INTERVENTIONAL PEDIATRIC CARDIOLOGY

Late Superior Vena Cava Perforation and Aortic Laceration After Stent Placement for Treatment of Superior Vena Cava Syndrome Sec

Michael R. Recto, MD, *Michael Bousamra, MD, *Thomas Yeh, Jr., MD, PhD
Michael R. Recto, MD, *Michael Bousamra, MD, *Thomas Yeh, Jr., MD, PhD
Stent implantation has long been used to treat many types of vascular stenoses. Stents were initially used in the late 1980’s to treat patients with branch pulmonary artery stenoses.1–3 In the 1990’s, stents were used to relieve right ventricular outflow tract obstruction, systemic/pulmonary venous obstruction, systemic venous baffle obstruction and coarctation of the aorta.4–8 Follow-up studies have demonstrated good results following stent implantation.9,10 Stents have also been used for the treatment of superior vena cava (SVC) stenosis and SVC syndrome with good results.11–13 There have been few complications associated with the use of stents in the SVC and other systemic venous vessels.13–15 We describe a case of late perforation of the SVC and laceration of the ascending aorta after stent implantation for SVC syndrome. The etiology of the perforation is unclear, but could be secondary to either flaring of the trailing edge of the stent during the stent implantation procedure, or to chest trauma resulting in the trailing edge of the stent eroding into the wall of the SVC and ascending aorta. Case Report. A 13-year-old female presented with a 3-month history of increasing facial swelling and periorbital edema. Physical examination was significant for edema of the face, neck and arms. The patient also had dilated superficial chest veins. Chest x-ray demonstrated the presence of a large calcified mediastinal lymph node. CT scan of the chest confirmed the presence of a large, 2 x 3 cm calcified lymph node in the right paratracheal region. The calcified lymph node was located posterior to the SVC and anterior to the trachea. Magnetic resonance imaging demonstrated a severely narrowed SVC measuring approximately 3 mm. Prominent venous collateral vessels were seen arising from the SVC and left innominate vein. Work-up to determine the etiology of the SVC syndrome was performed by both the hematology/oncology and infectious disease services. The presumed etiology of the SVC syndrome was fibrosing mediastinitis secondary to histoplasmosis. Over the course of one week, both the facial swelling and periorbital edema worsened, and the patient could barely open her eyes. The decision was then made to proceed with cardiac catheterization to better delineate the level and degree of SVC obstruction, and to possibly perform SVC stent implantation. Of note, the patient had a past history complicated by behavioral problems and was a ward of the state. After obtaining informed consent, cardiac catheterization was performed. After obtaining femoral venous and arterial access, the patient received a bolus of heparin. An activated clotting time of > 200 seconds was maintained throughout the case. Cardiac catheterization demonstrated the presence of severe SVC stenosis (Figure 1). The mean pressure in the SVC (cephalad) was 35 mmHg, and the mean pressure in the right atrium was 11 mmHg (mean gradient of 24 mmHg across the region of stenosis). There were many venous collateral vessels arising from both the SVC and innominate vein. Angiographically, the region of stenosis was adjacent to the insertion site of the left innominate vein into the SVC (Figure 1). After discussion with the cardiothoracic service, the consensus was that the patient would benefit from stent implantation. One of the initial concerns was that the stent would have to be positioned across the insertion of the innominate vein into the SVC, possibly impeding venous return from the innominate vein. However, there appeared to be sufficient venous collateral circulation to prevent significant obstruction to venous return from the innominate vein. The marks of an NIH Cardiomarker catheter (Medtronic Vascular, Danvers, Massachusetts) were used to measure the SVC. The SVC measured 12 mm superiorly, 3.5 mm at its narrowest point, and 13 mm inferiorly. The SVC was measured from both the anteroposterior and lateral projections. A 7 French (Fr) Goodale Lubin (GL) end-hole catheter (Medtronic Vascular) and a 0.035´´ Magic Torque wire (Meditech, Watertown, Massachusetts) were used to cross the region of stenosis. The Magic wire was then exchanged for a 0.035´´ Amplatz super-stiff wire (Boston Scientific/Scimed, Inc., Maple Grove, Minnesota), which was positioned in the SVC (cephalad). The GL catheter was then exchanged for a 12 Fr Mullins transseptal sheath (TSS), which was advanced over the super-stiff wire and positioned across the region of stenosis. A Palmaz 4010 stent (Cordis Corporation, Miami Lakes, Florida) was then hand-crimped on a 12 x 40 mm (outer balloon)/6 x 30 mm (inner balloon) NuMED balloon-in-balloon (BIB) catheter (NuMED, Hopkinton, New York). The stent was then advanced through the TSS and positioned across the region of stenosis. The TSS was then withdrawn and repositioned in the low SVC. Prior to stent delivery, a hand injection of contrast was performed through the TSS demonstrating proper stent position. The stent was then deployed by inflating the inner balloon to 5 atmospheres (Figure 2), followed a few seconds later by inflation of the outer balloon to 7 atmospheres (Figure 3). Repeat balloon dilation was performed utilizing only the outer balloon to try to further dilate the stent. Because of residual narrowing in the central portion of the stent, the BIB balloon was exchanged for a 10 x 30 mm Marshal balloon dilation catheter (Boston Scientific/Scimed, Inc.). Two inflations at 12 atmospheres were performed with further improvement in stent morphology (Figure 4). After stent implantation, a residual gradient of 5 mmHg was measured across the SVC. Angiography demonstrated proper stent position with significant improvement in flow of contrast across the SVC. The patient received 1 dose of intravenous kefzol after stent implantation and 3 more doses over the next 16 hours. Chest x-ray performed the next morning showed proper stent position. Two days later, both the facial and periorbital edema completely resolved, and the patient was discharged home on one aspirin daily for 3 months. One-week post-stent implantation, the patient was brought to the emergency room for evaluation of pallor and chest pain. The patient had behavioral problems, and earlier in the evening had an emotional outburst, which resulted in her restraint by several individuals. The history was not clear; however, the patient appeared to sustain trauma to her chest. Three steel rods in her door were reportedly bent after the incident. In the emergency room, the patient was hypotensive and tachycardic. She received a fluid bolus and was started on a dopamine infusion. Chest x-ray was significant for the presence of a wide mediastinum and cardiomegaly. Echocardiography demonstrated the presence of a large pericardial effusion. Pericardiocentesis and placement of a 6 Fr pericardial drain (pig-tail catheter) were performed under echocardiographic guidance. A total of 1,250 cc of dark-colored blood was obtained. The patient was then transferred to the cardiac catheterization laboratory. Contrast injected through a peripheral intravenous line in the right arm was visualized in the subclavian vein and superior portion of the SVC. There was no flow visualized through the SVC. A Cardiomarker catheter was then positioned within the SVC stent. A hand injection of contrast demonstrated the presence of a clot within the SVC stent (Figure 5). An angiogram performed in the ascending aorta demonstrated the presence of a small aneurysm directly across the inferior edge of the SVC stent (Figure 6). There was no active bleeding demonstrated in the ascending aorta. The patient was brought to the operating room for repair of the SVC and ascending aorta. There were two pinpoint perforations seen in the SVC and a 1 cm laceration in the ascending aorta directly across the SVC perforation site. There was no active bleeding in the ascending aorta; however, after removal of the clot on the surface of the ascending aorta, blood started to come out of the ascending aorta. The ascending aorta was repaired. The SVC was then opened and the stent removed. The SVC stent was redilated to a diameter of 14 mm and a 1 cm circumferential opening was created with the aid of a wire cutter (Figure 7). The stent was then positioned in the SVC with the newly created opening facing the insertion site of the innominate vein into the SVC. Pericardial patch repair of the SVC was then performed. A Goretex patch was also placed on both the SVC and ascending aorta, and a piece of polytetrafluoroethylene (PTFE) felt was added between the two vessels to further reinforce their border. This was done to prevent the inferior edge of the stent from eroding through both the SVC and ascending aorta. The patient was transferred to the intensive care unit and extubated the following day. Five days post-op, the patient underwent cardiac catheterization, which demonstrated a patent SVC with good flow of contrast across the repaired site. There was mild stenosis at the insertion site of the innominate vein into the SVC (Figure 8). The patient was discharged home on coumadin for a minimum period of 6 months. Three month follow-up evaluation demonstrated the stent in good position from both the postero-anterior and lateral projections (CXR) with good flow demonstrated through the SVC by both pulse and color flow Doppler. Discussion. Aortic laceration secondary to SVC stent implantation has been previously described.14 In the aforementioned case, the authors postulated that aortic laceration was secondary to flaring of the proximal and distal edges of the Palmaz stent secondary to the differing stress characteristics across the length of the stent (causing expansion from the ends of the stent progressively inward).14 In an accompanying editorial, Cheatham disagreed with this statement, and instead suggested that it was not the stent’s intrinsic characteristics that promote flaring at the edges, but rather the long, single balloon delivery catheter that causes the unwanted action.15 The editorial went further and suggested the use of a new balloon dilation catheter called the NuMED Balloon-in-Balloon (BIB) catheter. The characteristics of this balloon dilation catheter are well known to pediatric interventional cardiologists; the catheter consists of an inner balloon half the diameter and 1 cm shorter than the outer balloon. One of the advantages of using the BIB catheter is that when the inner balloon is first inflated, the central portion (middle) of the stent is expanded evenly without flaring of the leading/trailing edges The outer balloon is then inflated, with less flaring at the leading/trailing edges of the stent. The inner balloon is selected such that it is shorter than the length of the stent, whereas the outer balloon is just longer than the stent. In the above case report, a Palmaz 4010 stent (4 cm length) was selected because the authors felt that the P-308 stent (3 cm length) would have been too short to successfully open the region of SVC stenosis (Figure 1). In order to minimize flaring of both the leading and trailing edges of the stent, the Palmaz 4010 stent was mounted on a BIB balloon. Despite utilizing the BIB balloon, flaring of both the leading and trailing edges was still present (Figure 3). Because of the presence of residual stenosis in the mid portion of the stent, a high-pressure balloon was then selected to further expand the central portion of the stent. The balloon selected was shorter than the length of the stent (3 cm length) to minimize flaring of the leading/trailing stent edges. A 5 mmHg residual gradient remained after stent implantation. Despite the presence of a residual gradient, and region of central stenosis, the decision was made to leave well enough alone and possibly redilate the stent in 6 months. Angiographically, the SVC at its narrowest measured 9 mm and there was good flow of contrast through the vessel. Unfortunately, one-week after stent implantation, the patient was brought to the emergency department and was noted to have a large pericardial effusion. Pericardiocentesis was performed, and dark colored blood was aspirated. Cardiac catheterization demonstrated the presence of a clot in the SVC, and ascending aorta angiography demonstrated a small aneurysm in the rightward aspect of the ascending aorta directly across the inferior/trailing edge of the SVC stent. In the operating room, under direct vision, two pinpoint perforations were visualized in the SVC. Directly across the SVC perforation, an ascending aorta laceration was noted. Of note, the ascending aorta was not actively bleeding during the time of exploration. It is also worth noting that the blood aspirated during performance of the pericardiocentesis was dark in color (appeared to be venous), and that there was no active bleeding from the ascending aorta either angiographically or under direct vision. One can postulate that the patient had a slow venous leak in the SVC that could have been secondary to flaring of the trailing edge of the stent during the initial stent implantation procedure, and that over time the trailing edge of the stent eroded into the wall of both the SVC and the ascending aorta. Alternatively, SVC perforation and aortic laceration could have occurred a few hours before the patient presented to the emergency department after sustaining chest trauma. We suspect that if the ascending aorta had been perforated earlier (at the time of stent implantation), then the patient would have presented sooner and would have been more hemodynamically unstable. The fact that the patient remained well for one week and then suffered trauma to her chest is disconcerting because this event could have caused the trailing edge of the stent to push against the wall of the SVC, leading to SVC perforation and laceration of the ascending aorta. It is important to note that most of the stents that have previously been implanted in the SVC were shorter (Palmaz 188 and 308 series) and less stiff (Wallstent) in comparison to the Palmaz 4010 stent. The Palmaz 4010 is a biliary stent, which is longer and stiffer than the Palmaz 188/308 series (iliac stent). In retrospect, because the P-4010 stent is much stiffer, the flaring encountered during stent deployment could have made it easier for the sharp ends of the stent to puncture or erode through a thin walled vessel like the SVC. The IntraStent Double Strut LD stent (Intratherapeutics, St. Paul, Minnesota) would have been an alternative for this procedure. This stent is a new FDA-approved biliary stent recently adapted for use in patients with congenital heart disease.16–18 The IntraStent is more flexible, and has rounder leading and trailing edges. However, the IntraStent does not appear to have sufficient radial strength when compared to the Palmaz stent, and has previously been reported to collapse/recoil after implantation.19 Recently, a larger version of the Palmaz Corinthian stent (Cordis Corporation) has become available that is called the Genesis XD (extra diameter) stent; this stent has similar radial strength to the P-308 series, but has rounder leading/trailing edges and may prove to be a safer alternative to the currently available stents. In conclusion, we report a case of late perforation of the SVC with ascending aorta laceration presenting one week after stent implantation. The etiology of the SVC perforation could have been secondary to either flaring of the trailing edge of the stent during the initial stent implantation procedure or to chest trauma resulting in the trailing edge of the stent eroding into the wall of the SVC and ascending aorta. After review of our stent implantation procedure, we recommend the selection of a balloon (standard or BIB) that is shorter than the length of the stent for the initial stent implant. If further dilation is required, a larger diameter balloon (not to exceed the diameter of the nearest normal vessel) of similar or equal length (but not longer than the implanted stent), is then selected to further dilate the implanted stent. We believe that if these principles are followed, the risk of flaring of either the leading or trailing edge of the stent will be further minimized. We also suggest that ascending aorta angiography be performed after SVC stent implantation to make sure that the edge of the stent does not impinge on the ascending aorta. Further studies utilizing the Palmaz 4010 stent are also needed before routinely advocating the use of this stent in thin-walled venous vessels.
References
1. Mullins CE, O’Laughlin MP, Vick GW, et al. Implantation of balloon expandable intravascular grafts by catheterization in pulmonary arteries and systemic veins. Circulation 1988;77:188–199. 2. Benson LN, Hamilton F, Dasmahapatra M, et al. Percutaneous implantation of a balloon-expandable endoprosthesis for pulmonary artery stenosis: An experimental study. J Am Coll Cardiol 1991;18:1303–1308. 3. O’Laughlin MP, Perry SB, Lock JE, Mullins CE. Use of endovascular stents in congenital heart disease. Circulation 1991;83:1923–1939. 4. Hosking MCK, Benson LN, Nakanishi T, et al. Intravascular stent prosthesis for right ventricular outflow tract obstruction. J Am Coll Cardiol 1992;20:373–380. 5. Ward CJB, Mullins CE, Nihill MR, et al. Use of intravascular stents in systemic venous and systemic venous baffle obstructions. Circulation 1995;91:2948–2954. 6. Suarez de Lezo J, Pan M, Romero M, et al. Balloon expandable stent repair of severe coarctation of the aorta. Am Heart J 1995;129:1002–1008. 7. Bulbul AR, Bruckheimer E, Love JC, et al. Implantation of balloon expandable stents for coarctation of the aorta: Implantation data and short term results. Cathet Cardiovasc Diagn 1996;39:36–42. 8. Ebeid MR, Prieto LR, Latson LA. Use of balloon expandable stents for coarctation of the aorta: Initial results and intermediate term follow-up. J Am Coll Cardiol 1997;30:1847–1852. 9. O’Laughlin MP, Slack MC, Grifka RG, et al. Implantation and intermediate-term follow-up of stents in congenital heart disease. Circulation 1993;88:606–614. 10. Shaffer KM, Mullins CE, Grifka RG, et al. Intravascular stents in congenital heart disease: Short- and long-term results from a large single-center experience. J Am Coll Cardiol 1998;31:661–667. 11. Dodds GA, Harrison JK, O’Laughlin MP, et al. Relief of superior vena cava syndrome due to fibrosing mediastinitis using the Palmaz stent. Chest 1994;106:315–318. 12. Kee ST, Kinoshita L, Razava MK, et al. Superior vena cava syndrome: Treatment with catheter-directed thrombolysis and endovascular stent placement. Radiology 1998;206:187–193. 13. Hochrein J, Bashore TM, O’Laughlin MP, Harrison JK. Percutaneous stenting of superior vena cava syndrome: A case report and review of the literature. Am J Med 1998;104:78–84. 14. Evans J, Saba Z, Rosenfeld H, et al. Aortic laceration secondary to Palmaz stent placement for treatment of superior vena cava syndrome. Cathet Cardiovasc Intervent 2000;49:160–162. 15. Cheatham J. Editorial comment. A tragedy during Palmaz stent implant for SVC syndrome: Was it the stent or was it the balloon delivery system? Cathet Cardiovasc Intervent 2000;49:163–166. 16. Rutledge JM, Grifka RG, Nihill MR, et al. Initial experience with Intratherapeutics DoubleStrut LD stents in patients with congenital heart defects. J Am Coll Cardiol 2001;37(2 Suppl A):461A. 17. Ing FF, Mathewson JW, Cocalis M, et al. The new DoubleStrut stent: In vitro evaluation of stent geometry following overdilation and initial clinical experience in congenital heart disease. J Am Coll Cardiol 2001;37(2 Suppl A):467A. 18. Ing FF, Perry JC, Mathewson JW, Cocalis M, et al. Percutaneous implantation of large stents for the treatment of infants with severe post-operative branch pulmonary artery stenoses. J Am Coll Cardiol 2001;37(2 Suppl A):461A. 19. Recto MR, Grifka RG. IntraStent collapse/recoil following use in post-operative stenoses (Abstr). Cathet Cardiovasc Intervent 2001;53:146.