- Email: [email protected]
A systematic review of symptomatic duodenal perforation by inferior vena cava filters
A systematic review of symptomatic duodenal perforation by inferior vena cava filters Rafael D. Malgor, MD, and Nicos Labropoulos, PhD, DIC, RVT, Stony Brook, NY Objective: A systematic review of the literature on symptomatic duodenal perforation caused by inferior vena cava (IVC) filters. Methods: Three databases, PubMed MEDLINE, Web of Sciences, and Literatura Latino-Americana e do Caribe em Ciências da Saúde (LILACS), reference lists of review articles and conference proceedings were searched. All articles containing data on clinical presentation, diagnostic strategy, and available treatment of symptomatic duodenal perforation caused by an IVC filter were included regardless of design, language, size, or length of follow-up. Results: Seventy-two articles were selected for full-text screening, being 21 case reports were selected. The median age was 46 years old (range, 21-83 years old). Abdominal pain was reported in 11 patients and gastrointestinal bleed in 5 patients. The indications for IVC filter placement in this cohort of patients were contraindication of anticoagulation and recurrent pulmonary embolism (PE) despite therapeutic levels in 8 and 5 patients, respectively. Three different imaging modalities were obtained in 9 patients (43%) before confirming the diagnosis. All but 1 patient underwent open approach through laparotomy with or without removal of the filter. No PEs or deaths were reported and only 1 patient had a severe clinical complication of IVC and bilateral iliac vein thrombosis with massive lower extremities edema. Conclusions: Duodenal perforation caused by IVC filters is a rare complication that frequently requires extensive workup. Excellent outcomes with low complication rate have been reported in cases where an open procedure was performed with either extraction of the filter or removal of the offending struts. ( J Vasc Surg 2012;55:856-61.)
Inferior vena cava (IVC) interruption is indicated in patients who are not candidates for anticoagulation due to potential risk of life-threatening bleeding or for those who develop pulmonary embolism (PE) regardless of therapeutic levels of anticoagulation.1 However, IVC filters are not exempt from complications that encompass migration or fracture of filter struts, IVC thrombosis, and perforation of the venous wall causing bleeding or penetration into surrounding structures such as the aorta, portal, and renal veins, vertebral body, kidney and liver parenchyma, duodenum, large intestine, diaphragm, urinary tract, and the retroperitoneum.2-7 The incidence of perforation of the IVC wall is found in about 0.2% of the patients who underwent Greenfield filter placement, but the actual occurrence of duodenal perforation has not been reported.8 Symptomatic duodenum perforation by an IVC filter is rare and data on diagnosis and treatment have been inconsistently published. We performed a systematic review of the literature to assess the clinical presentation, diagnostic strategy, and available options of treatment for duodenal perforation caused by IVC filters. From the Division of Vascular Surgery, Stony Brook Medical Center. Competition of interest: none. Additional material for this article may be found online at www.jvascsurg.org. Reprint requests: Rafael Demarchi Malgor, MD, Division of Vascular Surgery, Department of Surgery, HSC T19 Room 90, Stony Brook University Medical Center, Stony Brook, NY 11794-8191 (e-mail: rafael. [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a competition of interest. 0741-5214/$36.00 Copyright © 2012 by the Society for Vascular Surgery. doi:10.1016/j.jvs.2011.09.082
856
METHODS Search strategy. A comprehensive search of databases, including PubMed MEDLINE, Web of Sciences, and Literatura Latino-Americana e do Caribe em Ciências da Saúde (LILACS) was performed. The latter was added to the searched databases in order to seek publications in a scientific repository that has no overlap with indexed literature in English. The Manufacturer and User Facility Device Experience (MAUDE) database provided by the Food and Drug Administration was also searched.9 This database gathers the reports of adverse events involving medical devices provided by manufacturers, distributors, and user facilities, but contains limited amounts of information about the event. The search strategy was designed and conducted by the authors. We also reviewed reference lists of review articles and conference proceedings. In addition, two conceptual areas, inferior vena cava filter and duodenum, were used to perform an extensive hand search. Details of the search are included in Appendix 1 (online only). Eligibility criteria. We used guidelines for systematic review published by expert consensus such as the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement.10 Studies were eligible for this review only if they reported data on outcomes of interest, including clinical presentation, diagnostic, and treatment of symptomatic duodenal perforation caused by an IVC filter. Anticipating that a limited number of publications would meet the criteria above, articles were included regardless of design (ie, case report, case-control, and cohort), language, size, or length of follow up. The articles excluded and the search strategy is depicted in Appendix 2 (online only).
JOURNAL OF VASCULAR SURGERY Volume 55, Number 3
Malgor and Labropoulos 857
Table I. Clinical presentation and filter characteristics Author (year) reference number
Age, gender
Irvin (1972)11
48, M
Appleberg et al (1990)12 Taheri et al (1992)13
71, F 41, F
Tritsch et al (1993)14
66, F
Goldman et al (1994)5
58, F
Al Zaharani et al (1995)15 Bianchini et al (1996)16
55, M
17
29, M
Sarkar et al (1997) Dardik et al (1997)18 Guillem et al (2001)19
68, F 36, M 60, F
Feezor et al (2002)20
40, M
Formentini et al (2005)21 Mansour et al (2005)22
23, F 41, M
Durairaj et al (2006)23
83, F
Botsios et al (2006)24
77, F
Ibele et al (2008)25
48, F
Veroux et al (2008)6
46, F
Parkin et al (2009)26 Franz et al (2009)27 Obman et al (2010)28 Becher et al (2010)29
21, M 27, M 40, F 42, M
Clinical presentation Fever ⫹ RUQ pain ⫹ Rt CVA tenderness Diarrhea ⫹ weight loss CP ⫹ SOB ⫹ RUQ abdominal pain Fever ⫹ weight loss ⫹ epigastric pain RUQ abdominal ⫹ flank pain Melena ⫹ hematemesis Heartburn ⫹ hematemesis Anemia ⫹ GI bleed Nausea (SB obstruction) Abdominal ⫹ lumbar pain Epigastric/RUQ pain ⫹ weight loss Epigastric pain Hematochezia ⫹ abdominal pain Epigastric discomfort Epigastric pain ⫹ GI bleed RUQ pain Diffuse edema left lower extremity Lower back pain Abdominal ⫹ back pain Upper abdominal pain Back pain
Interval from placement to presentation
Indication for IVC filter
Type of IVC filter
7 days
Recurrent PE despite AC
Mobin-Uddin
6 years 7 months
Massive iliofemoral DVT DVT ⫹ PE ⫹ GI bleed
4 years
DVT ⫹ PE Intracranial bleed ⫹ DVT
Greenfield Greenfield (supra-renal) KimrayGreenfield Mobin-Uddin
Recurrent DVT ⫹ PE despite AC
Bird’s nest
16 months
DVT ⫹ PE despite full AC
Greenfield
11 years 2 years 10 years
DVT ⫹ PE postoperatively DVT ⫹ GI bleed Thrombophlebitis ⫹ recurrent PE DVT ⫹ stroke
Mobin-Uddin Greenfield N/R
5 years 4 years
Postpartum DVT ⫹ PE PE ⫹ GI bleed
N/R Bird’s nest
6 years
Recurrent DVT ⫹ stroke ⫹ epistaxis/hematuria DVT ⫹ massive PE
Greenfield
Severe trauma ⫹ retroperitoneum bleeda Recurrent DVT ⫹ PE despite full AC DVT ⫹ multiple PE despite AC DVT ⫹ irregular use of AC Severe traumaa Traumatic SAH ⫹ multiorgan injurya
Recovery
10 years 5 years
N/R
9 years 14 months 2 years 5 years 10 months 15 years 10 months
Bird’s nest
Greenfield
Recovery Günther-tulip Celect Greenfield Celect
AC, Anticoagulation; CP, chest pain; CVA, costovertebral angle; DVT, deep vein thrombosis; F, female; GI, gastrointestinal; IVC, inferior vena cava; M, male; N/R, nonreported; PE, pulmonary embolism; Rt, right; RUQ, right upper quadrant; SAH, subarachnoid hemorrhage; SB, small bowel; SOB, short of breath. a Prophylactic indication with no DVT documented.
Statistical analysis. Descriptive statistics were used to analyze the data. No comparisons were made as the sample size was small and no severe clinical outcome was reported after intervention. Continuous variables were reported as mean, median, and SD, and categorical variables as percentages. RESULTS Twenty-one case reports met the criteria to be included in the study. The median age of patients who sustained a duodenal perforation by an IVC filter being treated either by endovascular (n ⫽ 1) with retrieval of the filter or open surgery (n ⫽ 20) was 46 years old (range, 21-83 years old). Eleven patients presented with abdominal pain located in the upper abdomen, mainly in the epigastrium or right upper quadrant. Nonoperative management of persistent pain led to partial improvement, but no resolution that postponed the surgical treatment in 2 patients. Gastrointestinal (GI) bleed was found in the initial assessment in 5
patients (23%) with subsequent hypovolemic shock in 1 patient. The indications for IVC filter placement in this cohort of patients were contraindication of anticoagulation and recurrent PE, despite therapeutic levels in 8 and 5 patients, respectively. Filters were prophylactically placed in 3 patients (14%) who sustained severe multiorgan trauma but no deep venous thrombosis (DVT). The most frequent type of filter causing a duodenal perforation was the Greenfield filter (Boston Scientific Corp, Natick, Mass) in 8 patients, followed by the Bird’s nest filter (Cook, Bloomington, Ind), and the Mobin-Uddin (no longer sold) in 3 patients each. However, other filters such as the Recovery filter (Bard Peripheral Vascular, Tempe, Ariz) in 2 patients, the Celect filter (Cook) in 2 patients, and the GuntherTulip filter (Cook) in 1 patient were also reported as a culprit of a duodenal perforation. A summary table containing data on clinical presentation and filter characteristics is provided in Table I.5-6,11-29
JOURNAL OF VASCULAR SURGERY March 2012
858 Malgor and Labropoulos
Table II. Workup and treatment of patients with symptomatic duodenal perforation secondary to IVC filters Author (year) reference number
Age, gender
Reported imaging workup
Irvin (1972)11 Appleberg et al (1990)12 Taheri et al (1992)13
48, M 71, F 41, F
Plain AXR Plain AXR ⫹ EGD ⫹ cavogram ⫹ CT of the abdomen/pelvis Plain AXR ⫹ cavogram
Tritsch et al (1993)14
66, F
Plain AXR ⫹ EGD ⫹ CT of the abdomen/pelvis
Goldman et al (1994)5 al Zaharani et al (1995)15
58, F 55, M
EGD ⫹ BE ⫹ colonoscopy ⫹ abdominal US ⫹ cavogram ⫹ CT EGD ⫹ CT of the abdomen/pelvis
Bianchini et al (1996)16
29, M
Plain AXR ⫹ EGD
Sarkar et al (1997) Dardik et al (1997)18 Guillem et al (2001)19 Feezor et al (2002)20
68, F 36, M 60, F 40, M
EGD Plain AXR ⫹ UGI series ⫹ EGD ⫹ CT Abdominal US ⫹ EGD ⫹ CT of the abdomen/pelvis Plain AXR ⫹ abdominal US ⫹ CT ⫹ EGD ⫹ cavogram
Formentini et al (2005)21 Mansour et al (2005)22
23, F 41, M
Plain AXR ⫹ EGD ⫹ CT of the abdomen/pelvis UGI series ⫹ EGD ⫹ CT of the abdomen/pelvis
Durairaj et al (2006)23 Botsios et al (2006)24 Ibele et al (2008)25
83, F 77, F 48, F
Abdominal US ⫹ CT of the abdomen/pelvis ⫹ ERCP EGD ⫹ CT of the abdomen/pelvis CT of the abdomen/pelvis
Veroux et al (2008)6
46, F
Parkin et al (2009)26 Franz et al (2009)27 Obman et al (2010)28 Becher et al (2010)29
21, M 27, M 40, F 42, M
Duplex US of the lower extremity ⫹ CT of the chest/abdomen/pelvis CT of the abdomen/pelvis CT of the abdomen/pelvis EGD ⫹ CT of the abdomen/pelvis CT of the abdomen/pelvis
17
AXR, Abdominal X-ray; BE, barium enema; CT, computed tomography; EGD, esophagogastroduodenoscopy; ERCP, endoscopic retrograde cholangiopancreatography; F, female; GSW, gunshot wound; IVC, inferior vena cava; M, male; N/R, nonreported; PDA, pseudoaneurysm; UGI, upper gastrointestinal series; US, ultrasound; UTI, urinary tract infection.
Imaging workup was performed in all patients with abdominal pain or GI bleed who had previous IVC filter placement. Three different imaging modalities were completed before any intervention in 9 patients (43%). In all but 4 patients, a computed tomography (CT) was used to confirm the diagnosis and evaluate the anatomic relationship between the filter and the surrounding structures in order to plan the procedure. An esophagogastroduodenoscopy (EGD) was performed in 13 patients being able to visualize struts of the filter in the second or third portion of the duodenum. An open procedure was carried out performing a laparotomy with or without removal of the filter. At the time of the treatment, the majority of the patients had the filter placed more than 2 years ago (range, 7 days-180 months) apart from the initial clinical presentation. The only endovascular retrieval reported was performed in a 48-year-old female patient who had a Recovery filter prophylactically placed due to multiorgan trauma and retroperitoneum bleeding 14 months before the onset of right upper quadrant pain. However, 2 other patients underwent an unsuccessful attempt to have the filter retrieved through an endovascular approach before surgery but with no complications. All of those five retrievable filters found were in place for more than 10 months (range, 10-60 months) before the onset of symptoms.
Patients who underwent abdominal exploration often had intense inflammatory tissue around the filter precluding removal in 4 patients (19%). The IVC was explored and a venotomy performed to remove the filter in 10 patients (48%) and all but 1 patient (primarily closed) had a bovine or prosthetic patch used for angioplasty. In those patients who did not have the filter extracted, one or more struts were trimmed flush with the IVC using wire cutters. Table II provides a detailed workup list and treatment.5-6,11-29 Complications after the abdominal approach were reported in only 1 patient who had IVC and bilateral iliac vein thrombosis with massive lower extremities edema after trimming of the prongs but no venotomy. There were no pulmonary embolisms or deaths reported. DISCUSSION IVC filters have been used as a safe and efficacious device to prevent PE since the early 1970s.30 The Greenfield filter was popularized with a premise of minimal risk of IVC occlusion and a device that was inserted totally by endovascular approach with minimal complications compared to extensive open techniques such as ligation and venous clips used to narrow the vessel lumen with great risk of IVC thrombosis and its consequences (ie, postthrombotic syndrome). However, filters are also prone to complications related to malposition and misplacement.
JOURNAL OF VASCULAR SURGERY Volume 55, Number 3
Malgor and Labropoulos 859
Table II. Continued Migration/tilt
Remarkable findings
N/R Tilted ⫹ migrated Tilted ⫹ migrated
N/R N/R Involvement of left renal
Tilted
Cholestasis, UTI
Tilted Tilted
Cholelytiasis Hypovolemic shock ⫹ IVC thrombosis
Not tilted or migrated
Quadriplegia post-GSW
N/R Migrated ⫹ tilted Migrated Tilted
Supra-hepatic IVC thrombosis Crohn’s disease ⫹ involvement of pancreas Choledochal dilatation Stroke ⫹ bipolar disorder
Tilted Migrated
N/R Small duodenal ulcer
N/R Tilted N/R
Dilation of biliary duct Small bowel obstruction Infected retroperitoneum hematoma
N/R
Aortic mural thrombus
N/R N/R N/R Tilted
Aortic ⫹ L2 vertebra affected N/R Erosion of vertebral body Fluid collection, aortic PDA
The literature on the topic is scant because of the rarity of duodenal perforation caused by IVC filters. The actual incidence of duodenal perforation is unknown. However, the MAUDE database reports 3739 adverse events associated with IVC filters from January 1992 to December 2010. According to the MAUDE registry, isolated perforation of the cava wall or into its surrounding structures is responsible for 10% to 20% of all events reported. The drawbacks of this database are the wide range of sources of information varying from patients to manufacturers and limited information on outcomes. However, it perhaps outlines and estimates the potential risks and roughly predicts the incidence of the events found in the so-called “real-life” environment. Management of symptomatic patients can be cumbersome, requiring exposure of the IVC bounded by inflammatory reaction. To our knowledge, the literature has not been systematically searched to scrutinize the current diagnostic and treatment strategies available. Our inclusion criteria were met only by case reports due to higher level of detail on workup and management of a rare complication. Generally, a high clinical suspicion is advocated by the authors and should be raised for patients with a history of IVC filter placement, especially those complaining of atypical or persistent abdominal pain, or who had an episode of a GI bleed. One may question the indication of interven-
Treatment Ligation of the IVC just below renal veins Repair of the duodenum ⫹ extraction of the filter (cavotomy) Minimal duodenal perforation ⫹ extraction of the filter (cavotomy) Strut cut flush with the IVC ⫹ repair of the duodenum (no cavotomy) Extraction of filter found outside the IVC Strut cut flush with the IVC ⫹ repair of the duodenum (no cavotomy) Repair of the duodenum ⫹ partial extraction of the filter (cavotomy) Repair of the duodenum ⫹ extraction of the filter (cavotomy) Repair of the duodenum ⫹ extraction of the filter (cavotomy) Repair of the duodenum ⫹ extraction of the filter (cavotomy) Strut cut flush with the IVC ⫹ repair of the duodenum (no cavotomy) Repair of the duodenum ⫹ extraction of the filter (cavotomy) Strut cut flush with the IVC ⫹ repair of the duodenum (no cavotomy) Nonoperative management, long-term antibiotic Repair of the duodenum ⫹ extraction of the filter (cavotomy) Retrieved endovascularly (CT showing involvement of duodenum) Repair of the duodenum ⫹ extraction of the filter ⫹ aortic thrombectomy Extraction of the filter (no macro injury of duodenum) Repair of the duodenum ⫹ extraction of the filter (cavotomy) Repair of the duodenum ⫹ extraction of the filter (cavotomy) Repair of Ao PDA, extraction of the filter
tion (eg, trimming of filter struts or cavotomy) in patients with a GI bleed that perhaps were overly anticoagulated and concomitantly had a duodenal perforation caused by an IVC filter. However, data on levels of anticoagulation are not clear in those 5 patients who initially present with a GI bleed. Regardless, we found that at least two different imaging studies were obtained in order to rule out other common causes and establish a relationship between filters and symptoms before final diagnosis and treatment. The design of the filter is also an important risk factor for tilting and perforation, especially in those retrievable ones that require point contact with the vena cava. Another interesting current concept on IVC filters is surveillance and follow-up due to the increasing number of retrievable filters. Symptomatic patients with duodenal perforation had a filter placed 2 years ago or longer apart from clinical presentation, perhaps conferring low efficacy of midterm or long-term follow-up for this specific complication. In addition, all retrievable filters found in this review were placed more than 10 months before the onset of symptoms. Therefore, retrievable filters coupled with a standardized approach by specialized teams to ensure removal may play a role in the future to decrease the number of complications, including duodenal perforation. The treatment of symptomatic patients with duodenal perforation was done through endovascular approach only
JOURNAL OF VASCULAR SURGERY March 2012
860 Malgor and Labropoulos
in 1 patient. The rest of them were treated through laparotomy with or without venotomy. Very good results after extensive abdominal operations and manipulation of the IVC or other organs were obtained with only one major complication of IVC thrombosis. One should be aware that this may be likely a result of a selection bias that occurs when good results are available in the literature rather than due to a low-risk procedure. However, the importance of surgical treatment leading to symptoms relief and absolute failure of nonoperative management must be emphasized. Strikingly, the number of cases reported in the literature has increased over the past 4 decades. There are a few possible hypotheses to explain this temporal distribution. First, universal availability of routine workup using CT and EGD for abdominal complaints such as abdominal pain and GI bleed in patients with previous IVC filter may have augmented the likelihood of detection of filter misplacement, migration, or tilting. Second, a rise on availability of filters mainly over the past 2 decades and the so-called prophylactic IVC filters may have exposed more patients to treatment and, therefore, to duodenum perforation. Finally, a long course may be required from the time of placement to protrusion of struts of a tilted filter through the IVC wall. The latter occurrence is seen in patients that had filters placed up to 15 years before abdominal symptoms regardless if they were properly positioned in the past. Continuous improvement on spatial design aiming to more durable and stable filters are required in order to prevent filter tilting and its subsequent complications, including perforation of intra-abdominal structures such as the duodenum.
2.
3.
4.
5. 6.
7. 8. 9.
10.
11. 12. 13.
14.
CONCLUSIONS Duodenal perforation caused by IVC filters is a rare complication. Diagnosis may be challenging, requiring multiple imagining modalities due to unspecific clinical presentation. Symptomatic patients sustaining duodenal perforation are nearly all candidates for open repair due to scarring and inflammatory reaction that frequently preclude endovascular retrieval. The results reported in the literature are excellent with low complication rates for patients who underwent open procedure with either extraction of the filter or removal of the offending struts.
15.
16.
17.
18.
19.
AUTHOR CONTRIBUTIONS Conception and design: RM, NL Analysis and interpretation: RM, NL Data collection: RM Writing the article: RM, NL Critical revision of the article: RM, NL Final approval of the article: RM, NL Statistical analysis: RM, NL Obtained funding: Not applicable Overall responsibility: RM, NL REFERENCES 1. Kearon C, Kahn SR, Agnelli G, Goldhaber S, Raskob GE, Comerota AJ; American College of Chest Physicians. Antithrombotic therapy for
20.
21.
22.
23. 24.
venous thromboembolic disease: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th edition). Chest 2008;133(6 Suppl):454S-545S. White S, Lehrfeld T, Schwab C. Misplaced inferior vena caval filter in right renal vein with erosion into renal collecting system. J Endourol 2009;23:1899-901. Greenfield LJ. Perforation of the inferior vena cava with aortic and vertebral penetration by a suprarenal Greenfield filter. Radiology 1990; 175:287. Drevet D, Himed A, Marx P, Revel D, Pelissier F, Amiel M, et al. [Perforation of a loop of small-intestine by a Kim-Ray Greenfield endocaval filter. Report of a case. [Article in French] Ann Radiol (Paris) 1990;33:347-50. Goldman KA, Adelman MA. Retroperitoneal caval filter as a source of abdominal pain. Cardiovasc Surg 1994;2:85-7. Veroux M, Tallarita T, Pennisi M, Veroux P. Late complication from a retrievable inferior vena cava filter with associated caval, aortic, and duodenal perforation: a case report. J Vasc Surg 2008;48:223-5. Miller CL, Wechsler RJ. CT evaluation of Kimray-Greenfield filter complications. AJR Am J Roentgenol 1986;147:45-50. Greenfield LJ, Proctor MC. Twenty-year clinical experience with the Greenfield filter. Cardiovasc Surg 1995;3:199-205. MAUDE. (Manufacturer and User Facility Device Experience) [Internet]. Available from: http://www.accessdata.fda.gov/scripts/cdrh/ cfdocs/cfmaude/search.cfm. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med 2009;6;e1000100. Irvin GL 3rd. Duodenal perforation with a vena caval umbrella. Am Surg 1972;38:635-7. Appleberg M, Crozier JA. Duodenal penetration by a Greenfield caval filter. Aust N Z J Surg 1991;61:957-60. Taheri SA, Kulaylat MN, Johnson E, Hoover E. A complication of the Greenfield filter: fracture and distal migration of two struts—a case report. J Vasc Surg 1992;16:96-9. Tritsch L, Coumaros D, Sauder P, Kopferschmitt J, Jaeger A, Mantz JM. [Duodenal perforations by the hooks of a Kimray-Greenfield filter]. [Article in French] Ann Fr Anesth Reanim 1993;12:75-8. al Zahrani HA. Bird’s nest inferior vena caval filter migration into the duodenum: a rare cause of upper gastrointestinal bleeding. J Endovasc Surg 1995;2:372-5. Bianchini AU, Mehta SN, Mulder DS, Barkun AN, Mayrand S. Duodenal perforation by a Greenfield filter: endoscopic diagnosis. Am J Gastroenterol 1997;92:686-7. Sarkar MR, Lemminger FM. An unusual cause of upper gastrointestinal haemorrhage—perforation of a vena cava filter into the duodenum. Vasa 1997;26:305-7. Dardik A, Campbell KA, Yeo CJ, Lipsett PA. Vena cava filter ensnarement and delayed migration: an unusual series of cases. J Vasc Surg 1997;26:869-74. Guillem PG, Binot D, Dupuy-Cuny J, Laberenne JE, Lesage J, Triboulet JP, et al. Duodenocaval fistula: a life-threatening condition of various origins. J Vasc Surg 2001;33:643-5. Feezor RJ, Huber TS, Welborn MB 3rd, Schell SR. Duodenal perforation with an inferior vena cava filter: an unusual cause of abdominal pain. J Vasc Surg 2002;35:1010-2. Formentini A, Stanescu A, Staib L, Aschoff AJ, Scharrer-Pamler R, Henne-Bruns D. [An unusual foreign body in the upper gastrointestinal tract causing nonspecific abdominal pain]. [Article in German] Chirurg 2005;76:501-4. Mansour JC, Lee FT Jr, Chen H, Turnipseed WD, Weber SM. Chronic abdominal pain and upper gastrointestinal bleeding due to duodenal perforation caused by migrated inferior vena cava filter—a case report. Vasc Endovascular Surg 2004;38:381-4. DuraiRaj R, Fogarty S. A penetrating inferior vena caval filter. Eur J Vasc Endovasc Surg 2006;32:737-9. Botsios S, Erhart R, Walterbusch G. [Acute gastrointestinal bleeding caused by perforation of a Greenfield caval filter into the duode-
JOURNAL OF VASCULAR SURGERY Volume 55, Number 3
25.
26.
27.
28.
num]. [Article in German] Dtsch Med Wochenschr 2006; 131:2715-7. Ibele A, Hermsen J, Kudsk K. Resolution of chronic abdominal pain after percutaneous removal of recoverable inferior vena cava filters: a report of two cases. J Trauma 2008;64:215-6; discussion 216-7. Parkin E, Serracino-Inglott F, Chalmers N, Smyth V. Symptomatic perforation of a retrievable inferior vena cava filter after a dwell time of 5 years. J Vasc Surg 2009;50:417-9. Franz RW, Johnson JD, Shah KJ. Symptomatic inferior vena cava perforation by a retrievable filter: report of two cases and a literature review. Int J Angiol 2009;18:203-6. Obmann MA, Gray JL, Sheldon DG, Franklin DP. Duodenal perforation and vertebral body erosion by a Greenfield filter. J Vasc Surg 2010;51:1528.
Malgor and Labropoulos 861
29. Becher RD, Corriere MA, Edwards MS, Godshall CJ. Late erosion of a prophylactic Celect IVC filter into the aorta, right renal artery, and duodenal wall. J Vasc Surg 2010;52:1041-4. 30. Greenfield LJ, McCurdy JR, Brown PP, Elkins RC. A new intracaval filter permitting continued flow and resolution of emboli. Surgery 1973;73:599-606.
Submitted July 26, 2011; accepted September 24, 2011.
Additional material for this article may be found online at www.jvascsurg.org.
JOURNAL OF VASCULAR SURGERY March 2012
861.e1 Malgor and Labropoulos
PUBMED MEDLINE #
Searches
Results
1
"vena cava filters"[MeSH Terms]
1819
2
duoden*[MeSH Terms]
39290
3
perforaon
43098
4
penetrat*
86515
5
intesn*[MeSH Terms]
283486
6
abdom*[MeSH Terms]
72985
7
pain[MeSH Terms]
261049
8
complicaon*[MeSH Subheading]
1490259
9
bleeding[MeSH Terms]
227329
10
#2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9
2223553
11
#10 AND #11
709
*Word/Character expanding funcon (i.e. duoden*=duodenum or duodenal) Appendix 1 (online only).
JOURNAL OF VASCULAR SURGERY Volume 55, Number 3
Malgor and Labropoulos 861.e2
Web of Sciences (Databases=SCI-EXPANDED, SSCI, A&HCI Timespan=All Years) #
Searches
Results
1
TS=(vena cava)
17,065
2
TS=(filter*)
>100,000
3
#1 AND #2
1,632
4
TS=(perforaon)
25,244
5
TS=(duoden*)
57,362
6
TS=(penetrat*)
>100,000
7
TS=(intesn*)
>100,000
8
TS=(abdom*)
>100,000
9
TS=(pain)
>100,000
10
TS=(complicaon*)
>100,000
11
TS=(bleeding)
82,125
12
#4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10 OR #11
>100,000
12
#12 AND #3
772
*Word/Character expanding function (i.e. duoden*=duodenum or duodenal); TS, topic
LILACS #
Searches
Results
1
Filtros de veia cava (inferior vena cava filters) [MeSH Terms]
39
Appendix 1 (online only). Continued
JOURNAL OF VASCULAR SURGERY March 2012
861.e3 Malgor and Labropoulos
1,520 citations identified in databases including additional records from references of references
1,332 citations identified in databases and screened by abstract and title after duplicates removed
72 full-text articles screened
Studies excluded n = 51 40 IVC perforaons but no duodenal perforaon or related-symptoms 5 Small-bowel but no duodenal involvement 3 Incidental finding of duodenal involvement (no treatment required) 1 Duodenal perforaon but no treatment data available
21 studies selected by outcomes for systematic review Appendix 2 (online only). Diagram of systematic review for selection or exclusion of studies.
Inferior vena cava (IVC) interruption is indicated in patients who are not candidates for anticoagulation due to potential risk of life-threatening bleeding or for those who develop pulmonary embolism (PE) regardless of therapeutic levels of anticoagulation.1 However, IVC filters are not exempt from complications that encompass migration or fracture of filter struts, IVC thrombosis, and perforation of the venous wall causing bleeding or penetration into surrounding structures such as the aorta, portal, and renal veins, vertebral body, kidney and liver parenchyma, duodenum, large intestine, diaphragm, urinary tract, and the retroperitoneum.2-7 The incidence of perforation of the IVC wall is found in about 0.2% of the patients who underwent Greenfield filter placement, but the actual occurrence of duodenal perforation has not been reported.8 Symptomatic duodenum perforation by an IVC filter is rare and data on diagnosis and treatment have been inconsistently published. We performed a systematic review of the literature to assess the clinical presentation, diagnostic strategy, and available options of treatment for duodenal perforation caused by IVC filters. From the Division of Vascular Surgery, Stony Brook Medical Center. Competition of interest: none. Additional material for this article may be found online at www.jvascsurg.org. Reprint requests: Rafael Demarchi Malgor, MD, Division of Vascular Surgery, Department of Surgery, HSC T19 Room 90, Stony Brook University Medical Center, Stony Brook, NY 11794-8191 (e-mail: rafael. [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a competition of interest. 0741-5214/$36.00 Copyright © 2012 by the Society for Vascular Surgery. doi:10.1016/j.jvs.2011.09.082
856
METHODS Search strategy. A comprehensive search of databases, including PubMed MEDLINE, Web of Sciences, and Literatura Latino-Americana e do Caribe em Ciências da Saúde (LILACS) was performed. The latter was added to the searched databases in order to seek publications in a scientific repository that has no overlap with indexed literature in English. The Manufacturer and User Facility Device Experience (MAUDE) database provided by the Food and Drug Administration was also searched.9 This database gathers the reports of adverse events involving medical devices provided by manufacturers, distributors, and user facilities, but contains limited amounts of information about the event. The search strategy was designed and conducted by the authors. We also reviewed reference lists of review articles and conference proceedings. In addition, two conceptual areas, inferior vena cava filter and duodenum, were used to perform an extensive hand search. Details of the search are included in Appendix 1 (online only). Eligibility criteria. We used guidelines for systematic review published by expert consensus such as the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement.10 Studies were eligible for this review only if they reported data on outcomes of interest, including clinical presentation, diagnostic, and treatment of symptomatic duodenal perforation caused by an IVC filter. Anticipating that a limited number of publications would meet the criteria above, articles were included regardless of design (ie, case report, case-control, and cohort), language, size, or length of follow up. The articles excluded and the search strategy is depicted in Appendix 2 (online only).
JOURNAL OF VASCULAR SURGERY Volume 55, Number 3
Malgor and Labropoulos 857
Table I. Clinical presentation and filter characteristics Author (year) reference number
Age, gender
Irvin (1972)11
48, M
Appleberg et al (1990)12 Taheri et al (1992)13
71, F 41, F
Tritsch et al (1993)14
66, F
Goldman et al (1994)5
58, F
Al Zaharani et al (1995)15 Bianchini et al (1996)16
55, M
17
29, M
Sarkar et al (1997) Dardik et al (1997)18 Guillem et al (2001)19
68, F 36, M 60, F
Feezor et al (2002)20
40, M
Formentini et al (2005)21 Mansour et al (2005)22
23, F 41, M
Durairaj et al (2006)23
83, F
Botsios et al (2006)24
77, F
Ibele et al (2008)25
48, F
Veroux et al (2008)6
46, F
Parkin et al (2009)26 Franz et al (2009)27 Obman et al (2010)28 Becher et al (2010)29
21, M 27, M 40, F 42, M
Clinical presentation Fever ⫹ RUQ pain ⫹ Rt CVA tenderness Diarrhea ⫹ weight loss CP ⫹ SOB ⫹ RUQ abdominal pain Fever ⫹ weight loss ⫹ epigastric pain RUQ abdominal ⫹ flank pain Melena ⫹ hematemesis Heartburn ⫹ hematemesis Anemia ⫹ GI bleed Nausea (SB obstruction) Abdominal ⫹ lumbar pain Epigastric/RUQ pain ⫹ weight loss Epigastric pain Hematochezia ⫹ abdominal pain Epigastric discomfort Epigastric pain ⫹ GI bleed RUQ pain Diffuse edema left lower extremity Lower back pain Abdominal ⫹ back pain Upper abdominal pain Back pain
Interval from placement to presentation
Indication for IVC filter
Type of IVC filter
7 days
Recurrent PE despite AC
Mobin-Uddin
6 years 7 months
Massive iliofemoral DVT DVT ⫹ PE ⫹ GI bleed
4 years
DVT ⫹ PE Intracranial bleed ⫹ DVT
Greenfield Greenfield (supra-renal) KimrayGreenfield Mobin-Uddin
Recurrent DVT ⫹ PE despite AC
Bird’s nest
16 months
DVT ⫹ PE despite full AC
Greenfield
11 years 2 years 10 years
DVT ⫹ PE postoperatively DVT ⫹ GI bleed Thrombophlebitis ⫹ recurrent PE DVT ⫹ stroke
Mobin-Uddin Greenfield N/R
5 years 4 years
Postpartum DVT ⫹ PE PE ⫹ GI bleed
N/R Bird’s nest
6 years
Recurrent DVT ⫹ stroke ⫹ epistaxis/hematuria DVT ⫹ massive PE
Greenfield
Severe trauma ⫹ retroperitoneum bleeda Recurrent DVT ⫹ PE despite full AC DVT ⫹ multiple PE despite AC DVT ⫹ irregular use of AC Severe traumaa Traumatic SAH ⫹ multiorgan injurya
Recovery
10 years 5 years
N/R
9 years 14 months 2 years 5 years 10 months 15 years 10 months
Bird’s nest
Greenfield
Recovery Günther-tulip Celect Greenfield Celect
AC, Anticoagulation; CP, chest pain; CVA, costovertebral angle; DVT, deep vein thrombosis; F, female; GI, gastrointestinal; IVC, inferior vena cava; M, male; N/R, nonreported; PE, pulmonary embolism; Rt, right; RUQ, right upper quadrant; SAH, subarachnoid hemorrhage; SB, small bowel; SOB, short of breath. a Prophylactic indication with no DVT documented.
Statistical analysis. Descriptive statistics were used to analyze the data. No comparisons were made as the sample size was small and no severe clinical outcome was reported after intervention. Continuous variables were reported as mean, median, and SD, and categorical variables as percentages. RESULTS Twenty-one case reports met the criteria to be included in the study. The median age of patients who sustained a duodenal perforation by an IVC filter being treated either by endovascular (n ⫽ 1) with retrieval of the filter or open surgery (n ⫽ 20) was 46 years old (range, 21-83 years old). Eleven patients presented with abdominal pain located in the upper abdomen, mainly in the epigastrium or right upper quadrant. Nonoperative management of persistent pain led to partial improvement, but no resolution that postponed the surgical treatment in 2 patients. Gastrointestinal (GI) bleed was found in the initial assessment in 5
patients (23%) with subsequent hypovolemic shock in 1 patient. The indications for IVC filter placement in this cohort of patients were contraindication of anticoagulation and recurrent PE, despite therapeutic levels in 8 and 5 patients, respectively. Filters were prophylactically placed in 3 patients (14%) who sustained severe multiorgan trauma but no deep venous thrombosis (DVT). The most frequent type of filter causing a duodenal perforation was the Greenfield filter (Boston Scientific Corp, Natick, Mass) in 8 patients, followed by the Bird’s nest filter (Cook, Bloomington, Ind), and the Mobin-Uddin (no longer sold) in 3 patients each. However, other filters such as the Recovery filter (Bard Peripheral Vascular, Tempe, Ariz) in 2 patients, the Celect filter (Cook) in 2 patients, and the GuntherTulip filter (Cook) in 1 patient were also reported as a culprit of a duodenal perforation. A summary table containing data on clinical presentation and filter characteristics is provided in Table I.5-6,11-29
JOURNAL OF VASCULAR SURGERY March 2012
858 Malgor and Labropoulos
Table II. Workup and treatment of patients with symptomatic duodenal perforation secondary to IVC filters Author (year) reference number
Age, gender
Reported imaging workup
Irvin (1972)11 Appleberg et al (1990)12 Taheri et al (1992)13
48, M 71, F 41, F
Plain AXR Plain AXR ⫹ EGD ⫹ cavogram ⫹ CT of the abdomen/pelvis Plain AXR ⫹ cavogram
Tritsch et al (1993)14
66, F
Plain AXR ⫹ EGD ⫹ CT of the abdomen/pelvis
Goldman et al (1994)5 al Zaharani et al (1995)15
58, F 55, M
EGD ⫹ BE ⫹ colonoscopy ⫹ abdominal US ⫹ cavogram ⫹ CT EGD ⫹ CT of the abdomen/pelvis
Bianchini et al (1996)16
29, M
Plain AXR ⫹ EGD
Sarkar et al (1997) Dardik et al (1997)18 Guillem et al (2001)19 Feezor et al (2002)20
68, F 36, M 60, F 40, M
EGD Plain AXR ⫹ UGI series ⫹ EGD ⫹ CT Abdominal US ⫹ EGD ⫹ CT of the abdomen/pelvis Plain AXR ⫹ abdominal US ⫹ CT ⫹ EGD ⫹ cavogram
Formentini et al (2005)21 Mansour et al (2005)22
23, F 41, M
Plain AXR ⫹ EGD ⫹ CT of the abdomen/pelvis UGI series ⫹ EGD ⫹ CT of the abdomen/pelvis
Durairaj et al (2006)23 Botsios et al (2006)24 Ibele et al (2008)25
83, F 77, F 48, F
Abdominal US ⫹ CT of the abdomen/pelvis ⫹ ERCP EGD ⫹ CT of the abdomen/pelvis CT of the abdomen/pelvis
Veroux et al (2008)6
46, F
Parkin et al (2009)26 Franz et al (2009)27 Obman et al (2010)28 Becher et al (2010)29
21, M 27, M 40, F 42, M
Duplex US of the lower extremity ⫹ CT of the chest/abdomen/pelvis CT of the abdomen/pelvis CT of the abdomen/pelvis EGD ⫹ CT of the abdomen/pelvis CT of the abdomen/pelvis
17
AXR, Abdominal X-ray; BE, barium enema; CT, computed tomography; EGD, esophagogastroduodenoscopy; ERCP, endoscopic retrograde cholangiopancreatography; F, female; GSW, gunshot wound; IVC, inferior vena cava; M, male; N/R, nonreported; PDA, pseudoaneurysm; UGI, upper gastrointestinal series; US, ultrasound; UTI, urinary tract infection.
Imaging workup was performed in all patients with abdominal pain or GI bleed who had previous IVC filter placement. Three different imaging modalities were completed before any intervention in 9 patients (43%). In all but 4 patients, a computed tomography (CT) was used to confirm the diagnosis and evaluate the anatomic relationship between the filter and the surrounding structures in order to plan the procedure. An esophagogastroduodenoscopy (EGD) was performed in 13 patients being able to visualize struts of the filter in the second or third portion of the duodenum. An open procedure was carried out performing a laparotomy with or without removal of the filter. At the time of the treatment, the majority of the patients had the filter placed more than 2 years ago (range, 7 days-180 months) apart from the initial clinical presentation. The only endovascular retrieval reported was performed in a 48-year-old female patient who had a Recovery filter prophylactically placed due to multiorgan trauma and retroperitoneum bleeding 14 months before the onset of right upper quadrant pain. However, 2 other patients underwent an unsuccessful attempt to have the filter retrieved through an endovascular approach before surgery but with no complications. All of those five retrievable filters found were in place for more than 10 months (range, 10-60 months) before the onset of symptoms.
Patients who underwent abdominal exploration often had intense inflammatory tissue around the filter precluding removal in 4 patients (19%). The IVC was explored and a venotomy performed to remove the filter in 10 patients (48%) and all but 1 patient (primarily closed) had a bovine or prosthetic patch used for angioplasty. In those patients who did not have the filter extracted, one or more struts were trimmed flush with the IVC using wire cutters. Table II provides a detailed workup list and treatment.5-6,11-29 Complications after the abdominal approach were reported in only 1 patient who had IVC and bilateral iliac vein thrombosis with massive lower extremities edema after trimming of the prongs but no venotomy. There were no pulmonary embolisms or deaths reported. DISCUSSION IVC filters have been used as a safe and efficacious device to prevent PE since the early 1970s.30 The Greenfield filter was popularized with a premise of minimal risk of IVC occlusion and a device that was inserted totally by endovascular approach with minimal complications compared to extensive open techniques such as ligation and venous clips used to narrow the vessel lumen with great risk of IVC thrombosis and its consequences (ie, postthrombotic syndrome). However, filters are also prone to complications related to malposition and misplacement.
JOURNAL OF VASCULAR SURGERY Volume 55, Number 3
Malgor and Labropoulos 859
Table II. Continued Migration/tilt
Remarkable findings
N/R Tilted ⫹ migrated Tilted ⫹ migrated
N/R N/R Involvement of left renal
Tilted
Cholestasis, UTI
Tilted Tilted
Cholelytiasis Hypovolemic shock ⫹ IVC thrombosis
Not tilted or migrated
Quadriplegia post-GSW
N/R Migrated ⫹ tilted Migrated Tilted
Supra-hepatic IVC thrombosis Crohn’s disease ⫹ involvement of pancreas Choledochal dilatation Stroke ⫹ bipolar disorder
Tilted Migrated
N/R Small duodenal ulcer
N/R Tilted N/R
Dilation of biliary duct Small bowel obstruction Infected retroperitoneum hematoma
N/R
Aortic mural thrombus
N/R N/R N/R Tilted
Aortic ⫹ L2 vertebra affected N/R Erosion of vertebral body Fluid collection, aortic PDA
The literature on the topic is scant because of the rarity of duodenal perforation caused by IVC filters. The actual incidence of duodenal perforation is unknown. However, the MAUDE database reports 3739 adverse events associated with IVC filters from January 1992 to December 2010. According to the MAUDE registry, isolated perforation of the cava wall or into its surrounding structures is responsible for 10% to 20% of all events reported. The drawbacks of this database are the wide range of sources of information varying from patients to manufacturers and limited information on outcomes. However, it perhaps outlines and estimates the potential risks and roughly predicts the incidence of the events found in the so-called “real-life” environment. Management of symptomatic patients can be cumbersome, requiring exposure of the IVC bounded by inflammatory reaction. To our knowledge, the literature has not been systematically searched to scrutinize the current diagnostic and treatment strategies available. Our inclusion criteria were met only by case reports due to higher level of detail on workup and management of a rare complication. Generally, a high clinical suspicion is advocated by the authors and should be raised for patients with a history of IVC filter placement, especially those complaining of atypical or persistent abdominal pain, or who had an episode of a GI bleed. One may question the indication of interven-
Treatment Ligation of the IVC just below renal veins Repair of the duodenum ⫹ extraction of the filter (cavotomy) Minimal duodenal perforation ⫹ extraction of the filter (cavotomy) Strut cut flush with the IVC ⫹ repair of the duodenum (no cavotomy) Extraction of filter found outside the IVC Strut cut flush with the IVC ⫹ repair of the duodenum (no cavotomy) Repair of the duodenum ⫹ partial extraction of the filter (cavotomy) Repair of the duodenum ⫹ extraction of the filter (cavotomy) Repair of the duodenum ⫹ extraction of the filter (cavotomy) Repair of the duodenum ⫹ extraction of the filter (cavotomy) Strut cut flush with the IVC ⫹ repair of the duodenum (no cavotomy) Repair of the duodenum ⫹ extraction of the filter (cavotomy) Strut cut flush with the IVC ⫹ repair of the duodenum (no cavotomy) Nonoperative management, long-term antibiotic Repair of the duodenum ⫹ extraction of the filter (cavotomy) Retrieved endovascularly (CT showing involvement of duodenum) Repair of the duodenum ⫹ extraction of the filter ⫹ aortic thrombectomy Extraction of the filter (no macro injury of duodenum) Repair of the duodenum ⫹ extraction of the filter (cavotomy) Repair of the duodenum ⫹ extraction of the filter (cavotomy) Repair of Ao PDA, extraction of the filter
tion (eg, trimming of filter struts or cavotomy) in patients with a GI bleed that perhaps were overly anticoagulated and concomitantly had a duodenal perforation caused by an IVC filter. However, data on levels of anticoagulation are not clear in those 5 patients who initially present with a GI bleed. Regardless, we found that at least two different imaging studies were obtained in order to rule out other common causes and establish a relationship between filters and symptoms before final diagnosis and treatment. The design of the filter is also an important risk factor for tilting and perforation, especially in those retrievable ones that require point contact with the vena cava. Another interesting current concept on IVC filters is surveillance and follow-up due to the increasing number of retrievable filters. Symptomatic patients with duodenal perforation had a filter placed 2 years ago or longer apart from clinical presentation, perhaps conferring low efficacy of midterm or long-term follow-up for this specific complication. In addition, all retrievable filters found in this review were placed more than 10 months before the onset of symptoms. Therefore, retrievable filters coupled with a standardized approach by specialized teams to ensure removal may play a role in the future to decrease the number of complications, including duodenal perforation. The treatment of symptomatic patients with duodenal perforation was done through endovascular approach only
JOURNAL OF VASCULAR SURGERY March 2012
860 Malgor and Labropoulos
in 1 patient. The rest of them were treated through laparotomy with or without venotomy. Very good results after extensive abdominal operations and manipulation of the IVC or other organs were obtained with only one major complication of IVC thrombosis. One should be aware that this may be likely a result of a selection bias that occurs when good results are available in the literature rather than due to a low-risk procedure. However, the importance of surgical treatment leading to symptoms relief and absolute failure of nonoperative management must be emphasized. Strikingly, the number of cases reported in the literature has increased over the past 4 decades. There are a few possible hypotheses to explain this temporal distribution. First, universal availability of routine workup using CT and EGD for abdominal complaints such as abdominal pain and GI bleed in patients with previous IVC filter may have augmented the likelihood of detection of filter misplacement, migration, or tilting. Second, a rise on availability of filters mainly over the past 2 decades and the so-called prophylactic IVC filters may have exposed more patients to treatment and, therefore, to duodenum perforation. Finally, a long course may be required from the time of placement to protrusion of struts of a tilted filter through the IVC wall. The latter occurrence is seen in patients that had filters placed up to 15 years before abdominal symptoms regardless if they were properly positioned in the past. Continuous improvement on spatial design aiming to more durable and stable filters are required in order to prevent filter tilting and its subsequent complications, including perforation of intra-abdominal structures such as the duodenum.
2.
3.
4.
5. 6.
7. 8. 9.
10.
11. 12. 13.
14.
CONCLUSIONS Duodenal perforation caused by IVC filters is a rare complication. Diagnosis may be challenging, requiring multiple imagining modalities due to unspecific clinical presentation. Symptomatic patients sustaining duodenal perforation are nearly all candidates for open repair due to scarring and inflammatory reaction that frequently preclude endovascular retrieval. The results reported in the literature are excellent with low complication rates for patients who underwent open procedure with either extraction of the filter or removal of the offending struts.
15.
16.
17.
18.
19.
AUTHOR CONTRIBUTIONS Conception and design: RM, NL Analysis and interpretation: RM, NL Data collection: RM Writing the article: RM, NL Critical revision of the article: RM, NL Final approval of the article: RM, NL Statistical analysis: RM, NL Obtained funding: Not applicable Overall responsibility: RM, NL REFERENCES 1. Kearon C, Kahn SR, Agnelli G, Goldhaber S, Raskob GE, Comerota AJ; American College of Chest Physicians. Antithrombotic therapy for
20.
21.
22.
23. 24.
venous thromboembolic disease: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th edition). Chest 2008;133(6 Suppl):454S-545S. White S, Lehrfeld T, Schwab C. Misplaced inferior vena caval filter in right renal vein with erosion into renal collecting system. J Endourol 2009;23:1899-901. Greenfield LJ. Perforation of the inferior vena cava with aortic and vertebral penetration by a suprarenal Greenfield filter. Radiology 1990; 175:287. Drevet D, Himed A, Marx P, Revel D, Pelissier F, Amiel M, et al. [Perforation of a loop of small-intestine by a Kim-Ray Greenfield endocaval filter. Report of a case. [Article in French] Ann Radiol (Paris) 1990;33:347-50. Goldman KA, Adelman MA. Retroperitoneal caval filter as a source of abdominal pain. Cardiovasc Surg 1994;2:85-7. Veroux M, Tallarita T, Pennisi M, Veroux P. Late complication from a retrievable inferior vena cava filter with associated caval, aortic, and duodenal perforation: a case report. J Vasc Surg 2008;48:223-5. Miller CL, Wechsler RJ. CT evaluation of Kimray-Greenfield filter complications. AJR Am J Roentgenol 1986;147:45-50. Greenfield LJ, Proctor MC. Twenty-year clinical experience with the Greenfield filter. Cardiovasc Surg 1995;3:199-205. MAUDE. (Manufacturer and User Facility Device Experience) [Internet]. Available from: http://www.accessdata.fda.gov/scripts/cdrh/ cfdocs/cfmaude/search.cfm. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med 2009;6;e1000100. Irvin GL 3rd. Duodenal perforation with a vena caval umbrella. Am Surg 1972;38:635-7. Appleberg M, Crozier JA. Duodenal penetration by a Greenfield caval filter. Aust N Z J Surg 1991;61:957-60. Taheri SA, Kulaylat MN, Johnson E, Hoover E. A complication of the Greenfield filter: fracture and distal migration of two struts—a case report. J Vasc Surg 1992;16:96-9. Tritsch L, Coumaros D, Sauder P, Kopferschmitt J, Jaeger A, Mantz JM. [Duodenal perforations by the hooks of a Kimray-Greenfield filter]. [Article in French] Ann Fr Anesth Reanim 1993;12:75-8. al Zahrani HA. Bird’s nest inferior vena caval filter migration into the duodenum: a rare cause of upper gastrointestinal bleeding. J Endovasc Surg 1995;2:372-5. Bianchini AU, Mehta SN, Mulder DS, Barkun AN, Mayrand S. Duodenal perforation by a Greenfield filter: endoscopic diagnosis. Am J Gastroenterol 1997;92:686-7. Sarkar MR, Lemminger FM. An unusual cause of upper gastrointestinal haemorrhage—perforation of a vena cava filter into the duodenum. Vasa 1997;26:305-7. Dardik A, Campbell KA, Yeo CJ, Lipsett PA. Vena cava filter ensnarement and delayed migration: an unusual series of cases. J Vasc Surg 1997;26:869-74. Guillem PG, Binot D, Dupuy-Cuny J, Laberenne JE, Lesage J, Triboulet JP, et al. Duodenocaval fistula: a life-threatening condition of various origins. J Vasc Surg 2001;33:643-5. Feezor RJ, Huber TS, Welborn MB 3rd, Schell SR. Duodenal perforation with an inferior vena cava filter: an unusual cause of abdominal pain. J Vasc Surg 2002;35:1010-2. Formentini A, Stanescu A, Staib L, Aschoff AJ, Scharrer-Pamler R, Henne-Bruns D. [An unusual foreign body in the upper gastrointestinal tract causing nonspecific abdominal pain]. [Article in German] Chirurg 2005;76:501-4. Mansour JC, Lee FT Jr, Chen H, Turnipseed WD, Weber SM. Chronic abdominal pain and upper gastrointestinal bleeding due to duodenal perforation caused by migrated inferior vena cava filter—a case report. Vasc Endovascular Surg 2004;38:381-4. DuraiRaj R, Fogarty S. A penetrating inferior vena caval filter. Eur J Vasc Endovasc Surg 2006;32:737-9. Botsios S, Erhart R, Walterbusch G. [Acute gastrointestinal bleeding caused by perforation of a Greenfield caval filter into the duode-
JOURNAL OF VASCULAR SURGERY Volume 55, Number 3
25.
26.
27.
28.
num]. [Article in German] Dtsch Med Wochenschr 2006; 131:2715-7. Ibele A, Hermsen J, Kudsk K. Resolution of chronic abdominal pain after percutaneous removal of recoverable inferior vena cava filters: a report of two cases. J Trauma 2008;64:215-6; discussion 216-7. Parkin E, Serracino-Inglott F, Chalmers N, Smyth V. Symptomatic perforation of a retrievable inferior vena cava filter after a dwell time of 5 years. J Vasc Surg 2009;50:417-9. Franz RW, Johnson JD, Shah KJ. Symptomatic inferior vena cava perforation by a retrievable filter: report of two cases and a literature review. Int J Angiol 2009;18:203-6. Obmann MA, Gray JL, Sheldon DG, Franklin DP. Duodenal perforation and vertebral body erosion by a Greenfield filter. J Vasc Surg 2010;51:1528.
Malgor and Labropoulos 861
29. Becher RD, Corriere MA, Edwards MS, Godshall CJ. Late erosion of a prophylactic Celect IVC filter into the aorta, right renal artery, and duodenal wall. J Vasc Surg 2010;52:1041-4. 30. Greenfield LJ, McCurdy JR, Brown PP, Elkins RC. A new intracaval filter permitting continued flow and resolution of emboli. Surgery 1973;73:599-606.
Submitted July 26, 2011; accepted September 24, 2011.
Additional material for this article may be found online at www.jvascsurg.org.
JOURNAL OF VASCULAR SURGERY March 2012
861.e1 Malgor and Labropoulos
PUBMED MEDLINE #
Searches
Results
1
"vena cava filters"[MeSH Terms]
1819
2
duoden*[MeSH Terms]
39290
3
perforaon
43098
4
penetrat*
86515
5
intesn*[MeSH Terms]
283486
6
abdom*[MeSH Terms]
72985
7
pain[MeSH Terms]
261049
8
complicaon*[MeSH Subheading]
1490259
9
bleeding[MeSH Terms]
227329
10
#2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9
2223553
11
#10 AND #11
709
*Word/Character expanding funcon (i.e. duoden*=duodenum or duodenal) Appendix 1 (online only).
JOURNAL OF VASCULAR SURGERY Volume 55, Number 3
Malgor and Labropoulos 861.e2
Web of Sciences (Databases=SCI-EXPANDED, SSCI, A&HCI Timespan=All Years) #
Searches
Results
1
TS=(vena cava)
17,065
2
TS=(filter*)
>100,000
3
#1 AND #2
1,632
4
TS=(perforaon)
25,244
5
TS=(duoden*)
57,362
6
TS=(penetrat*)
>100,000
7
TS=(intesn*)
>100,000
8
TS=(abdom*)
>100,000
9
TS=(pain)
>100,000
10
TS=(complicaon*)
>100,000
11
TS=(bleeding)
82,125
12
#4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10 OR #11
>100,000
12
#12 AND #3
772
*Word/Character expanding function (i.e. duoden*=duodenum or duodenal); TS, topic
LILACS #
Searches
Results
1
Filtros de veia cava (inferior vena cava filters) [MeSH Terms]
39
Appendix 1 (online only). Continued
JOURNAL OF VASCULAR SURGERY March 2012
861.e3 Malgor and Labropoulos
1,520 citations identified in databases including additional records from references of references
1,332 citations identified in databases and screened by abstract and title after duplicates removed
72 full-text articles screened
Studies excluded n = 51 40 IVC perforaons but no duodenal perforaon or related-symptoms 5 Small-bowel but no duodenal involvement 3 Incidental finding of duodenal involvement (no treatment required) 1 Duodenal perforaon but no treatment data available
21 studies selected by outcomes for systematic review Appendix 2 (online only). Diagram of systematic review for selection or exclusion of studies.