Aneurysm formation is a significant complication after. Endovascular Stent Grafts for Large Thoracic Aneurysms After Coarctation Repair

Transcription

1 Endovascular Stent Grafts for Large Thoracic Aneurysms After Coarctation Repair Shelby Kutty, MD, Roy K. Greenberg, MD, Scott Fletcher, MD, Lars G. Svensson, MD, and Larry A. Latson, MD Center for Pediatric and Congenital Heart Diseases, and Departments of Vascular Surgery and Cardiac Surgery, Cleveland Clinic Foundation, Cleveland, Ohio; and Creighton University University of Nebraska Joint Division of Pediatric Cardiology, Omaha, Nebraska Background. Aneurysm formation is a complication not bypass and subclavian ligation. Endografts were placed infrequently seen after repair of aortic coarctation and abutting the origin of the left common carotid artery. some may enlarge over time. Conventional management Seven patients were treated with Zenith endografts of large thoracic aneurysms after aortic coarctation repair (Cook, Inc, Bloomington, IN), and 2 with TAG devices has been akin to the surgical treatment of nonspecific(w.l. Gore & Associates, Flagstaff, AZ). Left common aneurysms; however, hypothermic circulatory arrest has carotid angioplasty and stenting was performed in 4 been more frequently required because of reoperations. patients. No major complications occurred. A mean We describe the treatment of a series of patients withfollow-up of 24 months (range, 6.4 to 48 months) demonstrated no late endoleaks, ruptures, conversions, or large aneurysms using novel endovascular techniques. Methods. The database of patients undergoing thoracic endograft placement was reviewed to identify those with thoracic aneurysms after aortic coarctation repair. Clinical, operative, and radiographic data were assessed. Follow-up imaging included spiral computed tomography (CT) scans immediately after deployment, at 6 months, and yearly thereafter. Results. Of 9 patients that were identified, 7 presented for elective repair and 2 were emergencies. The aneurysms measured 4.7 to 7.3 cm in diameter on spiral CT scans. Seven patients underwent carotid to subclavian migration. Conclusions. Placement of endovascular stent grafts is a less invasive approach for patients with thoracic aneurysm after aortic coarctation repair, provided there is no residual coarctation or arch hypoplasia. The potential to diminish the magnitude of the surgical procedure and consequences of aortic exposure in a reoperative field is promising and mandates further investigation. (Ann Thorac Surg 2008;85:1332 8) 2008 by The Society of Thoracic Surgeons Aneurysm formation is a significant complication after therapy for aortic coarctation (CoA). Aneurysms develop after surgical as well as transcatheter repair procedures of CoA. The reported prevalence of aneurysms after surgical repair of CoA is between 11% and 24%. The incidence is reported to increase with advancing interval after surgical or transcatheter therapy [1 3]. Older surgical techniques have been associated with development of aneurysms, as well as acute rupture and sudden death in long-term follow-up [4]. Although newer surgical techniques have supplanted older methods of open repair [1, 5], aneurysm development occurs independent of the surgical technique and has now been described after nearly every technique, including resection and end-to-end anastomosis, subclavian patch repair, and synthetic onlay patch repair [3, 6]. The reported incidence of aneurysms after transcatheter balloon procedures has varied greatly among different studies [2, 6 8], with some authors finding much higher Accepted for publication Jan 2, Address correspondence to Dr Greenberg, Department of Vascular Surgery/S61, Cleveland Clinic Foundation, 9500 Euclid Ave, Cleveland, OH 44195; greenbr@ccf.org. incidences [9, 10]. The subset of patients treated for recurrent CoA with balloon dilation also share risk of later aneurysm formation of 0% to 14% [2]. Untreated, aneurysms may enlarge and result in rupture or aortic dissection. Although maximal aortic diameter remains the strongest predictor of rupture for nonspecific aneurysms, the unique morphology of post-coa aneurysms causes skepticism with regard to the application of such data to the postoperative young patient population. The treatment of such problems can be complex given the inherent reoperative field (many patients have had multiple CoA repairs) with adherent pulmonary tissue, risk for aortobronchial fistula, the hypoplastic proximal arch, and the proximity of the lesion to the left common carotid artery, which may necessitate a staged approach using an elephant trunk graft technique. The potential advantage of an endovascular means of aneurysm exclusion in these circumstances is a marked reduction in the associated morbidity and mortality. Dr Greenberg discloses that he has a financial relationship with Cook, Inc and Boston Scientific by The Society of Thoracic Surgeons /08/$34.00 Published by Elsevier Inc doi: /j.athoracsur

2 Ann Thorac Surg KUTTY ET AL 2008;85: ENDOVASCULAR THERAPY FOR LARGE ANEURYSMS 1333 Material and Methods Clinical, radiographic, and laboratory data were collected prospectively, within the context of two physiciansponsored investigational device exemption studies, on all patients treated for thoracic aortic pathology with investigational devices. The study was approved by the Institutional Review Board. The database was retrospectively reviewed to identify patients who had aneurysms of previously treated CoA. All patients were considered high risk for an open surgical repair and agreed to participate in the study. All patients signed a consent form approved by our Institutional Review Board. The determination of high risk for open repair was made after the patient was evaluated by a surgeon and interventionalist. In general, patients were considered high risk if they had undergone multiple prior repairs, had unfavorable body habitus, or if the anatomy of their aortic lesions mandated a two-stage (elephant trunk graft) approach. Several patients were treated with open surgical repair concurrently with enrollment into an endovascular treatment paradigm. Devices Zenith TX1 or TX2 thoracic endografts (Cook Inc, Bloomington, IN) and the TAG device (W. L. Gore & Associates, Flagstaff, AZ), were used for aneurysm exclusion. Zenith thoracic grafts are made from stainless steel Z-stents attached to a polyester fabric in a one-piece (TX1) or a two-piece (TX2) design. The device requires a 20F to 22F delivery system. The TAG device is an expanded polytetrafluoroethylene (eptfe) graft with an outer selfexpanding nitinol support structure. It is delivered through a 20F to 24F sheath. It can be used as a single device or multiple components can be overlapped to achieve the desired length. Procedure and Device Delivery General anesthesia was used in 7 of 9 patients. Procedures were performed in an endovascular operating room with a fixed imaging system (Siemens Angiostar or AXIOM Artis FA, Siemens AG, Erlangen, Germany). Graft lengths were designed using three-dimensional image processing on a vascular-based workstation (Aquarius, TeraRecon Inc, San Mateo, CA, or Leonardo, Siemens, Inc, Malvern, PA). The prosthesis diameters were intended to be 10% to 20% larger than the measured aortic diameters, and the device lengths exceeded the aneurysm length by at least 20 mm proximally and distally whenever possible. In the setting of proximal disease, incorporation of the left carotid artery through a fenestration in the endograft enabled extension of the sealing and fixation zones. Femoral or iliac access was obtained after arterial exposure with an oblique incision in conjunction with a counter incision. A second brachial arterial access site was also used in all patients. Heparin anticoagulation was used to maintain activated clotting time exceeding 300 seconds. Endografts were inserted over a stiff wire (Amplatz or Lunderquist Wire, Cook, Inc, Bloomington, IN) and deployed after ensuring accurate positioning with the aid of small-bolus angiography. Device conformity at the sealing locations was modified by selective ballooning. After deployment, a completion angiogram was performed. The delivery system for nonfenestrated Zenith thoracic graft evolved into a braided sheath covering a nitinolbased cannula (Z-Track Plus), and the TAG was inserted through a 20F to 24F Check-flow sheath (Cook, Inc, Bloomington, IN). The Zenith devices incorporating a fenestration for the left common carotid artery included a preloaded catheter and wire that was used to selectively cannulate the left common carotid from within the device before graft deployment. This was snared from the brachial access to establish through and through access to align the carotid fenestration. A balloon-expandable stent was deployed in a retrograde fashion through a carotid subclavian bypass graft accessed by way of a brachial puncture to ensure alignment of the fenestration with the carotid orifice. Follow-Up Four-view chest roentgenograms and spiral computed tomography (CT) scans were performed before the hospital discharge, at 1, 6, and 12 months of follow-up, and yearly thereafter. Measurements of aortic size and device migration were performed using three-dimensional imaging and centerline of flow techniques. Patients who required carotid stent placement in addition to their endovascular graft were maintained on clopidogrel for 30 days. Duplex ultrasound (DUS) scans were performed to assess carotid subclavian bypass grafts, when necessary. Endovascular outcomes are reported in accordance with the most recent reporting standards document for endovascular aneurysm repair [11], and migration was assessed using a modification of this standard [12]. Results Between May 2001 and September 2005, post-coa aneurysms were treated with endovascular stent grafts in 9 patients. Patients were a mean age of 36 years (range, 22 to 43 years) at the time of diagnosis. All patients had previously undergone surgical repair for CoA one or more times as an infant or young adult, or both. Dacron (DuPont, Wilmington, DE) patch aortoplasty was the surgical technique used in 4 patients and the type of prior repair was unknown in the remaining 5 patients. Seven patients reported no symptoms because aneurysms were detected on routine follow-up screening (Fig 1). Two patients presented as emergencies: 1 with sudden onset of severe chest pain and presumed rapid aneurysm growth and the other with massive hemoptysis due to aortobronchial fistula. Five patients were hypertensive at the time of presentation. None of the patients had symptoms attributable to aortic narrowing. The mean aneurysm diameter on spiral CT scan at the time of presentation was 6.3 cm (range, 4.7 to 7.3 cm). Zenith endografts were used in 7 patients and TAG devices were used in

3 1334 KUTTY ET AL Ann Thorac Surg ENDOVASCULAR THERAPY FOR LARGE ANEURYSMS 2008;85: Follow-Up The mean duration of follow-up after stent graft placement was 24 months (range, 6.4 to 48 months). No instance of graft migration or late endoleak was found on follow-up. Aneurysm shrinkage was noted in all patients on follow-up CT scans within 12 months of the procedure (Fig 6). The mean aneurysm diameter had decreased to 4.4 cm (range, 3.9 to 5 cm) in the most recent follow-up scan. Subclavian steal syndrome (dizziness associated with left arm use on follow-up evaluation at 3 months) developed in 1 patient. This patient had undergone emergency stent graft placement after presentation with an impending aneurysm rupture. A carotid subclavian bypass graft was performed after symptoms developed. This patient continues to have minimal left arm symptoms, but claudication is largely resolved. Fig 1. A reconstructed spiral computed tomography image shows a large aneurysm involving the left subclavian artery. the 2 urgently treated patients. Patient characteristics and procedural results are reported in Table 1. Carotid subclavian bypasses using a eptfe prosthesis (Fig 2) (W. L. Gore & Associates, Flagstaff, AZ) were done in 6 patients before the endovascular procedure, and a similar bypass graft was done in 1 patient after repair to alleviate arm claudication. One patient underwent a staged hybrid approach where an elephant trunk graft was placed before completion with an endovascular graft because of an ascending aorta and arch aneurysm. Four patients underwent placement of left common carotid artery stents (Figs 3 and 4) to ensure patency of that vessel because of potential device encroachment. One patient required placement of a Palmaz (Cordis Corp/ Johnson & Johnson Health Care Systems Inc, Piscataway, NJ) stent in the mid arch for alignment. A single patient was treated with an arch fenestrated device to accommodate a hypoplastic arch that involved the left common carotid artery. Only one patient required an iliac conduit for graft introduction as a result of small iliac vasculature. Graft deployment was technically successful in all patients, and no adverse events occurred during implantation (Fig 5). The mean operative time was minutes (range, 70 to 170 minutes). The mean fluoroscopy time was 22.9 minutes (range, 9.5 to 45.5 minutes). The mean volume of contrast used was ml (range, 85 to 266 ml). All patients survived, and no major complications (local vascular, renal failure, spinal cord injury, and other neurologic) occurred after stent graft placement. Postprocedural imaging demonstrated residual endoleaks in 2 patients due to retrograde flow through the left subclavian artery. Leak in 1 patient was abolished with Cyanoacrylate glue embolization (Cordis Neurovascular), and the other resolved without treatment. Comment The development of late complications after surgical repair of CoA is not uncommon. Potential issues include recurrent CoA, aneurysm formation, and aortobronchial or aortoesophageal fistula. Aneurysm formation has been attributed to intrinsic aortic medial degeneration or failure of the suture line (aortic aortic or aortic prosthetic) or material (polyester patch or graft dilation). Pinson and colleagues [3] found a 30% incidence of aneurysms after surgical repair of CoA. In their series of 215 patients, follow-up angiography showed aneurysms had developed in 64 patients. The aneurysms appeared as diffuse enlargements, focal anterior bulges, or focal posterior bulges [3]. Knyshov and colleagues [6] reported a 5.4% Fig 2. A spiral computed tomography performed in the patient in Fig 1 immediately after deployment of the endovascular stent graft also shows a carotid subclavian bypass graft.

4 Table 1. Patient and Implantation Characteristics Patient Ages at CoA Repair Age at Endograft Placement, y Presentation Comorbidities Size on CT, cm Device Name Secondary Procedure Endoleak, Comments 1 23 y 43.8 Routine screening Hypertension, 5.4 Zenith,TX1 LCCA stent hyperlipidemia 2 9 d; 4 mon 26.8 Routine screening None 6.7 Zenith,TX2 CSBPG Iliac conduit 3 6 y; 16 y 41.9 Routine screening None 6.4 Zenith,TX1 CSBPG Small endoleak, resolved 4 2 y 27.7 Routine screening Hypertension 4.7 Zenith,TX2 CSBPG, LCCA stent 5 1 mon; 8 y 36.9 Routine screening AVR with homograft at 18 y; hyperlipidemia, hypertension 6 22 y 46.0 Routine screening Bicuspid Ao valve, severe AS, mod AI, hypertension 7.3 Zenith,TX2 CSBPG, LCCA stent, Asc Ao stent 6 Zenith,TX2 AVR with 23-mm bovine, pericardial valve and elephant trunk graft, 2 weeks before stent graft 7 1 mon 22.9 Routine screening None 7 Zenith,TX2 CSBPG, LCCA stent 8 13 y 39.4 Contained aortic rupture 9 2 in childhood; 1at20y 43 Rupture and aortobronchial fistula Retrograde subclavian flow treated with glue embolization Hypertension 6.7 TAG CSBPG CSBPG after stent graft placement None 6 TAG CSBPG Stent graft placement while patient on ECMO AI aortic insufficiency; AS aortic stenosis; Asc Ao ascending aorta; AVR Aortic valve replacement; CoA coarctation of the aorta; CSBPG carotid subclavian bypass graft; CT computerized tomography; ECMO extracorporeal membrane oxygenation; LCCA left common carotid artery. Ann Thorac Surg KUTTY ET AL 2008;85: ENDOVASCULAR THERAPY FOR LARGE ANEURYSMS 1335

5 1336 KUTTY ET AL Ann Thorac Surg ENDOVASCULAR THERAPY FOR LARGE ANEURYSMS 2008;85: Fig 3. An intraoperative angiogram demonstrates placement of the endovascular stent graft and left common carotid artery stent for a large aneurysm with involvement of the left subclavian artery. Fig 5. A reconstructed spiral computed tomography image of a scan done immediately after placement of the stent graft and left common carotid artery stent. incidence (48 of 891 patients) of aneurysm formation in a series of patients with CoA monitored for up to 20 years postoperatively. Patch repair techniques were clearly more prone to late aneurysm formation. Ninety percent of patients who had presented with aneurysms had undergone synthetic patch aortoplasty, and 8% had resection, followed by aortic end-to-end anastomosis. Extensive resection of the intima with or without patch angioplasty may also be an important predisposing factor for aneurysm formation opposite the aortotomy [13]. Regardless of the cause, ruptures with subsequent deleterious outcomes have been noted in this relatively young patient population. Fatal rupture occurred in 38% of patients noted to have large aneurysms after CoA repair who did not undergo treatment [6]. Rupture and Fig 4. An intraoperative angiogram demonstrates placement of the endovascular stent graft and left common carotid artery stent for a large aneurysm with involvement of the left subclavian artery. Fig 6. This spiral computed tomography was done in the patient in Fig 5 at 1-year of follow-up.

6 Ann Thorac Surg KUTTY ET AL 2008;85: ENDOVASCULAR THERAPY FOR LARGE ANEURYSMS 1337 Fig 7. This computed tomography scan was obtained immediately after implantation of an endovascular graft for an aortobronchial fistula that developed following a coarctation repair. The patient was treated with long-term antibiotics and remains well, without evidence of infection 5 years after the procedure. dissection can occur at smaller diameters in these patients than with other aneurysm types. Such observations mandate that follow-up studies be performed on all patients treated for CoA, yet the timing of aneurysm formation and thus the interval of requisite follow-up studies remain to be determined. Most surgeons advocate treatment of large aneurysms when the size exceeds 5 cm or the morphology is concerning for potential rupture (asymmetry or dissection). The surgical approach is dependent on the proximal and distal extent of the disease. When the arch is involved, either aneurysmal or markedly hypoplastic, a staged approach is generally taken whereby a median sternotomy is performed and an elephant trunk graft is placed under hypothermic circulatory arrest. Subsequent completion of the elephant trunk graft is accomplished through a left lateral thoracotomy, but in postsurgical CoA repair patients, scar tissue from prior thoracotomy may be encountered. Endovascular therapy is a relatively new treatment modality for thoracic aortic disease. Prospective studies evaluating endovascular repair of thoracic aortic aneurysms have been reported [14 16] with variable success. Many of these repairs are complex. The inherent tortuosity of the arch, proximity of the supraaortic trunk vessels, and hypoplastic nature of the proximal aorta can create technical difficulties. Access vessels may also be small, particularly with regard to the necessity of sheaths 20F or larger, creating the potential need for an iliac conduit [15]. Challenges with accurate device deployment, device migration [12], and aneurysm growth despite complete exclusion [16] have all been described. The typically young ages of post CoA repair patients magnify these challenges. Iliac vessels in such patients are often small, and the proximity of the disease to the arch vessels mandates accurately placed devices with particularly durable systems to ensure stability and prolonged sealing. The left subclavian artery was covered in every case in this series, although only 6 of the 9 patients underwent extraanatomic reconstructions (carotid subclavian bypass grafting) before endograft deployment. Such a bypass graft, when performed before deployment, ensures retrograde access to the proximal left carotid system should the aortic device encroach on the vessel. Additional justification includes a higher likelihood of claudication in a young active population in contrast to older, perhaps less active patients with nonspecific or atherosclerotic aneurysms. However, the long-term patency of these bypass grafts has not been defined in a young and otherwise healthy population. In all cases where there was a short proximal neck, which mandated graft material placement proximal to the ostium of the left common carotid artery, a scalloped fenestration was created to incorporate the left common carotid artery, allowing the graft material to traverse the entire arch to abut the innominate artery. A carotid subclavian bypass was performed not only to avoid claudication but, more important, to provide retrograde brachial access to the left carotid system by traversing the bypass graft with catheters and wires. We used fenestrated grafts over other techniques such as carotid carotid bypass grafting in an effort to avoid single-vessel cerebrovascular supply supplemented by the concept that a failed fenestration could always be treated with a subsequent extraanatomic bypass graft. Arch angulation in these young patients with prior CoA repairs tend to be rather steep, and thus, not all endovascular prostheses would be suitable for treatment of this pathology. Improvements in the delivery systems and devices were marked throughout enrollment of these patients. Flexibility was conferred by the use of braided delivery sheaths and nitinol cannulas. Owing to the conformational changes that are likely to occur, we tried to avoid balloon-expandable stents. Only 1 patient required such a stent to help align the proximal self-expanding stent graft with the arch angulation. Although the endovascular approach offers advantages compared with open surgery in terms of dissection, physiologic response, recovery time, and length of stay, the durability of the endovascular approach remains in question. Devices with proximal barbs, which discourage migration, were preferentially used in this series for all but 2 emergency patients. Most of the devices were also tapered to conform to the patients anatomy (small proximal arch diameters of 22 to 32 mm) with larger distal aortic diameters (up to 40 mm). The overall device length was short (less than 20 cm) in all but one patient, who presented with aortobronchial fistula arising from either a proximal or distal anastomosis of a Dacron graft bypassing the narrowed aortic segment, to avoid any unnecessary risk of paralysis that has been noted with longer devices [17]. We believe from past experience that no average sized male patient should be left with a residual aortic size of less than 20 to 22 mm diameter, and for females, 18 to 20 mm [18]. Indeed, placement of small sized stent grafts in these patients may be contraindicated with the current devices owing to the risk of a bird s beak deformity or collapse of devices within a too small aortic lumen. The endovascular approach should not be the first choice of

7 1338 KUTTY ET AL Ann Thorac Surg ENDOVASCULAR THERAPY FOR LARGE ANEURYSMS 2008;85: therapy if a patient will be left with a significant residual stenosis from proximal aortic arch hypoplasia. We observed no cases of stenoses within the carotid stents placed during our follow-up assessments using both CT and DUS scans. Although the potential for such stenoses exists, it is likely that the incidence of de novo stenosis within a stent in a patient without atherosclerotic disease would be uncommon and likely with a low risk of emboli generation. In 1 patient, an elephant trunk graft was placed before the endovascular repair. Arch involvement with concomitant coronary issues prompted this approach, which has been noted to be durable in more traditional aneurysm patients [18,19]. Finally, the treatment of presumed contaminated fields, such as the patient presenting with an aortobronchial fistula, may not eliminate the infectious source but is useful in rapidly ameliorating hemorrhage. The single patient in this series that had this problem presented in extremis, was treated urgently, and ultimately required extracorporeal membrane oxygenation for 5 days. Although a large postoperative lung abscess was noted (Fig 7), the abscess completely resolved over time and the patient remains well at more than 5 years after the endovascular procedure. Our experience suggests that treatment with endovascular stent grafts is a novel, less invasive approach for patients who develop thoracic aneurysms after repair of CoA. Endovascular therapy of this complex patient population offers some advantages compared with conventional treatment of late, large aneurysms that may be sequelae of CoA repair. The endovascular approach can potentially diminish the magnitude of the reparative procedure and consequences of aortic exposure in a reoperative field. Given the long life expectancy of most patients after CoA repair, the durability of endovascular devices is of paramount interest. Long-term follow-up using magnetic resonance imaging (MRI) or CT angiography is important in these patients. Either modality provides adequate information, but MRI could be advantageous owing to avoidance of repeated radiation exposure in this young population. Although surgery may remain the standard of care for many years, endovascular techniques have been developed to treat failures of surgical techniques. Endovascular devices are currently limited to use in patients with appropriate anatomy who are willing to comply with follow-up studies. References 1. Bromberg BI, Beekman RH, Rocchini AP, Snider AR, Bank ER, Heidelberger K, et al. Aortic aneurysm after patch aortoplasty repair of coarctation: A prospective analysis of prevalence, screening tests and risks. J Am Coll Cardiol 1989;14: Ovaert C, Benson LN, Nykanen D, Freedom RM. Transcatheter treatment of coarctation of the aorta: A review. Pediatr Cardiol 1998 Jan-Feb;19:27 44; discussion Pinzon JL, Burrows PE, Benson LN, Moes CA, Lightfoot NE, Williams WG, et al. Repair of coarctation of the aorta in children: Postoperative morphology. Radiology 1991;180: Emmrich K, Herbst M, Trenckmann H, Schulz HG, Wohlgemuth B. Severe late complications after operative correction of aortic coarctation by interposition of prosthesis. J Cardiovasc Surg (Torino) 1982;23: del Nido PJ, Williams WG, Wilson GJ, Coles JG, Moes CA, Hosokawa Y, et al. Synthetic patch angioplasty for repair of coarctation of the aorta: Experience with aneurysm formation. Circulation 1986;74:I Knyshov GV, Sitar LL, Glagola MD, Atamanyuk MY. Aortic aneurysms at the site of the repair of coarctation of the aorta: A review of 48 patients. Ann Thorac Surg 1996;61: Mendelsohn AM, Lloyd TR, Crowley DC, Sandhu SK, Kocis KC, Beekman RH, 3rd. Late follow-up of balloon angioplasty in children with a native coarctation of the aorta. Am J Cardiol 1994;74: Rao PS, Galal O, Smith PA, Wilson AD. Five- to nine-year follow-up results of balloon angioplasty of native aortic coarctation in infants and children. J Am Coll Cardiol 1996;27: Brandt B, 3rd, Marvin WJ, Jr, Rose EF, Mahoney LT. Surgical treatment of coarctation of the aorta after balloon angioplasty. J Thorac Cardiovasc Surg 1987;94: Cooper RS, Ritter SB, Rothe WB, Chen CK, Griepp R, Golinko RJ. Angioplasty for coarctation of the aorta: Longterm results. Circulation 1987;75: Chaikof EL, Blankensteijn JD, Harris PL, White GH, Zarins CK, Bernhard VM, et al. Reporting standards for endovascular aortic aneurysm repair. J Vasc Surg 2002;35: O Neill S, Greenberg RK, Resch T, Bathurst S, Fleming D, Kashyap V, et al. An evaluation of centerline of flow measurement techniques to assess migration after thoracic endovascular aneurysm repair. J Vasc Surg 2006; 43: DeSanto A, Bills RG, King H, Waller B, Brown JW. Pathogenesis of aneurysm formation opposite prosthetic patches used for coarctation repair. an experimental study. J Thorac Cardiovasc Surg 1987;94: Dake MD, Miller DC, Mitchell RS, Semba CP, Moore KA, Sakai T. The first generation of endovascular stent-grafts for patients with aneurysms of the descending thoracic aorta. J Thorac Cardiovasc Surg 1998;116:689,703; discussion Greenberg RK, O Neill S, Walker E, Haddad F, Lyden SP, Svensson LG, et al. Endovascular repair of thoracic aortic lesions with the zenith TX1 and TX2 thoracic grafts: Intermediate-term results. J Vasc Surg 2005;41: Makaroun MS, Dillavou ED, Kee ST, Sicard G, Chaikof E, Bavaria J, et al. Endovascular treatment of thoracic aortic aneurysms: Results of the phase II multicenter trial of the GORE TAG thoracic endoprosthesis. J Vasc Surg 2005; 41: Greenberg R, Resch T, Nyman U, Lindh M, Brunkwall J, Brunkwall P, et al. Endovascular repair of descending thoracic aortic aneurysms: An early experience with intermediate-term follow-up. J Vasc Surg 2000;31: Svensson LG, Kim KH, Blackstone EH, Alster JM, McCarthy PM, Greenberg RK, et al. Elephant trunk procedure: newer indications and uses. Ann Thorac Surg 2004;78: Greenberg RK, Haddad F, Svensson LG, O Neill S, Walker E, Lyden SP, et al. Hybrid approaches to thoracic aortic aneurysms: The role of endovascular elephant trunk completion. Circulation 2005;112: