Case Report and Brief Review

Myopericarditis Secondary to Embolization of Fractured Inferior Vena Cava Filter Limbs

N. Nathani, MD, M. Barzallo, MD, Sudhir Mungee, MD

N. Nathani, MD, M. Barzallo, MD, Sudhir Mungee, MD

ABSTRACT: Vena cava filters are placed for the prevention of pulmonary embolism in the setting of venous thromboembolism, especially when there is a contraindication to anticoagulation or complication secondary to anticoagulation. Here we discuss a rare complication of vena cava filters. We present the first documented case of inferior vena cava filter limb fracture presenting as myopericarditis. Six other cases of embolization of vena cava filter limbs are reviewed and a mechanism for fracture and embolization is proposed.
J INVASIVE CARDIOL 2011;23:E225–E228
Key words: myopericarditis, filter fracture, filter embolization, filter complication
————————————————————
Case Report. A 47-year-old male presented to the emergency department with sudden onset, sharp chest pain while exercising at home. Interestingly, this patient presented 2 weeks earlier with pleuritic pain, CK-MB of 1.5 U/L, Troponin-I of 0.64 ng/ml, and total creatinine phosphokinase (CPK) of 133 U/L, and electrocardiogram (ECG) suggestive of pericarditis. In spite of appropriate treatment with non-steroidal anti-inflammatory drugs (NSAIDs) and corticosteroids, he continued to be symptomatic. His past medical history was significant for deep vein thromboses (DVT) in the setting of a motor vehicle accident 3 years ago, at which time a Bard (C.R. Bard, Lowell, Massachusetts) inferior vena cava (IVC) filter was placed. His hemodynamic status and physical examination were unremarkable and there was absence of pericardial rub. He had significant ST-segment elevation in the lateral and inferior leads (Figure 1) with elevated cardiac markers, CK-MB and Troponin-I of 36.5 U/L and 13.93 ng/ml, respectively. Based on these findings, ST elevation myocardial infarction (STEMI) diagnosis was suspected. An echocardiogram showed normal wall-motion and ejection fraction, but presence of a moderate-sized circumferential pericardial effusion. The patient was taken to the cardiac catheterization lab for coronary angiography, which revealed angiographically normal coronaries. However, an incidental finding of 2 linear opacities was visualized in the region of cardiac silhouette (Figure 2). Subsequently, computed tomography (CT) scan (see report) of the chest without contrast revealed a moderate-sized hemorrhagic pericardial effusion (Figures 3 and 4). Within the anterior wall of the right ventricle, two metallic foreign bodies were identified (Figures 3 and 4). Position of the Bard filter was noted to be appropriate in the IVC; however, there was absence of 2 limbs. Based on normal angiographic findings, transient ECG changes, elevated cardiac markers and the above CT findings, a diagnosis of myopericarditis was entertained. The patient was taken to interventional radiology for percutaneous removal of the filter, which confirmed 2 missing limbs. Cardiothoracic surgery was consulted for limb retrieval and open heart surgery with cardiopulmonary bypass was performed to retrieve the foreign bodies. One limb of the IVC filter was immediately visualized perforating the free wall of the right ventricle medial to the left anterior descending artery (Figure 5). The second limb, however, could not be directly visualized despite fluoroscopic guidance and was fixated with 4-0 sutures. The patient was successfully removed from bypass and discharged home pain free. Discussion. This report highlights a serious complication related to IVC filters, discusses the mechanism involved, and offers a review of literature relevant to this complication. Delayed fracture of an IVC filter is a rare complication occurring in about 1% of the patient population.6 We describe a case of an IVC filter with late limb fractures embolizing to the right ventricle resulting in perforation, myopericarditis and hemodynamically insignificant pericardial effusion. A review of the literature shows few cases of IVC filter limb fracture and that even fewer cases embolize to the heart.13 To understand the risk factors associated with IVC filter fracture, migration, and the associated complications, we reviewed 6 case reports fitting our criteria. The search was performed using PubMed MeSH search with parameters “vena cava filter,” “migration,” “embolization,” and “fracture.” Within the 39 results, 6 papers matched our criteria (Table 1). The IVC filter types reviewed in these cases were 1 Greenfield (Boston Scientific Corp., Natick, Massachusetts), 4 Bard Recovery (C.R. Bard), and 1 Gunther-Tulip (Cook, Inc., Bloomington, Indiana). The design and characteristics of each filter are listed in Table 2. The Greenfield filters are permanent, while Gunther-Tulip and Bard Recovery filters are removable by design. All filters are conical shape, but differ in their composition. Elgioly and Nitinol are lighter alloy metals compared to stainless-steel. Elgiloy is composed of nickel, cobalt, and chromium predominantly, while Nitinol is composed of nickel and titanium. The advantage of alloys is that their shape is compressible at room temperature, but can re-expand at body temperatures. This permits easy loading into a catheter, yet a flush interface on expansion inside the vena cava. Due to a lack of data regarding IVC filter fracture, it is difficult to attribute this complication to specific characteristics of a filter, i.e., material or design. Consistent with this data, however, Hull et al reported a 21% incidence of filter arm fracture and embolization of Bard Recovery filters compared to the overall average of 1%.3 They attributed the mechanism to repetitive cyclical deformations of the filter caused by breathing.2 Previous studies suggest external forces can be a risk factor for filter fracture. Increased intra-abdominal pressure or strenuous physical activity may cause fracture and migration of an IVC filter.10 In our case, the patient was doing sit-ups when he first experienced symptoms. A filter limb could potentially embolize and/or perforate the IVC, superior vena cava, right atrium, right ventricle or pulmonary arteries. Hull et al reported cases of fractured filter limb embolizing to the right superior pulmonary artery, retroperitoneum, and IVC.3 The literature reviewed showed all clinically significant cases with filter limb fractures localize to the right ventricle, not to be confused with intact IVC filter migration and/or perforation. However, this is the first documented case of IVC filter fracture presenting as myopericarditis. Clinical presentation varies from cardiac tamponade to ventricular tachycardia to acute coronary syndrome. It is important to note that intact IVC filter migration can cause a similar presentation. Muhammad et al compiled 28 reports of migration of IVC filters. Of these cases, only 3 of 17 were composed of fractured filter wires. Excluding the fractured filters, the intact IVC filters presented as chest pain, ventricular tachycardia, and hypotension, though none specifically as pericardial tamponade.8 Ventricular perforation from an IVC filter limb can result in serious consequences. Our patient was in stable condition on presentation; however, decompensation and sudden death can result. Further studies are necessary to elucidate the cause of IVC filter fracture and migration. Conclusion. IVC filters are placed for the prevention of pulmonary embolism in the setting of venous thromboembolism (VTE), especially when there is a contraindication to anticoagulation or complication secondary to anticoagulation. The indications can be subdivided into absolute, relative and prophylactic categories (Table 3).6 A potentially disastrous complication of IVC filters prompts the question whether routine removal of all IVC filters is warranted once the indication is no longer present. In our case, serious consequence of delayed fracture of an IVC filter occurred less than 1% of the time.6 However, the follow-up to the PREPIC study illustrated more common long-term consequences of IVC filters. The study’s results show a higher rate of symptomatic recurrent DVT (35.7% versus 27.5%; p = 0.042), lower incidence of symptomatic pulmonary embolism (6.2% versus 15.1%; p = 0.008), but no effect on survival (48.1% versus 51.0%; p = 0.83) after 8 years with long-term IVC filter as opposed to anticoagulation alone.12 Long-term permanent IVC filter dwelling is not benign. Further studies performed on the efficacy and safety of retrievable IVC filters showed a rate of pulmonary embolism from 1.5–12.5% and recurrent deep vein thrombosis up to 7% with mean follow-up from 90–540 days.1 Furthermore, in support of routine removal, Yavuz et al reported a 2.8% incidence of VTE at an average of 20.58 months follow-up post-IVC filter retrieval. These 2 cases, comprising 2.8%, were from inadvertent suspension of anticoagulation and factor-V Leiden deficiency in the setting of a trans-Atlantic flight.14 Though rare, embolization of the entire filter versus a limb can result in fatal complications. Careful consideration of usage and timely removal may be one of the ways to avoid such complications. Patient’s knowledge of these complications and certain physical limitations may help in avoiding such circumstances. The physical nature of the alloy/metal involved in these filters to withstand the shearing stress forces may need to be further addressed to maintain stability. In patients with chronically placed filters, chest pain, palpitations, or hemodynamic compromise should alert the clinician to the cardiac complications such as pericarditis, arrhythmia, and shock secondary to filter or filter limb embolization.

References

1. Berczi V, Bottomley JR, Thomas SM, et al. Long-term retrievability of IVC filters: Should we abandon permanent devices? Cardiovasc Intervent Radiol 2007;30:620–627. 2. Chandra PA, Nwokolo C, Chuprun D, et al. Cardiac tamponade caused by fracture and migration of inferior vena cava filter. South Med J 2008;101:1163–1164. 3. Hull J, Robertson S. Bard recovery filter: Evaluation and management of vena cava limb perforation, fracture, and migration. J Vasc Intervent Radiol 2009;20:52–60. 4. Hull J, Han J, Gissel G. Retrieval of the recovery filter after arm perforation, fracture, and migration to the right ventricle. J Vasc Intervent Radiol 2008;19:1107–1111. 5. Kalavakunta J, Thomas C, Gupta V. A needle through the heart: Rare complication of inferior vena cava filters. J Invasive Cardiol 2009;21:221–223. 6. Kinney T. Update on inferior vena cava filters. J Vasc Intervent Radiol 2003;14:425–440. 7. Kumar S, Mahtabifard A, Young N. Fractured inferior vena cava filter strut presenting as a penetrating foreign body in the right ventricle: Report of a case. J Card Surg 2008;23:378–381. 8. Muhammad J, Omran F, Kastoon T, et al. Inferior vena cava filter migration: Updated review and case presentation. J Invasive Cardiol 2009;21:606–610. 9. Rogers NA, Nguyen L, Minniefield NE, et al. Fracture and embolization of an inferior vena cava filter strut leading to cardiac tamponade. Circulation 2009;119:2535–2536. 10. Rossi P, Arata FM, Bonaiuti P, Pedicini V. Fatal outcome in atrial migration of the Tempofilter. Cardiovasc Intervent Radiol 1999;22:227–231. 11. Saeed I, Garcia M, McNicholas K. Right ventricular migration of a recovery IVC filter’s fractured wire with subsequent pericardial tamponade. Cardiovasc Intervent Radiol 2006;29:685–686. 12. The PREPIC Study Group. Eight year follow-up of patients with permanent vena cava filters in the prevention of pulmonary embolism: The PREPIC (Prevention du Risque d’Embolic Pulmonaire par Interruption Cave) randomized study. Circulation 2005;112:416–422. 13. Vergara G, Wallace W, Bennett K. Spontaneous migration of an inferior vena cava filter resulting in cardiac tamponade and percutaneous filter retrieval. Catheter Cardiovasc Interv 2007;69:300–302. 14. Yavuz K, Geyik S, Hoppe H, et al. Venous thromboembolism after retrieval of inferior vena cava filters. J Vasc Intervent Radiol 2008;19:504–508.
————————————————————
From the University of Illinois College of Medicine Peoria, HeartCare Midwest, Peoria, Illinois. The authors report no conflicts of interest regarding the content herein. Manuscript submitted February 12, 2010, provisional acceptance given March 11, 2010, final version April 1, 2010. Author for correspondence: Sudhir Mungee, MD, FACC, FSCAI, Assistant Clinical Professor, University of Illinois College of Medicine Peoria, HeartCare Midwest, 5405 N. Knoxville Ave, Peoria, IL 61615. E-mail: docsudh@hotmail.com