Intracardiac Migration of Retrievable Vena Cava Filter Theresa Gelzinis, MD,* Kathirvel Subramaniam, MD,* William E. Katz, MD,† and Lawrence Wei, MD‡
NDICATIONS FOR THE insertion of inferior vena cava (IVC) filters are increasing. The Gunther-Tulip filter (GTF; Cook, Bloomington, IN) is one of the optional retrievable IVC filters available for clinical use and is placed under fluoroscopic guidance. Although intracardiac migration has been reported with all filters, there are only 2 reports concerning the GTF.1 The presentation, diagnosis, and management of 2 other patients with this complication are described. CASE REPORTS
Case 1 A 72-year-old man with polycythemia vera presented with acute deep vein thrombosis in his left leg. This occurred despite bimonthly phlebotomy and therapy with coumadin. Intravenous heparin therapy was started but discontinued when a paraspinal hematoma developed. GTF was placed under fluoroscopic guidance. Five days later, the patient developed sudden respiratory arrest before discharge. He was intubated and transferred to the intensive care unit (ICU) after resuscitation. In the ICU, the patient remained hemodynamically unstable requiring epinephrine and vasopressin infusions as well as intermittent chest compressions. He had also been having periods of atrioventricular block. On a chest x-ray, the IVC filter was seen above the diaphragm. An urgent transthoracic echocardiogram showed the IVC filter obstructing the tricuspid valve. The patient was emergently brought to the operating room for removal of the filter. An intraoperative transesophageal echocardiogram confirmed the location of the IVC filter and showed severe right ventricular (RV) dilatation and dysfunction (Fig 1 and Video 1 [supplementary videos accompanying this article are available online]). The patient was heparinized and emergently placed on cardiopulmonary bypass (CPB) using aortic and bicaval cannulae. A right atriotomy was performed, the IVC filter extracted, and the tricuspid valve inspected. Because the tricuspid valve was intact, the atrium was closed, and the patient was weaned from CPB. A postbypass TEE examination showed normal function of the tricuspid valve with no significant regurgitation, but RV dysfunction, which was treated with epinephrine, milrinone, and vasopressin infusions. Chest closure resulted in deterioration of RV function and hemodynamics. The chest was kept open, and the patient was transferred to the ICU in
Fig 1. Transgastric basal short-axis TEE image with scanning angle 77° showing the tricuspid valve entangled with an IVC filter, RV dilatation, and flattened interventricular septum.
critical condition. RV function improved, the inotropic support was weaned, and the chest was closed in the operating room on the second postoperative day. The postoperative course was complicated by respiratory failure (requiring tracheostomy), atrial fibrillation (thought to be secondary to trauma from the IVC filter), mild gastrointestinal bleeding, dysphagia (requiring tube feeds), parkinsonian symptoms (treated with Sinemet), transient delusions, cerebral salt wasting (salt therapy), and orthostatic hypotension (requiring compression stockings and abdominal binders). He was treated in the hospital for 3.5 months before being discharged home on coumadin therapy.
Case 2 A 52-year-old man was referred to the authors’ university medical center after IVC filter placement for left lower extremity deep vein thrombosis at another facility. The IVC filter apparently dislodged and presented as a “foreign body in the heart.” The patient had a history of spinal osteomyelitis, quadriparesis, hypertension, diabetes mellitus, intravenous drug abuse, hepatitis C, morbid obesity with cardiomyopathy, and tricuspid valvulectomy for infective endocarditis. The patient was in a stable hemodynamic condition and evaluated for possible filter removal by interventional radiologists. Fluoroscopic images showed that the filter was located superior to the pulmonary valve with the removal hook facing anteriorly, and the filter was oriented in an anteroposterior direction (Fig 2). The radiologist believed that attempts to remove the filter could be difficult and pose additional risk to the patient. The patient was referred to cardiothoracic surgery for surgical removal of the filter device. The patient had left-heart catheterization, which showed no significant coronary artery disease with preserved left ventricular function. He was clinically stable and scheduled to have surgery in the afternoon. While waiting for surgery, he suddenly became unresponsive and bradycardic (heart rate 20 beats/min). He was resuscitated and urgently transferred to the operating room for intervention. An intraoperative TEE revealed vena caval, right atrial, and right ventricular dilatation; absent tricuspid valve; and the IVC filter in the right ventricular outflow and main pulmonary artery (Figs 3 and 4 and Video 2). The patient underwent redo sternotomy, CPB with aortic and bicaval cannulations, mild hypothermia, antegrade cardioplegia, removal of IVC filter, and bioprosthetic 31-mm tricuspid valve replacement. A postoperative TEE confirmed the removal of the filter with no pulmonary insufficiency and a well-seated prosthetic tricuspid valve with no significant gradient or paravalvular leak. The postoperative period was complicated by coagulopathy, acute respiratory distress syndrome, renal failure and wound infection. The patient required tracheostomy, dialysis, and broad-spectrum antibiotics. He became stable enough for transfer to a facility after 2 months of hospitalization.
From the Departments of *Anesthesiology and ‡Cardiac Surgery, and †Cardiovascular Institute, University of Pittsburgh Medical Center, Presbyterian University Hospital, Pittsburgh, PA. Address reprint requests to Kathirvel Subramaniam, MD, Department of Anesthesiology, University of Pittsburgh Medical Center, Presbyterian University Hospital, 200 Lothrop Street, C-Wing, Pittsburgh, PA 15213. E-mail: [email protected]
© 2009 Elsevier Inc. All rights reserved. 1053-0770/09/2303-0020$36.00/0 doi:10.1053/j.jvca.2008.04.010 Key words: vena cava filter, migration, intracardiac
Journal of Cardiothoracic and Vascular Anesthesia, Vol 23, No 3 (June), 2009: pp 381-383
GELZINIS ET AL
Fig 4. A TEE midesophageal long-axis view showing the IVC filter in the right ventricle.
Fig 2. Fluoroscopic image showing the filter in the chest oriented in the anteroposterior direction.
Absolute indications for the use of IVC filters include contraindications to anticoagulation, failure of anticoagulation, and complications of anticoagulation including hemorrhage or thrombocytopenia.2 IVC filters can be permanent, temporary, or optional-retrievable filters. The GTF is an optional-retrievable filter, which can be removed if the indication for insertion no longer exists. It consists of 4 struts with a hook on the end of each strut to attach to the IVC and a hook on top to facilitate removal (Fig 5). Advantages of this filter are its magnetic reso-
nance imaging compatibility, ease of insertion and retrievability, low access site thrombosis, and high filtering efficiency.3 Complications of IVC filter placement include IVC perforation, worsening of proximal thrombosis, filter migration, and infection.4 Stein et al5 described an incidence of 2% to 3% for filter migration, but migration into the heart was rare. The majority of filter migrations were asymptomatic, but reports of myocardial infarction because of posterior descending artery dissection,6 massive pulmonary embolism, cardiac failure because of tricuspid valve obstruction or regurgitation, and vena caval or myocardial perforation with cardiac tamponade7 have been described. Patients with these complications can present with hemodynamic instability after IVC filter placement. Cardiac arrhythmias have been described with intracardiac filter migration.8 Both patients in the present report developed bradyarrhythmias preoperatively and atrial fibrillation postopera-
Fig 3. A TEE midesophageal RV inflowoutflow view showing the absent tricuspid valve. (Color version of figure is available online.)
RETRIEVABLE VENA CAVA FILTER
Fig 5. A Gunther-Tulip IVC filter extracted from the heart. The filter has 4 primary struts, 4 secondary struts, and a retrieving hook at one end. (Color version of figure is available online.)
tively; in patient 1, this could be related to trauma from the filter. A chest x-ray can be a very useful screening test in suspected patients. The definitive diagnosis of intracardiac filter migration can be made in stable patients using fluoroscopy or computed tomography scans. Echocardiography at the bedside is the diagnostic method of choice in hemodynamically unstable patients. A transthoracic echocardiogram and, if possible, TEE may be recommended because these patients will need immediate surgery with CPB. Echocardiography can also rule out other cardiac causes of sudden hemodynamic instability such as myocardial ischemia with wall motion changes, large pulmonary embolism, or acute valvular regurgitation. Echocardiography is particularly useful in determining the exact location of the filter and consequences of filter migration including tricuspid or pulmonary valve damage (valve obstruction and regurgitation), RV dysfunction, or pericardial tamponade. Patient 1 was hemodynamically unstable, and the supradiaphragmatic location of the filter was suggested by a chest x-ray,
which was further confirmed with transthoracic echocardiography. In patient 2, a chest x-ray and fluoroscopy provided the diagnosis. An intraoperative TEE revealed a dilated IVC in this patient. Ideally, IVC diameter measurement should be done before filter placement. GTF filters can be inserted in patients with IVC diameters only up to 30 mm.9 TEE confirmed an absent tricuspid valve that resulted in distal migration of the filter into the RV outflow and pulmonary artery. James et al10 reviewed the literature of intracardiac migration of Greenfield permanent filters. The management ranged from observation in asymptomatic patients to open surgical removal in other patients. Nonsurgical management (the patient would not give consent for surgery) led to death in 1 of the patients with tricuspid regurgitation.10 In another series of 3 patients with intracardiac migration, long-term follow-up to 60 months did not result in any complication.11 Successful surgical removal should be weighed against the higher perioperative morbidity as described in both of these high-risk patients. Percutaneous removal by interventional radiologists provides an alternate method of treatment in asymptomatic patients and in high-risk patients who may not be surgical candidates. Bochenek et al1 described right atrial migration and successful percutaneous retrieval of GTF with a special snare in a trauma patient. A backup cardiac operating room and CPB machine should be available for these cases. Hemodynamically unstable patients with valve involvement (obstruction and regurgitation), patients with myocardial involvement (perforation and cardiac tamponade), and patients with cardiac arrhythmias will require prompt cardiac surgical consultation. The removal of intracardiac IVC filters alone can be done using CPB on a beating heart with bicaval cannulations unless the patient needs other procedures such as valve repair or replacement. The patent foramen ovale should be looked for carefully in TEE examination because its presence will be a contraindication for open beating-heart surgery. In the post-CPB period, TEE examination should focus on completeness of removal of the filter, residual valve damage, and RV dysfunction.
REFERENCES 1. Bochenek KM, Aruny JE, Tal MG: Right atrial migration and percutaneous retrieval of a Gunther Tulip inferior vena cava filter. J Vasc Interv Radiol 14:1207-1209, 2003 2. Kinney TB: Update on inferior vena cava filters. J Vasc Interv Radiol 14:425-440, 2003 3. Yamagami T, Kato T, Iida S, et al: Gunther Tulip inferior vena cava filter placement during treatment for deep venous thrombosis of the lower extremity. Cardiovasc Intervent Radiol 28:442-453, 2005 4. Chiou AC, Biggs KL, Matsumura JS: Vena cava filters: Why, when, what, how? Perspect Vasc Surg Endovasc Ther 17:329-339, 2005 5. Stein PD, Kayali F, Olsen RE: Twenty-one year trends in the use of inferior vena cava filters. Arch Intern Med 164:1541-1545, 2004
6. Puram B, Maley TJ, White NM, et al: Acute myocardial infarction resulting from the migration of a Greenfield filter. Chest 98:1510-1511, 1990 7. Urena R, Greenwood L: Bird’s nest filter migration to the right atrium. AJR Am J Roentgenol 183:1037-1039, 2004 8. Bach JR, Zaneuski R, Lee H: Cardiac arrhythmias from a malpositioned Greenfield filter in a traumatic quadriplegic. Am J Phys Med Rehabil 69:251-253, 1990 9. Matthews BD, Joels CS, LeQuire MH: Inferior vena cava filter placement: Preinsertion inferior vena cava imaging. Am Surg 69:649653, 2003 10. James KV, Sobolewski AP, Lohr JM, et al: Tricuspid insufficiency after intracardiac migration of a Greenfield filter: Case report and review of the literature J Vasc Surg 24:494-498, 1996 11. Gelbfish GA, Ascer E: Intracardiac and intrapulmonary Greenfield filters: A long-term follow-up. J Vasc Surg 14:614-617, 1991