVCF in 24 (6.1%). Singlepuncture technique was performed in 388 (97.4%); additional dual access was required in 10 (2.6%). Periprocedural complications were rare (3.0%) and included malpositioning that required retrieval and repositioning or an additional IVCF (3), filter tilt $20 degrees (4), arteriovenous fistulas (2), insertion site thrombosis (2), and hematoma (1). Comparison of the first 100 procedures performed within the sample population with the last 100 procedures revealed an overall success rate of 96% in the first 100 compared with 100% in the last 100 (P [ .043). There were no deaths related to pulmonary embolism or IVCFrelated problems. Conclusions: On the basis of 5 years of experience with bedside IVCF placement in critically ill patients, the IVUS-guided IVCF technique continues to be a safe and effective option in this highrisk population, with a time-dependent improvement in outcome measures. (J Vasc Surg: Venous and Lym Dis 2014;-:1-6.)
Inferior vena cava filter (IVCF) placement by intravascular ultrasound (IVUS) imaging has been previously described and is being used with increasing frequency, especially in critically ill patients. Initial studies of these techniques have demonstrated technical feasibility and procedural efficacy.1-4 Bedside placement also eliminates the risks, costs, and resource use required to transport a patient to the operating room or fluoroscopy suite.5-7 We have previously reported successful prospective implementation of an algorithm for bedside IVCF placement by IVUS, thereby limiting the need for transport of critically ill patients.8 However, this preliminary experience was based on analysis of the initial year of protocol
implementation. The purpose of this study was to evaluate our experience with IVUS-guided IVCF placement during a longer time to determine procedural effectiveness and complication rates in this larger population sample and to confirm that safety and efficacy are sustainable.
From the Division of Vascular Surgery and Endovascular Therapy, University of Alabama at Birmingham. Author conflict of interest: none. Presented as a free paper at the Thirty-eighth Annual Meeting of the Southern Association for Vascular Surgery, The Breakers, Palm Beach, Fla, January 15-18, 2014. Reprint requests: Marc A. Passman, MD, Division of Vascular Surgery and Endovascular Therapy, University of Alabama at Birmingham, 503 Boshell Bldg, 1720 2nd Ave S, Birmingham, AL 35294-0012 (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the Journal policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 2213-333X/$36.00 Copyright Ó 2014 by the Society for Vascular Surgery. http://dx.doi.org/10.1016/j.jvsv.2014.04.007
METHODS All consecutive bedside IVCF placements by IVUS from January 1, 2008, to December 31, 2012, were identified from a prospectively maintained institutional single-center registry. This sample population included overlap with our previously published study on protocol implementation with patients from January 1 to December 31, 2008.8 Records were reviewed on the basis of IVCF reporting standards.9 Patient demographics, IVCF indications, available preoperative images, procedural data, periprocedural and late complications (any serious device-related or adverse event, such as death), pulmonary embolism (PE), groin complications, filter migration, erosion, thrombosis, and mortality were determined. Filter algorithm. A clinical decision-making algorithm for bedside IVCF placement by IVUS imaging has been previously described.8 Patients were initially evaluated for placement of an IVCF on the basis of evidencebased guidelines including standard indications of deep venous thrombosis (DVT) or PE with a contraindication to anticoagulation or recurrent PE despite adequate anticoagulation. Additional consideration for prophylactic IVCF placement was made on a selective basis for patients at 1
2 Glocker et al
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS --- 2014
Fig 1. Intravascular ultrasound (IVUS) images of inferior vena cava (IVC) showing anatomic landmarks. A, Right atrium. B, Hepatic veins. C, Left renal vein. D, Right renal vein. E, Measurement of infrarenal IVC diameter.
high risk for venous thromboembolism (VTE) with contraindications to pharmacologic preventive measures.10,11 A permanent IVCF (stainless steel Greenfield filter; Boston Scientific, Natick, Mass) was placed when the patient’s risk for VTE was for an extended time. Optional IVCFs (Celect Filter and Günther Tulip Filter; Cook Inc, Bloomington, Ind) were used when the patient’s VTE risk was determined to be transient and when there was a defined retrieval end point within 6 months of placement. The decision to place the IVCF at the bedside by IVUS was based on an overall assessment of the risks, benefits, and feasibility. Patients who were critically ill, with increased injury severity, or potentially unstable for transport were considered for bedside IVCF placement. Preprocedure lower extremity duplex ultrasound examinations were performed to evaluate for the presence of pre-existing DVT and potential thrombus at the intended access site. Computed tomography scans already in the imaging system were reviewed before IVCF placement to identify any potential problems that would preclude bedside IVCF placement, including inferior vena cava (IVC) or renal vein anomalies, IVC thrombosis, and IVC diameter of >28 mm. Bedside IVCF placement technique with IVUS. Our technique for IVCF placement has been previously described and was standardized among placement surgeons during the protocol implementation phase and was consistent for the entire study period.8,12 After sterile preparation,
percutaneous common femoral vein access was obtained with placement of an 8F 10-cm sheath. Right femoral access was preferred if possible, providing a straighter path to the IVC, but the left common femoral vein was used if thrombus was present on the right. A 0.035-inch guidewire was advanced into the IVC to a length that corresponded to an external measurement from groin to heart. The IVUS probe (Volcano s5 with a Visions PV8.2F catheter; Volcano Corp, Rancho Cordova, Calif) was advanced over the wire to the level of the right atrium, followed by a slow pullback to allow identification of anatomic landmarks, including the right atrium, hepatic veins, renal veins, and iliac vein confluence (Fig 1). The IVUS probe was then positioned immediately caudal to the lowest renal vein, and a diameter measurement was taken to ensure that the caval diameter was <28 mm before the IVCF kit was opened onto the field. If there was difficulty in determining the location of iliac confluence or renal vein levels, contralateral dual access was obtained to allow more direct positioning of catheters. If anatomic landmarks were not able to be adequately visualized with IVUS, the procedure was aborted and IVCF placement by contrast venography was performed at a later time. For the Günther Tulip and Celect filters, the 8F 10-cm sheath was exchanged for the 8F Cook sheath, which was advanced over the wire to a length corresponding to the position of the renal veins. More precise localization of the end of the sheath at the renal veins was confirmed by
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS Volume -, Number -
Table. Patient demographics and inferior vena cava filter (IVCF) placement data Patient characteristics Male:female Age, years Mean 6 standard deviation Range Indication for IVCF placement VTE prophylaxis DVT with contraindication to anticoagulation PE despite anticoagulation Type of IVCF placed Cook Celect Cook Günther Tulip Greenfield Total IVCFs placed by IVUS 2008 2009 2010 2011 2012
N ¼ 398 279:119 (70%:30%) 76.5 6 19.4 14-91 262 (66%) 91 (23%) 45 (11%) 222 (56%) 152 (38%) 24 (6%) 109 129 65 58 35
DVT, Deep venous thrombosis; IVUS, intravascular ultrasound; PE, pulmonary embolism; VTE, venous thromboembolism.
IVUS, which was positioned at the renal vein as the sheath was moved back and forth, with the end of the sheath visualized as a change in imaging intensity. The IVUS catheter and wire were then removed, and the filter delivery catheter was advanced into the sheath and mated appropriately on the basis of predetermined marks on the catheter, thereby aligning the apical hook with the renal vein. With the filter delivery catheter held steady, the sheath was then pulled back to the second mark and the filter deployed. For the Greenfield filter, the 15F sheath was advanced to the level of the renal vein as confirmed by IVUS. The sheath was then pulled back over the IVUS catheter a distance of 7 cm, corresponding to the distance that the filter delivery catheter protrudes from the delivery sheath during placement. The IVUS catheter was removed, but the wire was left in place. The filter delivery catheter was then loaded over the wire and Luer locked to the sheath, and the Greenfield filter was deployed. After IVCF deployment, the delivery catheter was withdrawn. For postdeployment imaging, the IVUS catheter was carefully advanced through the IVCF to confirm position of the IVCF based on apposition of the filter legs to the IVC wall and the level of the filter apex in relation to the level of the lowest renal vein. If any resistance was felt on passage of the IVUS catheter, any further advancement was stopped to avoid IVCF dislodgment. The sheath was removed and gentle pressure held to ensure hemostasis. A postprocedure abdominal radiograph was obtained to document position and alignment of the IVCF relative to the vertebral body column. Statistical analysis. Technical feasibility was defined as the intent to place the IVCF at the bedside with IVUS based on adequate access and visualization by IVUS. Technical success was defined as an appropriately aligned IVCF, with the tip at the renal vein level and full expansion of the
Glocker et al 3
filter base in the infrarenal IVC confirmed directly by IVUS. A malpositioned filter was defined as a suprarenal position (filter base above the level of the renal veins), tilt of the IVCF of more than 20 degrees relative to the spinous processes, or a filter deployed within the iliac system. The c2 (or Fisher exact test where appropriate) was used to compare frequencies between groups. A P value < .05 was used to indicate statistical significance. SPSS 20.0 for Windows (IBM SPSS, New York, NY) was used for data analysis. This study was approved by the University of Alabama at Birmingham internal Institutional Review Board. RESULTS During the 5-year period, 398 patients underwent intended bedside IVCF placement by IVUS. Mean age was 76.5 6 19 (standard deviation) years (range, 14-91 years; median, 46 years), with 279 male patients (70%) and 119 female patients (30%). Indications for placement included VTE prophylaxis in the setting of multisystem injury with a contraindication to pharmacologic anticoagulation in 262 patients (66%), DVT with a contraindication to anticoagulation in 91 patients (23%), and PE despite therapeutic anticoagulation in 45 patients (11%). A single-puncture technique was used in 388 patients (97.4%), with dual access required in 10 (2.6%). For IVCF device selection, 372 (94%) were optional (222 Celect [56%] and 150 Gunther Tülip [38%]) and 24 (6%) were permanent (Greenfield). In 2008, 109 IVCFs were placed; that number increased to 129 in 2009 and then decreased to 65 in 2010, 58 in 2011, and 35 in 2012 (Table). Among the 398 patients receiving IVCFs, 237 (59.5%) had major limb trauma, 168 (42.2%) were recently postoperative from a major surgical procedure, 116 (29.1%) had a vertebral fracture, 74 (18.8%) had a spinal cord injury, and 201 (50.5%) had a closed head injury. IVCF placement was technically feasible in 396 patients (99.5%); two bedside procedures were aborted because of inadequate IVUS visualization. Of these 396 patients, overall technical success was achieved in 393 patients (99.2%) (Fig 2). Among the three technical failures, the IVCF was placed in the left common iliac vein in two patients; one was removed percutaneously followed by replacement, and the other was left in place with placement of a separate right common iliac vein IVCF. The third patient had a right common iliac vein filter placed that was not able to be removed percutaneously and required an exploratory laparotomy with concern for IVC injury after the IVCF was noted to be partially extraluminal on venography. In the remaining 393 successfully placed IVCFs, the tip was located at the T12 vertebral body in 15 patients (3.8%), L1 in 131 patients (33.3%), L2 in 199 patients (50.7%), L3 in 40 patients (10.2%), and L4 in 8 patients (2.0%). Periprocedural complications occurred in 12 patients (3.0%), including the three previously mentioned malpositioned IVCFs, four IVCFs having a tilt >20 degrees, two arteriovenous fistulas, two insertion site thromboses, and one access site hematoma. No long-term complications were identified in a query of all postplacement clinical visit notes within our electronic medical record. There were no
4 Glocker et al
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS --- 2014
Fig 2. Technical success and feasibility of bedside placement of inferior vena cava filter (IVCF).
deaths related to VTE and no IVCF-related complications. Thirty patients (7.5%) died within 30 days of IVCF placement because of underlying illness or injury. Comparison of the first 100 procedures performed within the sample population with the last 100 procedures revealed procedural success rate of 96% in the first 100 compared with 100% in the last 100, a difference that was significant (P ¼ .043). There was also a trend toward decreased complication rates, with a 7.0% complication rate occurring during the first 100 procedures and 2.0% occurring during the last 100 procedures, although this difference was not significant (P ¼ .08). DISCUSSION Transport of critically ill patients within the hospital, even by specially trained intensive care unit staff, carries risk.13 Stearley6 noted that complications can occur during 15.5% of patient transfers; these include vital sign derangements, airway disruptions, cardiopulmonary arrest, and missed medication doses. Szem et al7 found that patients requiring transport represented a more critically ill subpopulation with a higher overall mortality compared with patients who did not require transport. Given the issues with patient transport in regard to patient safety, bedside IVCF placement offers a potential advantage by minimizing the risks of transport as well as administration of contrast material and fluoroscopy. Bedside IVCF placement techniques with both transabdominal duplex ultrasound and IVUS have been previously shown to be safe and efficacious.4,8,14 A clinical decision algorithm for bedside IVCF placement by IVUS was effectively implemented at our institution in 2008.8 The initial decision to use IVUS was based on improved reported technical feasibility of IVUS over transabdominal ultrasound because of limitations in visualization due to body habitus and bowel gas. Conners et al14 reported a series of IVCFs placed by transabdominal duplex ultrasound with technical feasibility in 88% and technical success rates of 98% when IVCF placement was technically feasible. In contrast, collective published experience for IVCF placement by IVUS has shown higher technical feasibility and success rates ranging from 98.1% to 100% and 92.0% to 97.2% respectively.1,8 In our initial published report with prospective implementation of our bedside IVCF placement protocol by
IVUS during 1 year, technical feasibility was 98.0% and technical success was 97.2%. These high rates were maintained in this current 5-year follow-up study, which is the largest published experience with IVCF placement by IVUS to date, with reported technical feasibility of 99.4% and success of 99.2%.8 In comparison of the first 100 patients and the last 100 patients in our cohort, technical success was also noted to improve with time, approaching 100% in the later portion of the experience, and there was a trend toward improved complication rates. With an overall complication rate of 3%, our series of IVUS-guided IVCF placements compares well with large series of venographically placed IVCFs.15-17 The described technique for IVCF placement by IVUS at our institution has been consistent during the study period. With our large experience, there are some important technical notes that have been learned to assist with successful placement and that are likely to have contributed to our improved success rate during the study period. A thorough understanding of IVCF device and deployment systems is critical to safe performance of the procedures. We prefer to use the single-puncture (97.4%) technique with right-sided access, allowing a more direct straight alignment into the IVC. We reserved dual access (2.6%) for difficult visualization of anatomic landmarks or to facilitate confirmation of the iliac confluence with direct imaging of a contralateral wire. For single-puncture techniques, because a low-megahertz IVUS probe is needed for adequate imaging of the entire IVC diameter plus adjacent landmarks, an 8F sheath access is required for imaging. With use of the Greenfield, Günther Tulip, and Celect filters, the larger sheath required in these kits facilitates single-puncture access. Although lower profile devices can still be placed by IVUS, a dual access technique is required but can potentially increase access site complications.18 Because the IVCF is not deployed under direct vision with the single-puncture technique, the key is correct positioning of the sheath at the renal level. Because the delivery catheters either interlock with the sheath (Greenfield) or have marks noting position in the sheath (Günther Tulip and Celect), if the sheath is correctly positioned, the filter will be correctly positioned on deployment. Deviation from understanding this relationship between the IVCF delivery catheter and the sheath mechanism will lead to potential misdeployment or malpositioning. The three malpositioned IVCFs and four tilted IVCFs in our study were probably due to incorrect identification of IVC anatomy or deviations in technique. For providers looking to begin placing IVCFs by IVUS, a thorough understanding of the IVUS device and its use is critical. We frequently use IVUS in other aspects of our vascular practice and think that this familiarity is of assistance in placing an IVCF. Initially, if a placement physician is uncertain about the anatomy with use of IVUS, the IVCF can be placed by a dual modality approach in the angiography suite, allowing real-time correlation between venographic and IVUS representations of the patient’s anatomy. Whereas this study focuses on IVCF placement techniques by IVUS and associated feasibility and success, it does not address the role of IVCF in the current algorithm
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS Volume -, Number -
of IVCF use. With the recent Food and Drug Administration warning about retrievable IVCF use and updated 2012 American College of Chest Physician evidence-based guidelines recommending against IVCF use for prophylaxis, there has been a general trend nationally for decreased IVCF use, especially for venous thromboprophylaxis in high-risk patients.10,19 Interestingly, during the course of our study period, there was a decrease in the number of IVCFs placed per year. Although no defined changes in our IVCF placement algorithm took place, we think that this trend probably represents an expanded role of anticoagulation for thromboprophylaxis and contraction of IVCF use in critically ill patients at our institution. There are several limitations with our study. As a single-arm retrospective experience, there is no comparable control group (such as venography-guided placement) for comparison. There is also selection bias based on protocol preprocedure screening of bedside feasibility. With preprocedure screening of available abdominal CT scans, there is improved preprocedure likelihood of success and technical feasibility. Finally, IVCF-defined end points were obtained when available from procedure reports and postprocedure imaging; however, long-term follow-up was limited to clinically relevant need for additional venous evaluation outside our standard protocol. CONCLUSIONS This study is the largest experience to date with bedside IVCF placement by IVUS in critically ill patients and confirms that bedside IVCF placement by IVUS continues to be a safe and effective option in this high-risk population, with a time-dependent improvement in outcome measures. AUTHOR CONTRIBUTIONS Conception and design: RG, OA, MAP Analysis and interpretation: RG, OA, ZN, BP, MP, TM, WJ, MAP Data collection: RG, OA, MAP Writing the article: RG, MAP Critical revision of the article: RG, OA, ZN, BP, MP, TM, WJ, MAP Final approval of the article: MAP Statistical analysis: ZN, MAP Obtained funding: Not applicable Overall responsibility: MAP REFERENCES 1. Ebaugh J, Chiou A, Morasch M, Matsumura J, Pearce W. Bedside vena cava filter placement guided with intravascular ultrasound. J Vasc Surg 2001;34:21-6.
Glocker et al 5
2. Garrett J, Passman MA, Guzman RJ, Dattilo JB, Naslund TC. Expanding options for bedside placement of inferior vena cava filters with intravascular ultrasound when transabdominal Duplex ultrasound imaging is inadequate. Ann Vasc Surg 2004;18:329-34. 3. Wellons ED, Rosenthal D, Shuler FW, Levitt AB, Matsumura J, Henderson VJ. Real-time intravascular ultrasound guided placement of a removable inferior vena cava filter. J Trauma 2004;57:20-5. 4. Oppat W, Chiou A, Matsumura J. Intravascular ultrasound guided vena cava filter placement. J Endovasc Surg 1999;6:285-7. 5. Karmy-Jones R, Jurkovich GJ, Velmahos GC, Burdick T, Spaniolas K, Todd SR. Practice patterns and outcomes of retrievable vena cava filters in trauma patients: an AAST multicenter study. J Trauma 2007;62: 17-24. 6. Stearley HE. Patients’ outcomes: intrahospital transportation and monitoring of critically ill patients by a specially trained ICU and nursing staff. Am J Crit Care 1998;7:282-7. 7. Szem JW, Hydo LJ, Fischer E, Kapur S, Klemperer J, Barie BS. High risk intrahospital transport of critically ill patients: safety and outcome of the necessary “road trip”. Crit Care Med 1995;23:1660-6. 8. Killingsworth CD, Taylor SM, Patterson MA, Weinberg JA, McGwin G, Melton SM, et al. Prospective implementation of an algorithm for bedside intravascular ultrasound-guided filter placement in critically ill patients. J Vasc Surg 2010;51:1215-21. 9. Participants in the Vena Caval Filter Consensus Conference. Recommended reporting standards for vena caval filter placement and patient follow up. J Vasc Surg 1999;30:573-9. 10. Kearon C, Akl EA, Comerota AJ, Prandoni P, Bounameaux H, Goldhaber SZ, et al. Antithrombotic therapy for VTE disease: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141:e419S-94S. 11. Rogers FB, Cipolle MD, Velmahos G, Rozycki G, Luchette FA. Practice management guidelines for the prevention of venous thromboembolism in trauma patients: the EAST Practice Management Guidelines Work Group. J Trauma 2002;53:142-64. 12. Passman MA. IVUS-guided vena cava filter placement: technique and clinical decision algorithm. Vasc Dis Manag 2011;3:57-61. 13. Beckmann U, Gillies D, Berenholtz S, Wu A, Pronovost P. Incidents relating to the intra-hospital transfer of critically ill patients. Intensive Care Med 2004;30:1579-85. 14. Conners M, Becker S, Guzman R, Passman M, Pierce R, Kelly T, et al. Duplex scan-directed placement of inferior vena cava filters: a five-year institutional experience. J Vasc Surg 2002;35:286-91. 15. Nazzal M, Chan E, Nazzal M, Abbas J, Erikson G, Sedige S, et al. Complications related to inferior vena cava filters: a single-center experience. Ann Vasc Surg 2010;24:480-6. 16. Joels CS, Sing RF, Heniford BT. Complications of inferior vena cava filters. Am Surg 2003;69:654-9. 17. Corriere MA, Passman MA, Guzman RJ, Dattilo JB, Naslund TC. Comparison of bedside transabdominal duplex ultrasound versus contrast venography for inferior vena cava filter placement: what is the best imaging modality? Ann Vasc Surg 2005;19:229-34. 18. Jacobs DL, Motaganahalli RL, Peterson BG. Bedside vena cava filter placement with intravascular ultrasound: a simple, accurate, single venous access method. J Vasc Surg 2007;46:1284-6. 19. Inferior vena cava (IVC) filters: initial communication: risk of adverse events with long term use. Available at: http://wwwfdagov/Safety/ MedWatch/SafetyInformation/SafetyAlertsforHumanMedicalProducts/ ucm221707htm. Accessed September 1, 2013. Submitted Feb 6, 2014; accepted Apr 27, 2014.
DISCUSSION Dr Lazar J. Greenfield, Sr (Tucson, Ariz). I wish to compliment Dr Glocker on his presentation and commend the authors on their excellent results. During the past 40 years, filter evolution has followed the unusual path of little change in basic design but
remarkable improvements in placement techniques, moving from operating room to radiology suite to bedside. Of course, when you see 100% success in a series, you wonder about selection, so my first question is, How many filter patients were excluded and
6 Glocker et al
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS --- 2014
for what reasons? Did you exclude the patients with venous anomalies seen on computed tomography (CT) scan? Although you mention retrieval only of misplaced filters, your recent publication on 121 retrieval attempts during the same period had a 76% success rate, leaving 24% as longterm implants. In our current financial environment, how costeffective is this two-procedure approach as well as your selection process for venous thromboembolism risk when none actually developed that complication? Finally, having told those patients with venous thromboembolism of the need for a filter, what level of risk remains for those who no longer have that protection? Dr Roan J. Glocker. Thank you, Dr Greenfield. With regard to your first question, I think it relates to patient selection. Unfortunately, we do not know the overall number of patients we evaluated for bedside filter placement, but during the course of our study period, we did note that appropriate patient selection is perhaps the most important part of procedural success. The patients in whom we placed these filters are, as we saw, a significantly traumatically injured population, many of them with blunt trauma. Those patients almost always have an abdominal CT scan on admission to the hospital, and a thorough review of that abdominal CT scan is critical for bedside filter placement. We
believe that we have a reasonably high degree of procedural success before we begin a bedside filter placement on the basis of our view of that CT scan, ensuring that there are no venous anomalies and making sure the cava is of adequate diameter. Your second question relates to the venous anomalies. Those patients who did have venous anomalies were not considered for bedside filter placement by intravascular ultrasound, and those patients, if their anomaly was such that it would allow filter placement, underwent filter placement in the angiography suite. The question relating to filter retrieval is a focus of current work for us. We thought that looking at the rate of retrievals in this study was a little bit outside the scope. Our previous work looking at retrieval rates kind of compares apples and oranges because that patient cohort includes all filters we placed including those placed in the operating room. With regard to the need for filter placement and the overall trend toward placement of a decreasing number of filters, when we get these consults, we make a decision about whether to place the filter in conjunction with our critical care colleagues. I think the trend toward placing a decreased number of filters represents a perhaps better overall understanding of which patients benefit from a prophylactic filter placement and a trend toward decreasing the number of those prophylactic filters placed.
Bedside inferior vena cava filter placement by intravascular ultrasound in critically ill patients is safe and effective for an extended time Roan J. Glocker, MD, Oluwafunmi Awonuga, MD, Zdenek Novak, MD, Benjamin J. Pearce, MD, Mark Patterson, MD, Thomas C. Matthews, MD, William D. Jordan, MD, and Marc A. Passman, MD, Birmingham, Ala Background: Bedside inferior vena cava filter (IVCF) placement by intravascular ultrasound (IVUS) guidance has previously been shown to be a safe and effective technique, especially for critically ill patients, with initial experience of a prospectively implemented algorithm. The purpose of this study was to evaluate the effectiveness of IVUS-guided filter placement in critically ill patients with experience now extending out 5 years from implementation. Methods: All patients undergoing bedside IVUS-guided IVCF placement from 2008 to 2012 were identified. Records were reviewed on the basis of IVCF reporting standards. Outcomes data including technical success, complications, and mortality were analyzed at 30 days. Results: During the 5-year period, 398 patients underwent attempted bedside IVCF placement by IVUS. Technical feasibility was possible in 396 cases (99.5%); two bedside procedures were aborted because of inadequate IVUS visualization. Overall technical success was achieved in 393 of 396 (99.2%), with
malpositioned IVCF in three cases. An optional IVCF was used in 372 (93.9%) and a permanent IVCF in 24 (6.1%). Singlepuncture technique was performed in 388 (97.4%); additional dual access was required in 10 (2.6%). Periprocedural complications were rare (3.0%) and included malpositioning that required retrieval and repositioning or an additional IVCF (3), filter tilt $20 degrees (4), arteriovenous fistulas (2), insertion site thrombosis (2), and hematoma (1). Comparison of the first 100 procedures performed within the sample population with the last 100 procedures revealed an overall success rate of 96% in the first 100 compared with 100% in the last 100 (P [ .043). There were no deaths related to pulmonary embolism or IVCFrelated problems. Conclusions: On the basis of 5 years of experience with bedside IVCF placement in critically ill patients, the IVUS-guided IVCF technique continues to be a safe and effective option in this highrisk population, with a time-dependent improvement in outcome measures. (J Vasc Surg: Venous and Lym Dis 2014;-:1-6.)
Inferior vena cava filter (IVCF) placement by intravascular ultrasound (IVUS) imaging has been previously described and is being used with increasing frequency, especially in critically ill patients. Initial studies of these techniques have demonstrated technical feasibility and procedural efficacy.1-4 Bedside placement also eliminates the risks, costs, and resource use required to transport a patient to the operating room or fluoroscopy suite.5-7 We have previously reported successful prospective implementation of an algorithm for bedside IVCF placement by IVUS, thereby limiting the need for transport of critically ill patients.8 However, this preliminary experience was based on analysis of the initial year of protocol
implementation. The purpose of this study was to evaluate our experience with IVUS-guided IVCF placement during a longer time to determine procedural effectiveness and complication rates in this larger population sample and to confirm that safety and efficacy are sustainable.
From the Division of Vascular Surgery and Endovascular Therapy, University of Alabama at Birmingham. Author conflict of interest: none. Presented as a free paper at the Thirty-eighth Annual Meeting of the Southern Association for Vascular Surgery, The Breakers, Palm Beach, Fla, January 15-18, 2014. Reprint requests: Marc A. Passman, MD, Division of Vascular Surgery and Endovascular Therapy, University of Alabama at Birmingham, 503 Boshell Bldg, 1720 2nd Ave S, Birmingham, AL 35294-0012 (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the Journal policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 2213-333X/$36.00 Copyright Ó 2014 by the Society for Vascular Surgery. http://dx.doi.org/10.1016/j.jvsv.2014.04.007
METHODS All consecutive bedside IVCF placements by IVUS from January 1, 2008, to December 31, 2012, were identified from a prospectively maintained institutional single-center registry. This sample population included overlap with our previously published study on protocol implementation with patients from January 1 to December 31, 2008.8 Records were reviewed on the basis of IVCF reporting standards.9 Patient demographics, IVCF indications, available preoperative images, procedural data, periprocedural and late complications (any serious device-related or adverse event, such as death), pulmonary embolism (PE), groin complications, filter migration, erosion, thrombosis, and mortality were determined. Filter algorithm. A clinical decision-making algorithm for bedside IVCF placement by IVUS imaging has been previously described.8 Patients were initially evaluated for placement of an IVCF on the basis of evidencebased guidelines including standard indications of deep venous thrombosis (DVT) or PE with a contraindication to anticoagulation or recurrent PE despite adequate anticoagulation. Additional consideration for prophylactic IVCF placement was made on a selective basis for patients at 1
2 Glocker et al
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS --- 2014
Fig 1. Intravascular ultrasound (IVUS) images of inferior vena cava (IVC) showing anatomic landmarks. A, Right atrium. B, Hepatic veins. C, Left renal vein. D, Right renal vein. E, Measurement of infrarenal IVC diameter.
high risk for venous thromboembolism (VTE) with contraindications to pharmacologic preventive measures.10,11 A permanent IVCF (stainless steel Greenfield filter; Boston Scientific, Natick, Mass) was placed when the patient’s risk for VTE was for an extended time. Optional IVCFs (Celect Filter and Günther Tulip Filter; Cook Inc, Bloomington, Ind) were used when the patient’s VTE risk was determined to be transient and when there was a defined retrieval end point within 6 months of placement. The decision to place the IVCF at the bedside by IVUS was based on an overall assessment of the risks, benefits, and feasibility. Patients who were critically ill, with increased injury severity, or potentially unstable for transport were considered for bedside IVCF placement. Preprocedure lower extremity duplex ultrasound examinations were performed to evaluate for the presence of pre-existing DVT and potential thrombus at the intended access site. Computed tomography scans already in the imaging system were reviewed before IVCF placement to identify any potential problems that would preclude bedside IVCF placement, including inferior vena cava (IVC) or renal vein anomalies, IVC thrombosis, and IVC diameter of >28 mm. Bedside IVCF placement technique with IVUS. Our technique for IVCF placement has been previously described and was standardized among placement surgeons during the protocol implementation phase and was consistent for the entire study period.8,12 After sterile preparation,
percutaneous common femoral vein access was obtained with placement of an 8F 10-cm sheath. Right femoral access was preferred if possible, providing a straighter path to the IVC, but the left common femoral vein was used if thrombus was present on the right. A 0.035-inch guidewire was advanced into the IVC to a length that corresponded to an external measurement from groin to heart. The IVUS probe (Volcano s5 with a Visions PV8.2F catheter; Volcano Corp, Rancho Cordova, Calif) was advanced over the wire to the level of the right atrium, followed by a slow pullback to allow identification of anatomic landmarks, including the right atrium, hepatic veins, renal veins, and iliac vein confluence (Fig 1). The IVUS probe was then positioned immediately caudal to the lowest renal vein, and a diameter measurement was taken to ensure that the caval diameter was <28 mm before the IVCF kit was opened onto the field. If there was difficulty in determining the location of iliac confluence or renal vein levels, contralateral dual access was obtained to allow more direct positioning of catheters. If anatomic landmarks were not able to be adequately visualized with IVUS, the procedure was aborted and IVCF placement by contrast venography was performed at a later time. For the Günther Tulip and Celect filters, the 8F 10-cm sheath was exchanged for the 8F Cook sheath, which was advanced over the wire to a length corresponding to the position of the renal veins. More precise localization of the end of the sheath at the renal veins was confirmed by
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS Volume -, Number -
Table. Patient demographics and inferior vena cava filter (IVCF) placement data Patient characteristics Male:female Age, years Mean 6 standard deviation Range Indication for IVCF placement VTE prophylaxis DVT with contraindication to anticoagulation PE despite anticoagulation Type of IVCF placed Cook Celect Cook Günther Tulip Greenfield Total IVCFs placed by IVUS 2008 2009 2010 2011 2012
N ¼ 398 279:119 (70%:30%) 76.5 6 19.4 14-91 262 (66%) 91 (23%) 45 (11%) 222 (56%) 152 (38%) 24 (6%) 109 129 65 58 35
DVT, Deep venous thrombosis; IVUS, intravascular ultrasound; PE, pulmonary embolism; VTE, venous thromboembolism.
IVUS, which was positioned at the renal vein as the sheath was moved back and forth, with the end of the sheath visualized as a change in imaging intensity. The IVUS catheter and wire were then removed, and the filter delivery catheter was advanced into the sheath and mated appropriately on the basis of predetermined marks on the catheter, thereby aligning the apical hook with the renal vein. With the filter delivery catheter held steady, the sheath was then pulled back to the second mark and the filter deployed. For the Greenfield filter, the 15F sheath was advanced to the level of the renal vein as confirmed by IVUS. The sheath was then pulled back over the IVUS catheter a distance of 7 cm, corresponding to the distance that the filter delivery catheter protrudes from the delivery sheath during placement. The IVUS catheter was removed, but the wire was left in place. The filter delivery catheter was then loaded over the wire and Luer locked to the sheath, and the Greenfield filter was deployed. After IVCF deployment, the delivery catheter was withdrawn. For postdeployment imaging, the IVUS catheter was carefully advanced through the IVCF to confirm position of the IVCF based on apposition of the filter legs to the IVC wall and the level of the filter apex in relation to the level of the lowest renal vein. If any resistance was felt on passage of the IVUS catheter, any further advancement was stopped to avoid IVCF dislodgment. The sheath was removed and gentle pressure held to ensure hemostasis. A postprocedure abdominal radiograph was obtained to document position and alignment of the IVCF relative to the vertebral body column. Statistical analysis. Technical feasibility was defined as the intent to place the IVCF at the bedside with IVUS based on adequate access and visualization by IVUS. Technical success was defined as an appropriately aligned IVCF, with the tip at the renal vein level and full expansion of the
Glocker et al 3
filter base in the infrarenal IVC confirmed directly by IVUS. A malpositioned filter was defined as a suprarenal position (filter base above the level of the renal veins), tilt of the IVCF of more than 20 degrees relative to the spinous processes, or a filter deployed within the iliac system. The c2 (or Fisher exact test where appropriate) was used to compare frequencies between groups. A P value < .05 was used to indicate statistical significance. SPSS 20.0 for Windows (IBM SPSS, New York, NY) was used for data analysis. This study was approved by the University of Alabama at Birmingham internal Institutional Review Board. RESULTS During the 5-year period, 398 patients underwent intended bedside IVCF placement by IVUS. Mean age was 76.5 6 19 (standard deviation) years (range, 14-91 years; median, 46 years), with 279 male patients (70%) and 119 female patients (30%). Indications for placement included VTE prophylaxis in the setting of multisystem injury with a contraindication to pharmacologic anticoagulation in 262 patients (66%), DVT with a contraindication to anticoagulation in 91 patients (23%), and PE despite therapeutic anticoagulation in 45 patients (11%). A single-puncture technique was used in 388 patients (97.4%), with dual access required in 10 (2.6%). For IVCF device selection, 372 (94%) were optional (222 Celect [56%] and 150 Gunther Tülip [38%]) and 24 (6%) were permanent (Greenfield). In 2008, 109 IVCFs were placed; that number increased to 129 in 2009 and then decreased to 65 in 2010, 58 in 2011, and 35 in 2012 (Table). Among the 398 patients receiving IVCFs, 237 (59.5%) had major limb trauma, 168 (42.2%) were recently postoperative from a major surgical procedure, 116 (29.1%) had a vertebral fracture, 74 (18.8%) had a spinal cord injury, and 201 (50.5%) had a closed head injury. IVCF placement was technically feasible in 396 patients (99.5%); two bedside procedures were aborted because of inadequate IVUS visualization. Of these 396 patients, overall technical success was achieved in 393 patients (99.2%) (Fig 2). Among the three technical failures, the IVCF was placed in the left common iliac vein in two patients; one was removed percutaneously followed by replacement, and the other was left in place with placement of a separate right common iliac vein IVCF. The third patient had a right common iliac vein filter placed that was not able to be removed percutaneously and required an exploratory laparotomy with concern for IVC injury after the IVCF was noted to be partially extraluminal on venography. In the remaining 393 successfully placed IVCFs, the tip was located at the T12 vertebral body in 15 patients (3.8%), L1 in 131 patients (33.3%), L2 in 199 patients (50.7%), L3 in 40 patients (10.2%), and L4 in 8 patients (2.0%). Periprocedural complications occurred in 12 patients (3.0%), including the three previously mentioned malpositioned IVCFs, four IVCFs having a tilt >20 degrees, two arteriovenous fistulas, two insertion site thromboses, and one access site hematoma. No long-term complications were identified in a query of all postplacement clinical visit notes within our electronic medical record. There were no
4 Glocker et al
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS --- 2014
Fig 2. Technical success and feasibility of bedside placement of inferior vena cava filter (IVCF).
deaths related to VTE and no IVCF-related complications. Thirty patients (7.5%) died within 30 days of IVCF placement because of underlying illness or injury. Comparison of the first 100 procedures performed within the sample population with the last 100 procedures revealed procedural success rate of 96% in the first 100 compared with 100% in the last 100, a difference that was significant (P ¼ .043). There was also a trend toward decreased complication rates, with a 7.0% complication rate occurring during the first 100 procedures and 2.0% occurring during the last 100 procedures, although this difference was not significant (P ¼ .08). DISCUSSION Transport of critically ill patients within the hospital, even by specially trained intensive care unit staff, carries risk.13 Stearley6 noted that complications can occur during 15.5% of patient transfers; these include vital sign derangements, airway disruptions, cardiopulmonary arrest, and missed medication doses. Szem et al7 found that patients requiring transport represented a more critically ill subpopulation with a higher overall mortality compared with patients who did not require transport. Given the issues with patient transport in regard to patient safety, bedside IVCF placement offers a potential advantage by minimizing the risks of transport as well as administration of contrast material and fluoroscopy. Bedside IVCF placement techniques with both transabdominal duplex ultrasound and IVUS have been previously shown to be safe and efficacious.4,8,14 A clinical decision algorithm for bedside IVCF placement by IVUS was effectively implemented at our institution in 2008.8 The initial decision to use IVUS was based on improved reported technical feasibility of IVUS over transabdominal ultrasound because of limitations in visualization due to body habitus and bowel gas. Conners et al14 reported a series of IVCFs placed by transabdominal duplex ultrasound with technical feasibility in 88% and technical success rates of 98% when IVCF placement was technically feasible. In contrast, collective published experience for IVCF placement by IVUS has shown higher technical feasibility and success rates ranging from 98.1% to 100% and 92.0% to 97.2% respectively.1,8 In our initial published report with prospective implementation of our bedside IVCF placement protocol by
IVUS during 1 year, technical feasibility was 98.0% and technical success was 97.2%. These high rates were maintained in this current 5-year follow-up study, which is the largest published experience with IVCF placement by IVUS to date, with reported technical feasibility of 99.4% and success of 99.2%.8 In comparison of the first 100 patients and the last 100 patients in our cohort, technical success was also noted to improve with time, approaching 100% in the later portion of the experience, and there was a trend toward improved complication rates. With an overall complication rate of 3%, our series of IVUS-guided IVCF placements compares well with large series of venographically placed IVCFs.15-17 The described technique for IVCF placement by IVUS at our institution has been consistent during the study period. With our large experience, there are some important technical notes that have been learned to assist with successful placement and that are likely to have contributed to our improved success rate during the study period. A thorough understanding of IVCF device and deployment systems is critical to safe performance of the procedures. We prefer to use the single-puncture (97.4%) technique with right-sided access, allowing a more direct straight alignment into the IVC. We reserved dual access (2.6%) for difficult visualization of anatomic landmarks or to facilitate confirmation of the iliac confluence with direct imaging of a contralateral wire. For single-puncture techniques, because a low-megahertz IVUS probe is needed for adequate imaging of the entire IVC diameter plus adjacent landmarks, an 8F sheath access is required for imaging. With use of the Greenfield, Günther Tulip, and Celect filters, the larger sheath required in these kits facilitates single-puncture access. Although lower profile devices can still be placed by IVUS, a dual access technique is required but can potentially increase access site complications.18 Because the IVCF is not deployed under direct vision with the single-puncture technique, the key is correct positioning of the sheath at the renal level. Because the delivery catheters either interlock with the sheath (Greenfield) or have marks noting position in the sheath (Günther Tulip and Celect), if the sheath is correctly positioned, the filter will be correctly positioned on deployment. Deviation from understanding this relationship between the IVCF delivery catheter and the sheath mechanism will lead to potential misdeployment or malpositioning. The three malpositioned IVCFs and four tilted IVCFs in our study were probably due to incorrect identification of IVC anatomy or deviations in technique. For providers looking to begin placing IVCFs by IVUS, a thorough understanding of the IVUS device and its use is critical. We frequently use IVUS in other aspects of our vascular practice and think that this familiarity is of assistance in placing an IVCF. Initially, if a placement physician is uncertain about the anatomy with use of IVUS, the IVCF can be placed by a dual modality approach in the angiography suite, allowing real-time correlation between venographic and IVUS representations of the patient’s anatomy. Whereas this study focuses on IVCF placement techniques by IVUS and associated feasibility and success, it does not address the role of IVCF in the current algorithm
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS Volume -, Number -
of IVCF use. With the recent Food and Drug Administration warning about retrievable IVCF use and updated 2012 American College of Chest Physician evidence-based guidelines recommending against IVCF use for prophylaxis, there has been a general trend nationally for decreased IVCF use, especially for venous thromboprophylaxis in high-risk patients.10,19 Interestingly, during the course of our study period, there was a decrease in the number of IVCFs placed per year. Although no defined changes in our IVCF placement algorithm took place, we think that this trend probably represents an expanded role of anticoagulation for thromboprophylaxis and contraction of IVCF use in critically ill patients at our institution. There are several limitations with our study. As a single-arm retrospective experience, there is no comparable control group (such as venography-guided placement) for comparison. There is also selection bias based on protocol preprocedure screening of bedside feasibility. With preprocedure screening of available abdominal CT scans, there is improved preprocedure likelihood of success and technical feasibility. Finally, IVCF-defined end points were obtained when available from procedure reports and postprocedure imaging; however, long-term follow-up was limited to clinically relevant need for additional venous evaluation outside our standard protocol. CONCLUSIONS This study is the largest experience to date with bedside IVCF placement by IVUS in critically ill patients and confirms that bedside IVCF placement by IVUS continues to be a safe and effective option in this high-risk population, with a time-dependent improvement in outcome measures. AUTHOR CONTRIBUTIONS Conception and design: RG, OA, MAP Analysis and interpretation: RG, OA, ZN, BP, MP, TM, WJ, MAP Data collection: RG, OA, MAP Writing the article: RG, MAP Critical revision of the article: RG, OA, ZN, BP, MP, TM, WJ, MAP Final approval of the article: MAP Statistical analysis: ZN, MAP Obtained funding: Not applicable Overall responsibility: MAP REFERENCES 1. Ebaugh J, Chiou A, Morasch M, Matsumura J, Pearce W. Bedside vena cava filter placement guided with intravascular ultrasound. J Vasc Surg 2001;34:21-6.
Glocker et al 5
2. Garrett J, Passman MA, Guzman RJ, Dattilo JB, Naslund TC. Expanding options for bedside placement of inferior vena cava filters with intravascular ultrasound when transabdominal Duplex ultrasound imaging is inadequate. Ann Vasc Surg 2004;18:329-34. 3. Wellons ED, Rosenthal D, Shuler FW, Levitt AB, Matsumura J, Henderson VJ. Real-time intravascular ultrasound guided placement of a removable inferior vena cava filter. J Trauma 2004;57:20-5. 4. Oppat W, Chiou A, Matsumura J. Intravascular ultrasound guided vena cava filter placement. J Endovasc Surg 1999;6:285-7. 5. Karmy-Jones R, Jurkovich GJ, Velmahos GC, Burdick T, Spaniolas K, Todd SR. Practice patterns and outcomes of retrievable vena cava filters in trauma patients: an AAST multicenter study. J Trauma 2007;62: 17-24. 6. Stearley HE. Patients’ outcomes: intrahospital transportation and monitoring of critically ill patients by a specially trained ICU and nursing staff. Am J Crit Care 1998;7:282-7. 7. Szem JW, Hydo LJ, Fischer E, Kapur S, Klemperer J, Barie BS. High risk intrahospital transport of critically ill patients: safety and outcome of the necessary “road trip”. Crit Care Med 1995;23:1660-6. 8. Killingsworth CD, Taylor SM, Patterson MA, Weinberg JA, McGwin G, Melton SM, et al. Prospective implementation of an algorithm for bedside intravascular ultrasound-guided filter placement in critically ill patients. J Vasc Surg 2010;51:1215-21. 9. Participants in the Vena Caval Filter Consensus Conference. Recommended reporting standards for vena caval filter placement and patient follow up. J Vasc Surg 1999;30:573-9. 10. Kearon C, Akl EA, Comerota AJ, Prandoni P, Bounameaux H, Goldhaber SZ, et al. Antithrombotic therapy for VTE disease: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141:e419S-94S. 11. Rogers FB, Cipolle MD, Velmahos G, Rozycki G, Luchette FA. Practice management guidelines for the prevention of venous thromboembolism in trauma patients: the EAST Practice Management Guidelines Work Group. J Trauma 2002;53:142-64. 12. Passman MA. IVUS-guided vena cava filter placement: technique and clinical decision algorithm. Vasc Dis Manag 2011;3:57-61. 13. Beckmann U, Gillies D, Berenholtz S, Wu A, Pronovost P. Incidents relating to the intra-hospital transfer of critically ill patients. Intensive Care Med 2004;30:1579-85. 14. Conners M, Becker S, Guzman R, Passman M, Pierce R, Kelly T, et al. Duplex scan-directed placement of inferior vena cava filters: a five-year institutional experience. J Vasc Surg 2002;35:286-91. 15. Nazzal M, Chan E, Nazzal M, Abbas J, Erikson G, Sedige S, et al. Complications related to inferior vena cava filters: a single-center experience. Ann Vasc Surg 2010;24:480-6. 16. Joels CS, Sing RF, Heniford BT. Complications of inferior vena cava filters. Am Surg 2003;69:654-9. 17. Corriere MA, Passman MA, Guzman RJ, Dattilo JB, Naslund TC. Comparison of bedside transabdominal duplex ultrasound versus contrast venography for inferior vena cava filter placement: what is the best imaging modality? Ann Vasc Surg 2005;19:229-34. 18. Jacobs DL, Motaganahalli RL, Peterson BG. Bedside vena cava filter placement with intravascular ultrasound: a simple, accurate, single venous access method. J Vasc Surg 2007;46:1284-6. 19. Inferior vena cava (IVC) filters: initial communication: risk of adverse events with long term use. Available at: http://wwwfdagov/Safety/ MedWatch/SafetyInformation/SafetyAlertsforHumanMedicalProducts/ ucm221707htm. Accessed September 1, 2013. Submitted Feb 6, 2014; accepted Apr 27, 2014.
DISCUSSION Dr Lazar J. Greenfield, Sr (Tucson, Ariz). I wish to compliment Dr Glocker on his presentation and commend the authors on their excellent results. During the past 40 years, filter evolution has followed the unusual path of little change in basic design but
remarkable improvements in placement techniques, moving from operating room to radiology suite to bedside. Of course, when you see 100% success in a series, you wonder about selection, so my first question is, How many filter patients were excluded and
6 Glocker et al
JOURNAL OF VASCULAR SURGERY: VENOUS AND LYMPHATIC DISORDERS --- 2014
for what reasons? Did you exclude the patients with venous anomalies seen on computed tomography (CT) scan? Although you mention retrieval only of misplaced filters, your recent publication on 121 retrieval attempts during the same period had a 76% success rate, leaving 24% as longterm implants. In our current financial environment, how costeffective is this two-procedure approach as well as your selection process for venous thromboembolism risk when none actually developed that complication? Finally, having told those patients with venous thromboembolism of the need for a filter, what level of risk remains for those who no longer have that protection? Dr Roan J. Glocker. Thank you, Dr Greenfield. With regard to your first question, I think it relates to patient selection. Unfortunately, we do not know the overall number of patients we evaluated for bedside filter placement, but during the course of our study period, we did note that appropriate patient selection is perhaps the most important part of procedural success. The patients in whom we placed these filters are, as we saw, a significantly traumatically injured population, many of them with blunt trauma. Those patients almost always have an abdominal CT scan on admission to the hospital, and a thorough review of that abdominal CT scan is critical for bedside filter placement. We
believe that we have a reasonably high degree of procedural success before we begin a bedside filter placement on the basis of our view of that CT scan, ensuring that there are no venous anomalies and making sure the cava is of adequate diameter. Your second question relates to the venous anomalies. Those patients who did have venous anomalies were not considered for bedside filter placement by intravascular ultrasound, and those patients, if their anomaly was such that it would allow filter placement, underwent filter placement in the angiography suite. The question relating to filter retrieval is a focus of current work for us. We thought that looking at the rate of retrievals in this study was a little bit outside the scope. Our previous work looking at retrieval rates kind of compares apples and oranges because that patient cohort includes all filters we placed including those placed in the operating room. With regard to the need for filter placement and the overall trend toward placement of a decreasing number of filters, when we get these consults, we make a decision about whether to place the filter in conjunction with our critical care colleagues. I think the trend toward placing a decreased number of filters represents a perhaps better overall understanding of which patients benefit from a prophylactic filter placement and a trend toward decreasing the number of those prophylactic filters placed.