- Email: [email protected]
Experimental Evaluation of Complete Endovascular Arch Reconstruction by In Situ Retrograde Fenestration
Ludovic Canaud, MD, PhD, Elsa Madeleine Faure, MD, Pascal Branchereau, MD, Baris Ata Ozdemir, BS, MRCS, Charles Marty-An e, MD, PhD, and Pierre Alric, MD, PhD Department of Thoracic and Vascular Surgery, Arnaud de Villeneuve Hospital, Montpellier; and INSERM U 1046, Montpellier, France
Background. The aim of this experimental study was to assess the feasibility of complete endovascular arch reconstruction by in situ retrograde fenestration and to investigate the impact of stent-graft material on stent-graft fenestrations. Methods. The experiments were performed using 8 cadaveric human thoracic aortas (aortic arch) using 2 different stent-graft types: woven polyester (Valiant Captivia; Medtronic Vascular, Santa Rosa, CA) and expanded polytetrafluoroethylene (conformable [C]-TAG; W.L. Gore & Associates, Flagstaff, AZ). A benchtop aortic pulsatile flow model was used. Stent-grafts were deployed into the aortic arch, covering the ostia of the supraaortic trunks. A 5-mm 30-degree angioscope was introduced into the ascending aorta to monitor the procedure. Retrograde fenestration and deployment of the balloon expandable stent-graft was performed sequentially for each supraaortic trunk. Subsequent to stent-graft
explantation, macroscopic evaluation of each fenestration was performed. Results. All attempts to fenestrate the C-TAG and Valiant stent-grafts and implant the covered stent through the supraaortic trunks were successful. In all cases, branch stents were patent and no endoleak was evident. The Valiant stent-graft was easier to puncture because of the higher radial force of the stent-graft providing better counterpressure; however, stent-graft material had no impact on the quality of fenestrations. Conclusions. Total endovascular repair of the aortic arch through in situ retrograde fenestration of stent-grafts is feasible. The behavior of the 2 types of stent-graft was significantly different while the fenestrations were fashioned, but stent-graft material had no impact on the quality of fenestrations.
D
conditions of the aortic arch, the majority of these adjunctive procedures remain major operations and are associated with significant perioperative mortality. Branched stent-grafts that permit completely percutaneous aortic arch repair have been proposed [3]. The disadvantages of this modular approach include the time required time to manufacture and deliver custom-made stent-grafts for urgent cases and the high costs associated these sorts of modular devices. Most notably, there is a high rate of embolism associated with this approach, which is probably related to the complexity of multibranched unibody stent-graft deployment. The concept of retrograde in situ fenestration combines intentional stent-graft coverage of the supraaortic trunks with subsequent reestablishment of blood flow by retrograde puncture of the device. This approach is an attractive alternative that eliminates the need for preoperative custom tailoring (allowing repair of emergent cases) and the inherent risk of cerebral embolism associated with catheterization of difficulties side branches. The aim of this experimental study was to assess the feasibility of complete endovascular arch reconstruction by in situ retrograde fenestration and to investigate whether the quality of stent-graft fenestrations varied with stent-graft material.
eep hypothermic circulatory arrest is required for cerebral protection during surgical repair of the aortic arch with replacement [1]. Mortality and morbidity associated with transverse aortic arch replacement in the standard-risk population has decreased over the past few decades with the implementation of various modifications of surgical technique. Despite these recent advances, aortic arch reconstruction remains challenging, particularly in elderly patients, those requiring emergency repair, and those with major preexisting comorbidities. Many patients are deemed unsuitable for open repair. More recently a combined endovascular and open approach has been adopted as a valuable alternative, consisting of supraaortic debranching and revascularization followed by stent-graft deployment [2]. Debranching is performed to provide an appropriate landing zone for the stent-graft while preserving perfusion to the supraaortic trunks. Although this approach provides an attractive alternative for the treatment of pathologic
Accepted for publication July 9, 2014. Address correspondence to Dr Canaud, Service de Chirurgie Vasculaire et Thoracique, Hopital A de Villeneuve, 191 av Doyen Gaston Giraud 34090, Montpellier, France; e-mail: [email protected]. ˇ
ADULT CARDIAC
Experimental Evaluation of Complete Endovascular Arch Reconstruction by In Situ Retrograde Fenestration
Ó 2014 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2014;98:2086–91) Ó 2014 by The Society of Thoracic Surgeons
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2014.07.024
Ann Thorac Surg 2014;98:2086–91
CANAUD ET AL COMPLETE ENDOVASCULAR AORTIC ARCH RECONSTRUCTION
2087
Material and Methods
ADULT CARDIAC
The study was approved by our institutional review committee. Consent was obtained from the relatives of each cadaveric aorta donor.
Harvesting and Preparation of Aortas In accordance with French regulation, 9 fresh nonaneurysmal human aortas were harvested at autopsy from men and women who had died up to 4 days previously. The aortas were harvested from 2 cm above the level of the aortic valve proximally to the iliac bifurcation distally. The brachiocephalic trunk, the left common carotid artery, and the left subclavian artery were harvested to maximal length. Each aorta was immediately packed in ice and maintained at 4 C. Experiments were performed within 2 hours of harvest. Aortic sections were sent to the Department of Pathology for microscopic evaluation (staining with hematoxylin and eosin) to confirm the presence of 3 layers of aortic wall comparable to fresh specimens.
Bench Test Model A previously described benchtop closed-system pulsatile flow model was used to mimic aortic flow and pressure as close to normal physiologic conditions as possible [4].
Fig 1. Benchtop closed-system pulsatile flow model. A graft is anastomosed to the distal segments of the aortic arch side branches. To monitor the procedure, a 5-mm angioscope connected to a camera was introduced into the ascending aorta.
Experimental Setup After harvest of the entire aorta, an 8-mm knitted polyester graft was anastomosed to the distal part of each supraaortic trunk. The distal end of the graft was then connected to a closed circuit to simulate antegrade flow into the aortic branch vessels during the experiment. Abdominal aortic branches were ligated 1 cm distal to their origin, and intercostal and lumbar arteries were oversewn. Each aorta was then coupled to the closed-system pulsatile flow model. Once each aorta was incorporated into the circulatory circuit, the pump was activated, leading to pulsatile flow: 60 pulses/min and pressure of 150/80 mm Hg. MONITORING THE PROCEDURE. To monitor the procedure, a 5-mm 30-degree angioscope (Richard Wolf, Vernon Hills, IL,) connected to a video camera was introduced into the ascending aorta through a simple 3-0 pursestring Vicryl suture (Ethicon, Somerville, NJ) (Fig 1). STENT-GRAFTS. Two types of commercial stent-grafts of 100-mm length (Fig 2) were used: the Valiant Captivia stent-graft (Medtronic Vascular, Santa Rosa, CA) and the conformable TAG (C-TAG) stent-graft (WL Gore & Associates, Flagstaff, AZ). The Valiant is composed of a nitinol stent framework between layers of polyester graft. Individual stents are sutured to the outside of the polyester graft material. The proximal end features an open bare stent segment. The C-TAG is composed of a symmetrically expanded polytetrafluoroethylene tube reinforced externally with a layer of expanded polytetrafluoroethylene. An exoskeleton consisting of nitinol stents is attached to cover the length of the graft. PREPARATION OF AORTA.
STENT-GRAFT PLACEMENT. Stent-grafts were deployed into the aortic arch, covering the ostia of the supraaortic trunks. Delivery of the covered stent was performed in a retrograde fashion. RETROGRADE FENESTRATION. Retrograde fenestration and deployment of the balloon expandable stent-graft was performed sequentially for each of the supraaortic trunks (brachiocephalic trunk, left common carotid artery, and subclavian artery) using the same technique. The puncture of the stent-graft was performed using a 20-gauge needle, and a 0.035-inch guide wire was advanced through the aperture into the descending aorta. The graft puncture site was subsequently dilated by advancing a 5F introducer sheath with its dilator tip over the 0.035inch guide wire. This was followed by dilation of the fenestration with a standard 4-mm angioplasty balloon (Wanda; Boston Scientific, Natick, MA). Thereafter, a balloon-expandable covered stent (Atrium Medical Corp, Hudson, NH) that was 22-mm in length (the diameter of which was selected by referencing the diameter of the supraaortic vessel treated; systematic oversizing of 1 mm), was passed and positioned across the fenestration. The balloon-expandable stent-graft was then deployed 5 mm into the aorta. Proximal flaring of the covered stent was initially performed using a noncompliant balloon (Coda; Cook Medical, Bloomington, IL) introduced through the supraaortic vessel. Final flaring of the covered stent was then performed using the same noncompliant balloon introduced through the abdominal aorta.
2088
CANAUD ET AL COMPLETE ENDOVASCULAR AORTIC ARCH RECONSTRUCTION
Ann Thorac Surg 2014;98:2086–91
ADULT CARDIAC
Fig 2. (A) Internal and (B) external views after stent-graft explantation after complete endovascular arch reconstruction by in situ retrograde fenestration. The image confirms secure fixation and sealing of the retrograde fenestration.
STENT-GRAFT EXPLANTATION. After completion of the endovascular arch reconstruction by in situ retrograde fenestration, each stent-graft was explanted. Macroscopic qualitative evaluation of each fenestration was performed.
Results Harvesting and Preparation of Aortas Eight fresh human nonaneurysmal aortas were harvested from 6 men and 2 women (mean age, 38.4 years; range, 29–51 years) who had died within a maximum of 4 days (mean, 1.2 days). The length of the brachiocephalic trunk, the left common carotid artery, and the left subclavian artery varied from 4 to 8 cm. The mean aortic diameter measured at the level of the left ascending aorta level was 26.5 2.2 mm (range, 24–29 mm) before explantation. Histologic analysis of the harvested aortas at the start of the study revealed that all 8 aortas consisted of 3 layers and were comparable to fresh aortas.
Comment An anatomically unsuitable proximal neck is the most common reason for patient exclusion from endovascular aortic repair. To maintain aortic side branch patency, thoracic stent-graft modifications have been proposed, featuring scallops, fenestrations, and branches. Fenestrated and branched stent-grafts have shown promising midterm results in selected patients [5]. All these systems are customized, requiring accurate preoperative planning. Target-vessel cannulation can be technically challenging and time-consuming even for experienced operators, especially in the presence of complex tortuous anatomy. Despite the custom-made nature of these devices and extensive preoperative planning, graft rotation and misalignment of the fenestration/vessel ostium interface can still occur. In addition, the aorta may change configuration after insertion of the semirigid stent-graft and thus alter alignment of the side branches. Furthermore, factors such as the delay in device planning
Initial Retrograde Fenestration Procedure Complete endovascular arch reconstruction by in situ retrograde fenestration was achieved in all cases (Fig 2). The mean procedure time was 10 2 minutes. A total of 24 balloon-expandable covered stents were successfully inserted. The Valiant stent-graft was easier to puncture because the higher radial force of the stent-graft provided a better counterbrace. Completion angioscopic assessment demonstrated the patency of stent-graft branches and no type III endoleak in all cases.
Stent-Graft Explantation Stent-graft explantation confirmed that secure fixation and seal of retrograde fenestrations was achieved in all cases. Balloon dilation and deployment of the balloonexpandable stent-grafts after fenestration was not associated with tears of the graft material, and the gross integrity of the fabric was always maintained. Endograft material (woven polyester versus expanded polytetrafluoroethylene) had no impact on the quality of fenestrations (Fig 3).
Fig 3. No tearing of the graft material was observed, and the gross integrity of the fabric was always maintained. Endograft material had no impact on the quality of fenestrations: (A) expanded polytetrafluoroethylene versus (B) woven polyester.
CANAUD ET AL COMPLETE ENDOVASCULAR AORTIC ARCH RECONSTRUCTION
and manufacturing, anatomic and technical limitations, and expense limit the widespread uptake of this technology, which additionally is unsuitable for emergent cases. Above all, the technical difficulty of side branch catheterization results in an inherently high risk of cerebral embolism. The use of readily available “off the shelf” endovascular materials would decrease the cost of endovascular repair and increase the availability of these minimally invasive techniques to a greater number of patients. Thus retrograde in situ fenestration seems to offer an appropriate solution to most of these issues, allowing more accurate fenestration placement with less reliance on preoperative imaging, as well as a reduction in the number of catheter and guide wire manipulations in the aortic arch. McWilliams and colleagues [6] were the first to report successful retrograde in situ fenestration in bench and animal models, followed by clinical application in the left subclavian artery. Sonesson and coworkers [7] used this approach in the emergency clinical setting to treat an acute aortic arch rupture. These clinical reports used widely varying techniques to fashion fenestrations, and the reproducibility of this approach, especially in an experimental setting, has to be demonstrated before more widespread clinical application. Our cadaveric model demonstrates the feasibility of complete aortic arch reconstruction using the retrograde approach. This approach, which was successful in all the models using “universal” noncustomized stent-grafts, render it suitable for urgent and emergency cases. A potential disadvantage of the retrograde technique is that the stent-graft that lies against the orifice might be pushed away during mechanical puncture. This was observed in our experimental study using the stent-graft with the lower radial force. As such, the use of stent-grafts with higher radial force would appear to be preferable to provide superior counterpressure. Enlargement of the initial puncture hole is a critical step in this technique. Balloon advancement to enlarge the hole through the fenestration is straightforward and successful. After graft fenestration, the possibility of propagation of the fabric tear is of greatest concern . This could result in a type III endoleak, especially during the balloon dilation of the initial puncture hole or during deployment of the balloon-expandable stent-graft. Our study demonstrated that the sealing of retrograde fenestrations was achieved in all explanted stent-grafts. Careful dilation of the balloon (4-mm diameter) and deployment of a moderately oversized balloonexpandable stent-graft (systematic oversizing of 1 mm) after fenestration produced no tearing of the graft material, and the gross integrity of the fabric was always maintained. In our study, macroscopic examination of the stent-grafts after explantation did not identify any effect of the endograft material (woven polyester versus expanded polytetrafluoroethylene) on the quality of fenestrations. Further work is being undertaken to characterize cutting balloon versus conventional angioplasty balloons, as well as the long-term fatigue characteristics of the textile graft materials [8]. Riga and associates [9]
2089
reported that needle puncture angles had an impact on the quality of the graft puncture and that the use of cutting balloons resulted in significantly more fabric tears and poor-quality fenestrations in all graft types. Improvements in puncture devices will probably accelerate the development of this approach. The use of energy devices (radiofrequency and laser) has the potential to decrease the mechanical force required for puncture and to facilitate graft perforation even in tortuous anatomy. Recently, Redlinger and colleagues [10] reported the outcomes of 22 patients with distal aortic arch aneurysms (zone 2) who underwent thoracic endovascular aortic repair with left subclavian artery revascularization with laser graft fenestration. Technical success was achieved in all cases, with no deaths and no major fenestration-related complications. Follow-up computed tomographic/angiographic imaging demonstrated 100% primary patency for the left subclavian artery stents. The durability of repair will also have to be ensured. Indeed, the long-term interactions between the stentgraft and the covered stent will need to be monitored closely because of the potential for stent collapse or stent breakage or the development of a late type III endoleak, or both, between the 2 components. Temporary brain perfusion is required during complete endovascular arch reconstruction by in situ retrograde fenestration. This can be achieved with the use of temporary extraanatomic brain perfusion: femoral artery–to–carotid artery bypass. In 1984, Walterbusch and associates [11] reported a case of restoration of carotid blood flow by femorocarotid bypass in acute aortic dissection before ascending aorta reconstruction. In 2008, Sch€ onholz and coworkers [12] described a percutaneous external shunt to restore carotid flow in a patient with acute type A aortic dissection and carotid occlusion. We have previously reported [13] the use of this temporary extraanatomic brain perfusion followed by total rerouting of the supraaortic vessels for hybrid repair of a ruptured aortic arch aneurysm with mediastinal hematoma underlying the sternum. The use of extracorporeal femoral artery–to–carotid artery perfusion using a standard cardiopulmonary bypass circuit can also be considered [14]. Although it was possible to penetrate and deploy the stents within 10 minutes in cadaveric arches, in vivo it is unlikely that the procedure could be performed as rapidly, and therefore temporary extraanatomic brain perfusion is likely to be mandatory. To decrease carotid territory ischemic time, it would be of interest to fenestrate the aortic graft before insertion. However, the aortic arch presents complex spatial geometry with curves and 3-dimensional angulations. Even with the current tremendous advances in imaging, it is not possible at present to fashion accurate fenestrations before aortic graft insertion in the acute setting. Preclinical testing has limitations regarding the ability to predict clinical failures, in part because of constraints inherent in replicating in vivo conditions. The use of cadaveric aortas rather than a nonbiological aortic model is relatively time-consuming and expensive. However,
ADULT CARDIAC
Ann Thorac Surg 2014;98:2086–91
2090
CANAUD ET AL COMPLETE ENDOVASCULAR AORTIC ARCH RECONSTRUCTION
ADULT CARDIAC
plastic tubing and other substrates bear little resemblance to basic tissue properties, such as compliance, of the human aorta. One weakness of our model is that stentgrafts were deployed within nonaneurysmal aortas. Thus, the “seal” zone for the aortic graft is the entire length of the graft, and therefore there is no room for a type I endoleak to appear. Additionally, retrograde fenestration would be more difficult in the setting of an aneurysm in which major portions of the stent-graft may not be immediately adjacent to the aortic wall. Therefore, the model used more closely parallels traumatic aortic injury or an acute type B dissection rather than an aneurysm. In conclusion, this study confirms the feasibility and reproducibility of performing complete endovascular arch reconstruction by in situ retrograde fenestration. This approach would produce more accurate fenestrations, with less reliance on preoperative imaging. It could potentially broaden availability and decrease costs through the use of noncustomized stent-grafts. Furthermore, retrograde perforation avoids the difficulties of side branch catheterization and the high inherent risk of cerebral embolism.
Ann Thorac Surg 2014;98:2086–91
4.
5. 6.
7. 8. 9.
10.
11. 12.
References 1. Estrera AL, Miller CC 3rd, Madisetty J, et al. Ascending and transverse aortic arch repair: the impact of glomerular filtration rate on mortality. Ann Surg 2008;247:524–9. 2. Canaud L, Hireche K, Berthet JP, Branchereau P, Marty-An e C, Alric P. Endovascular repair of aortic arch lesions in high-risk patients or after previous aortic surgery: midterm results. J Thorac Cardiovasc Surg 2010;140: 52–8. 3. Abraham CZ, Lioupis C. Treatment of aortic arch aneurysms with a modular transfemoral multibranched
13.
14.
stent-graft: initial experience. J Thorac Cardiovasc Surg 2013;145:110–7. Canaud L, Alric P, Laurent M, et al. Proximal fixation of thoracic stent-grafts as a function of oversizing and increasing aortic arch angulation in human cadaveric aortas. J Endovasc Ther 2008;15:326–34. Greenberg RK, Lytle B. Endovascular repair of thoracoabdominal aneurysms. Circulation 2008;117:2288–96. McWilliams RG, Fearn SJ, Harris PL, Hartley D, Semmens JB, Lawrence-Brown MM. Retrograde fenestration of endoluminal grafts from target vessels: feasibility, technique, and potential usage. J Endovasc Ther 2003;10:946–52. Sonesson B, Resch T, Allers M, Malina M. Endovascular total aortic arch replacement by in situ stent graft fenestration technique. J Vasc Surg 2009;49:1589–91. Saari P, Manninen H. Fenestration of aortic stent grafts-in vitro tests using various device combinations. J Vasc Interv Radiol 2011;22:89–94. Riga CV, Bicknell CD, Basra M, Hamady M, Cheshire NJ. In vitro fenestration of aortic stent-grafts: implications of puncture methods for in situ fenestration durability. J Endovasc Ther 2013;20:536–43. Redlinger RE Jr, Ahanchi SS, Panneton JM. In situ laser fenestration during emergent thoracic endovascular aortic repair is an effective method for left subclavian artery revascularization. J Vasc Surg 2013;58:1171–7. Walterbusch G, Oelert H, Borst HG. Restoration of cerebral blood flow by extraanatomic bypass in acute aortic dissection. Thorac Cardiovasc Surg 1984;32:381–2. Sch€ onholz C, Ikonomidis JS, Hannegan C, Mendaro E. Bailout percutaneous external shunt to restore carotid flow in a patient with acute type A aortic dissection and carotid occlusion. J Endovasc Ther 2008;15:639–42. Joyeux F, Canaud L, Hireche K, Berthet JP, Marty-Ane C, Alric P. Temporary extra-anatomic brain perfusion followed by total rerouting of the supra-aortic vessels for hybrid repair of a ruptured aortic arch aneurysm. J Vasc Surg 2011;54: 1145–7. Lownie SP, Menkis AH, Craen RA, Mezon B, MacDonald J, Steinman DA. Extracorporeal femoral to carotid artery perfusion in selective brain cooling for a giant aneurysm. Case report. J Neurosurg 2004;100:343–7.
INVITED COMMENTARY For about 60 years, surgical procedures on the aortic arch have been performed by different approaches with varied outcomes. The risks of death and stroke range from 1% to 12% and from 1% to 15%, respectively. Open surgical arch replacement continues to be done with cooling and circulatory arrest, but the optimum technique, the temperature, and the role of adjuvant brain perfusion remain debatable issues. Furthermore, the results of either arch debranching and stenting or custom-branched grafts have generally not been better compared with classic open operations; indeed, they are often worse. The predictors of outcome, however, that are certain include urgency of operation, rupture of the arch, older age, and cardiopulmonary bypass time. In this report, Canaud and colleagues [1] have built on their own and other authors’ experiences to develop a cadaver model for replacing the aortic arch. They have shown that, given complex angulation and size changes in the aortic arch, it is still feasible to place an aortic arch stent (with likely cardiopulmonary bypass) across the Ó 2014 by The Society of Thoracic Surgeons Published by Elsevier
arch, and they insert retrograde bridging stent grafts from the greater arch arteries back into the aortic arch stent graft. This novel approach should be particularly useful in those patients at greatest risk of adverse events, namely, patients who are elderly, require urgent operations, have aortic arch ruptures, or have some combination of these conditions. The authors used two currently approved aortic devices along with commercially available needles, wires, balloons, and a balloon expandable covered stent for the branches. As they have pointed out, one advantage of these devices is that they are all readily available “off the shelf.” With their model, they have demonstrated technical feasibility and no endoleaks on initial fluoroscopy. They are to be congratulated for creating a fine pulsatile model in fresh human aortas to mimic performance of the procedure. Of course, additional testing of this approach will be needed before its wider application in patients. For example, the more proximal aorta, arch, and arch branches are subject to great continuous hemodynamic
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2014.08.004
Background. The aim of this experimental study was to assess the feasibility of complete endovascular arch reconstruction by in situ retrograde fenestration and to investigate the impact of stent-graft material on stent-graft fenestrations. Methods. The experiments were performed using 8 cadaveric human thoracic aortas (aortic arch) using 2 different stent-graft types: woven polyester (Valiant Captivia; Medtronic Vascular, Santa Rosa, CA) and expanded polytetrafluoroethylene (conformable [C]-TAG; W.L. Gore & Associates, Flagstaff, AZ). A benchtop aortic pulsatile flow model was used. Stent-grafts were deployed into the aortic arch, covering the ostia of the supraaortic trunks. A 5-mm 30-degree angioscope was introduced into the ascending aorta to monitor the procedure. Retrograde fenestration and deployment of the balloon expandable stent-graft was performed sequentially for each supraaortic trunk. Subsequent to stent-graft
explantation, macroscopic evaluation of each fenestration was performed. Results. All attempts to fenestrate the C-TAG and Valiant stent-grafts and implant the covered stent through the supraaortic trunks were successful. In all cases, branch stents were patent and no endoleak was evident. The Valiant stent-graft was easier to puncture because of the higher radial force of the stent-graft providing better counterpressure; however, stent-graft material had no impact on the quality of fenestrations. Conclusions. Total endovascular repair of the aortic arch through in situ retrograde fenestration of stent-grafts is feasible. The behavior of the 2 types of stent-graft was significantly different while the fenestrations were fashioned, but stent-graft material had no impact on the quality of fenestrations.
D
conditions of the aortic arch, the majority of these adjunctive procedures remain major operations and are associated with significant perioperative mortality. Branched stent-grafts that permit completely percutaneous aortic arch repair have been proposed [3]. The disadvantages of this modular approach include the time required time to manufacture and deliver custom-made stent-grafts for urgent cases and the high costs associated these sorts of modular devices. Most notably, there is a high rate of embolism associated with this approach, which is probably related to the complexity of multibranched unibody stent-graft deployment. The concept of retrograde in situ fenestration combines intentional stent-graft coverage of the supraaortic trunks with subsequent reestablishment of blood flow by retrograde puncture of the device. This approach is an attractive alternative that eliminates the need for preoperative custom tailoring (allowing repair of emergent cases) and the inherent risk of cerebral embolism associated with catheterization of difficulties side branches. The aim of this experimental study was to assess the feasibility of complete endovascular arch reconstruction by in situ retrograde fenestration and to investigate whether the quality of stent-graft fenestrations varied with stent-graft material.
eep hypothermic circulatory arrest is required for cerebral protection during surgical repair of the aortic arch with replacement [1]. Mortality and morbidity associated with transverse aortic arch replacement in the standard-risk population has decreased over the past few decades with the implementation of various modifications of surgical technique. Despite these recent advances, aortic arch reconstruction remains challenging, particularly in elderly patients, those requiring emergency repair, and those with major preexisting comorbidities. Many patients are deemed unsuitable for open repair. More recently a combined endovascular and open approach has been adopted as a valuable alternative, consisting of supraaortic debranching and revascularization followed by stent-graft deployment [2]. Debranching is performed to provide an appropriate landing zone for the stent-graft while preserving perfusion to the supraaortic trunks. Although this approach provides an attractive alternative for the treatment of pathologic
Accepted for publication July 9, 2014. Address correspondence to Dr Canaud, Service de Chirurgie Vasculaire et Thoracique, Hopital A de Villeneuve, 191 av Doyen Gaston Giraud 34090, Montpellier, France; e-mail: [email protected]. ˇ
ADULT CARDIAC
Experimental Evaluation of Complete Endovascular Arch Reconstruction by In Situ Retrograde Fenestration
Ó 2014 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2014;98:2086–91) Ó 2014 by The Society of Thoracic Surgeons
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2014.07.024
Ann Thorac Surg 2014;98:2086–91
CANAUD ET AL COMPLETE ENDOVASCULAR AORTIC ARCH RECONSTRUCTION
2087
Material and Methods
ADULT CARDIAC
The study was approved by our institutional review committee. Consent was obtained from the relatives of each cadaveric aorta donor.
Harvesting and Preparation of Aortas In accordance with French regulation, 9 fresh nonaneurysmal human aortas were harvested at autopsy from men and women who had died up to 4 days previously. The aortas were harvested from 2 cm above the level of the aortic valve proximally to the iliac bifurcation distally. The brachiocephalic trunk, the left common carotid artery, and the left subclavian artery were harvested to maximal length. Each aorta was immediately packed in ice and maintained at 4 C. Experiments were performed within 2 hours of harvest. Aortic sections were sent to the Department of Pathology for microscopic evaluation (staining with hematoxylin and eosin) to confirm the presence of 3 layers of aortic wall comparable to fresh specimens.
Bench Test Model A previously described benchtop closed-system pulsatile flow model was used to mimic aortic flow and pressure as close to normal physiologic conditions as possible [4].
Fig 1. Benchtop closed-system pulsatile flow model. A graft is anastomosed to the distal segments of the aortic arch side branches. To monitor the procedure, a 5-mm angioscope connected to a camera was introduced into the ascending aorta.
Experimental Setup After harvest of the entire aorta, an 8-mm knitted polyester graft was anastomosed to the distal part of each supraaortic trunk. The distal end of the graft was then connected to a closed circuit to simulate antegrade flow into the aortic branch vessels during the experiment. Abdominal aortic branches were ligated 1 cm distal to their origin, and intercostal and lumbar arteries were oversewn. Each aorta was then coupled to the closed-system pulsatile flow model. Once each aorta was incorporated into the circulatory circuit, the pump was activated, leading to pulsatile flow: 60 pulses/min and pressure of 150/80 mm Hg. MONITORING THE PROCEDURE. To monitor the procedure, a 5-mm 30-degree angioscope (Richard Wolf, Vernon Hills, IL,) connected to a video camera was introduced into the ascending aorta through a simple 3-0 pursestring Vicryl suture (Ethicon, Somerville, NJ) (Fig 1). STENT-GRAFTS. Two types of commercial stent-grafts of 100-mm length (Fig 2) were used: the Valiant Captivia stent-graft (Medtronic Vascular, Santa Rosa, CA) and the conformable TAG (C-TAG) stent-graft (WL Gore & Associates, Flagstaff, AZ). The Valiant is composed of a nitinol stent framework between layers of polyester graft. Individual stents are sutured to the outside of the polyester graft material. The proximal end features an open bare stent segment. The C-TAG is composed of a symmetrically expanded polytetrafluoroethylene tube reinforced externally with a layer of expanded polytetrafluoroethylene. An exoskeleton consisting of nitinol stents is attached to cover the length of the graft. PREPARATION OF AORTA.
STENT-GRAFT PLACEMENT. Stent-grafts were deployed into the aortic arch, covering the ostia of the supraaortic trunks. Delivery of the covered stent was performed in a retrograde fashion. RETROGRADE FENESTRATION. Retrograde fenestration and deployment of the balloon expandable stent-graft was performed sequentially for each of the supraaortic trunks (brachiocephalic trunk, left common carotid artery, and subclavian artery) using the same technique. The puncture of the stent-graft was performed using a 20-gauge needle, and a 0.035-inch guide wire was advanced through the aperture into the descending aorta. The graft puncture site was subsequently dilated by advancing a 5F introducer sheath with its dilator tip over the 0.035inch guide wire. This was followed by dilation of the fenestration with a standard 4-mm angioplasty balloon (Wanda; Boston Scientific, Natick, MA). Thereafter, a balloon-expandable covered stent (Atrium Medical Corp, Hudson, NH) that was 22-mm in length (the diameter of which was selected by referencing the diameter of the supraaortic vessel treated; systematic oversizing of 1 mm), was passed and positioned across the fenestration. The balloon-expandable stent-graft was then deployed 5 mm into the aorta. Proximal flaring of the covered stent was initially performed using a noncompliant balloon (Coda; Cook Medical, Bloomington, IL) introduced through the supraaortic vessel. Final flaring of the covered stent was then performed using the same noncompliant balloon introduced through the abdominal aorta.
2088
CANAUD ET AL COMPLETE ENDOVASCULAR AORTIC ARCH RECONSTRUCTION
Ann Thorac Surg 2014;98:2086–91
ADULT CARDIAC
Fig 2. (A) Internal and (B) external views after stent-graft explantation after complete endovascular arch reconstruction by in situ retrograde fenestration. The image confirms secure fixation and sealing of the retrograde fenestration.
STENT-GRAFT EXPLANTATION. After completion of the endovascular arch reconstruction by in situ retrograde fenestration, each stent-graft was explanted. Macroscopic qualitative evaluation of each fenestration was performed.
Results Harvesting and Preparation of Aortas Eight fresh human nonaneurysmal aortas were harvested from 6 men and 2 women (mean age, 38.4 years; range, 29–51 years) who had died within a maximum of 4 days (mean, 1.2 days). The length of the brachiocephalic trunk, the left common carotid artery, and the left subclavian artery varied from 4 to 8 cm. The mean aortic diameter measured at the level of the left ascending aorta level was 26.5 2.2 mm (range, 24–29 mm) before explantation. Histologic analysis of the harvested aortas at the start of the study revealed that all 8 aortas consisted of 3 layers and were comparable to fresh aortas.
Comment An anatomically unsuitable proximal neck is the most common reason for patient exclusion from endovascular aortic repair. To maintain aortic side branch patency, thoracic stent-graft modifications have been proposed, featuring scallops, fenestrations, and branches. Fenestrated and branched stent-grafts have shown promising midterm results in selected patients [5]. All these systems are customized, requiring accurate preoperative planning. Target-vessel cannulation can be technically challenging and time-consuming even for experienced operators, especially in the presence of complex tortuous anatomy. Despite the custom-made nature of these devices and extensive preoperative planning, graft rotation and misalignment of the fenestration/vessel ostium interface can still occur. In addition, the aorta may change configuration after insertion of the semirigid stent-graft and thus alter alignment of the side branches. Furthermore, factors such as the delay in device planning
Initial Retrograde Fenestration Procedure Complete endovascular arch reconstruction by in situ retrograde fenestration was achieved in all cases (Fig 2). The mean procedure time was 10 2 minutes. A total of 24 balloon-expandable covered stents were successfully inserted. The Valiant stent-graft was easier to puncture because the higher radial force of the stent-graft provided a better counterbrace. Completion angioscopic assessment demonstrated the patency of stent-graft branches and no type III endoleak in all cases.
Stent-Graft Explantation Stent-graft explantation confirmed that secure fixation and seal of retrograde fenestrations was achieved in all cases. Balloon dilation and deployment of the balloonexpandable stent-grafts after fenestration was not associated with tears of the graft material, and the gross integrity of the fabric was always maintained. Endograft material (woven polyester versus expanded polytetrafluoroethylene) had no impact on the quality of fenestrations (Fig 3).
Fig 3. No tearing of the graft material was observed, and the gross integrity of the fabric was always maintained. Endograft material had no impact on the quality of fenestrations: (A) expanded polytetrafluoroethylene versus (B) woven polyester.
CANAUD ET AL COMPLETE ENDOVASCULAR AORTIC ARCH RECONSTRUCTION
and manufacturing, anatomic and technical limitations, and expense limit the widespread uptake of this technology, which additionally is unsuitable for emergent cases. Above all, the technical difficulty of side branch catheterization results in an inherently high risk of cerebral embolism. The use of readily available “off the shelf” endovascular materials would decrease the cost of endovascular repair and increase the availability of these minimally invasive techniques to a greater number of patients. Thus retrograde in situ fenestration seems to offer an appropriate solution to most of these issues, allowing more accurate fenestration placement with less reliance on preoperative imaging, as well as a reduction in the number of catheter and guide wire manipulations in the aortic arch. McWilliams and colleagues [6] were the first to report successful retrograde in situ fenestration in bench and animal models, followed by clinical application in the left subclavian artery. Sonesson and coworkers [7] used this approach in the emergency clinical setting to treat an acute aortic arch rupture. These clinical reports used widely varying techniques to fashion fenestrations, and the reproducibility of this approach, especially in an experimental setting, has to be demonstrated before more widespread clinical application. Our cadaveric model demonstrates the feasibility of complete aortic arch reconstruction using the retrograde approach. This approach, which was successful in all the models using “universal” noncustomized stent-grafts, render it suitable for urgent and emergency cases. A potential disadvantage of the retrograde technique is that the stent-graft that lies against the orifice might be pushed away during mechanical puncture. This was observed in our experimental study using the stent-graft with the lower radial force. As such, the use of stent-grafts with higher radial force would appear to be preferable to provide superior counterpressure. Enlargement of the initial puncture hole is a critical step in this technique. Balloon advancement to enlarge the hole through the fenestration is straightforward and successful. After graft fenestration, the possibility of propagation of the fabric tear is of greatest concern . This could result in a type III endoleak, especially during the balloon dilation of the initial puncture hole or during deployment of the balloon-expandable stent-graft. Our study demonstrated that the sealing of retrograde fenestrations was achieved in all explanted stent-grafts. Careful dilation of the balloon (4-mm diameter) and deployment of a moderately oversized balloonexpandable stent-graft (systematic oversizing of 1 mm) after fenestration produced no tearing of the graft material, and the gross integrity of the fabric was always maintained. In our study, macroscopic examination of the stent-grafts after explantation did not identify any effect of the endograft material (woven polyester versus expanded polytetrafluoroethylene) on the quality of fenestrations. Further work is being undertaken to characterize cutting balloon versus conventional angioplasty balloons, as well as the long-term fatigue characteristics of the textile graft materials [8]. Riga and associates [9]
2089
reported that needle puncture angles had an impact on the quality of the graft puncture and that the use of cutting balloons resulted in significantly more fabric tears and poor-quality fenestrations in all graft types. Improvements in puncture devices will probably accelerate the development of this approach. The use of energy devices (radiofrequency and laser) has the potential to decrease the mechanical force required for puncture and to facilitate graft perforation even in tortuous anatomy. Recently, Redlinger and colleagues [10] reported the outcomes of 22 patients with distal aortic arch aneurysms (zone 2) who underwent thoracic endovascular aortic repair with left subclavian artery revascularization with laser graft fenestration. Technical success was achieved in all cases, with no deaths and no major fenestration-related complications. Follow-up computed tomographic/angiographic imaging demonstrated 100% primary patency for the left subclavian artery stents. The durability of repair will also have to be ensured. Indeed, the long-term interactions between the stentgraft and the covered stent will need to be monitored closely because of the potential for stent collapse or stent breakage or the development of a late type III endoleak, or both, between the 2 components. Temporary brain perfusion is required during complete endovascular arch reconstruction by in situ retrograde fenestration. This can be achieved with the use of temporary extraanatomic brain perfusion: femoral artery–to–carotid artery bypass. In 1984, Walterbusch and associates [11] reported a case of restoration of carotid blood flow by femorocarotid bypass in acute aortic dissection before ascending aorta reconstruction. In 2008, Sch€ onholz and coworkers [12] described a percutaneous external shunt to restore carotid flow in a patient with acute type A aortic dissection and carotid occlusion. We have previously reported [13] the use of this temporary extraanatomic brain perfusion followed by total rerouting of the supraaortic vessels for hybrid repair of a ruptured aortic arch aneurysm with mediastinal hematoma underlying the sternum. The use of extracorporeal femoral artery–to–carotid artery perfusion using a standard cardiopulmonary bypass circuit can also be considered [14]. Although it was possible to penetrate and deploy the stents within 10 minutes in cadaveric arches, in vivo it is unlikely that the procedure could be performed as rapidly, and therefore temporary extraanatomic brain perfusion is likely to be mandatory. To decrease carotid territory ischemic time, it would be of interest to fenestrate the aortic graft before insertion. However, the aortic arch presents complex spatial geometry with curves and 3-dimensional angulations. Even with the current tremendous advances in imaging, it is not possible at present to fashion accurate fenestrations before aortic graft insertion in the acute setting. Preclinical testing has limitations regarding the ability to predict clinical failures, in part because of constraints inherent in replicating in vivo conditions. The use of cadaveric aortas rather than a nonbiological aortic model is relatively time-consuming and expensive. However,
ADULT CARDIAC
Ann Thorac Surg 2014;98:2086–91
2090
CANAUD ET AL COMPLETE ENDOVASCULAR AORTIC ARCH RECONSTRUCTION
ADULT CARDIAC
plastic tubing and other substrates bear little resemblance to basic tissue properties, such as compliance, of the human aorta. One weakness of our model is that stentgrafts were deployed within nonaneurysmal aortas. Thus, the “seal” zone for the aortic graft is the entire length of the graft, and therefore there is no room for a type I endoleak to appear. Additionally, retrograde fenestration would be more difficult in the setting of an aneurysm in which major portions of the stent-graft may not be immediately adjacent to the aortic wall. Therefore, the model used more closely parallels traumatic aortic injury or an acute type B dissection rather than an aneurysm. In conclusion, this study confirms the feasibility and reproducibility of performing complete endovascular arch reconstruction by in situ retrograde fenestration. This approach would produce more accurate fenestrations, with less reliance on preoperative imaging. It could potentially broaden availability and decrease costs through the use of noncustomized stent-grafts. Furthermore, retrograde perforation avoids the difficulties of side branch catheterization and the high inherent risk of cerebral embolism.
Ann Thorac Surg 2014;98:2086–91
4.
5. 6.
7. 8. 9.
10.
11. 12.
References 1. Estrera AL, Miller CC 3rd, Madisetty J, et al. Ascending and transverse aortic arch repair: the impact of glomerular filtration rate on mortality. Ann Surg 2008;247:524–9. 2. Canaud L, Hireche K, Berthet JP, Branchereau P, Marty-An e C, Alric P. Endovascular repair of aortic arch lesions in high-risk patients or after previous aortic surgery: midterm results. J Thorac Cardiovasc Surg 2010;140: 52–8. 3. Abraham CZ, Lioupis C. Treatment of aortic arch aneurysms with a modular transfemoral multibranched
13.
14.
stent-graft: initial experience. J Thorac Cardiovasc Surg 2013;145:110–7. Canaud L, Alric P, Laurent M, et al. Proximal fixation of thoracic stent-grafts as a function of oversizing and increasing aortic arch angulation in human cadaveric aortas. J Endovasc Ther 2008;15:326–34. Greenberg RK, Lytle B. Endovascular repair of thoracoabdominal aneurysms. Circulation 2008;117:2288–96. McWilliams RG, Fearn SJ, Harris PL, Hartley D, Semmens JB, Lawrence-Brown MM. Retrograde fenestration of endoluminal grafts from target vessels: feasibility, technique, and potential usage. J Endovasc Ther 2003;10:946–52. Sonesson B, Resch T, Allers M, Malina M. Endovascular total aortic arch replacement by in situ stent graft fenestration technique. J Vasc Surg 2009;49:1589–91. Saari P, Manninen H. Fenestration of aortic stent grafts-in vitro tests using various device combinations. J Vasc Interv Radiol 2011;22:89–94. Riga CV, Bicknell CD, Basra M, Hamady M, Cheshire NJ. In vitro fenestration of aortic stent-grafts: implications of puncture methods for in situ fenestration durability. J Endovasc Ther 2013;20:536–43. Redlinger RE Jr, Ahanchi SS, Panneton JM. In situ laser fenestration during emergent thoracic endovascular aortic repair is an effective method for left subclavian artery revascularization. J Vasc Surg 2013;58:1171–7. Walterbusch G, Oelert H, Borst HG. Restoration of cerebral blood flow by extraanatomic bypass in acute aortic dissection. Thorac Cardiovasc Surg 1984;32:381–2. Sch€ onholz C, Ikonomidis JS, Hannegan C, Mendaro E. Bailout percutaneous external shunt to restore carotid flow in a patient with acute type A aortic dissection and carotid occlusion. J Endovasc Ther 2008;15:639–42. Joyeux F, Canaud L, Hireche K, Berthet JP, Marty-Ane C, Alric P. Temporary extra-anatomic brain perfusion followed by total rerouting of the supra-aortic vessels for hybrid repair of a ruptured aortic arch aneurysm. J Vasc Surg 2011;54: 1145–7. Lownie SP, Menkis AH, Craen RA, Mezon B, MacDonald J, Steinman DA. Extracorporeal femoral to carotid artery perfusion in selective brain cooling for a giant aneurysm. Case report. J Neurosurg 2004;100:343–7.
INVITED COMMENTARY For about 60 years, surgical procedures on the aortic arch have been performed by different approaches with varied outcomes. The risks of death and stroke range from 1% to 12% and from 1% to 15%, respectively. Open surgical arch replacement continues to be done with cooling and circulatory arrest, but the optimum technique, the temperature, and the role of adjuvant brain perfusion remain debatable issues. Furthermore, the results of either arch debranching and stenting or custom-branched grafts have generally not been better compared with classic open operations; indeed, they are often worse. The predictors of outcome, however, that are certain include urgency of operation, rupture of the arch, older age, and cardiopulmonary bypass time. In this report, Canaud and colleagues [1] have built on their own and other authors’ experiences to develop a cadaver model for replacing the aortic arch. They have shown that, given complex angulation and size changes in the aortic arch, it is still feasible to place an aortic arch stent (with likely cardiopulmonary bypass) across the Ó 2014 by The Society of Thoracic Surgeons Published by Elsevier
arch, and they insert retrograde bridging stent grafts from the greater arch arteries back into the aortic arch stent graft. This novel approach should be particularly useful in those patients at greatest risk of adverse events, namely, patients who are elderly, require urgent operations, have aortic arch ruptures, or have some combination of these conditions. The authors used two currently approved aortic devices along with commercially available needles, wires, balloons, and a balloon expandable covered stent for the branches. As they have pointed out, one advantage of these devices is that they are all readily available “off the shelf.” With their model, they have demonstrated technical feasibility and no endoleaks on initial fluoroscopy. They are to be congratulated for creating a fine pulsatile model in fresh human aortas to mimic performance of the procedure. Of course, additional testing of this approach will be needed before its wider application in patients. For example, the more proximal aorta, arch, and arch branches are subject to great continuous hemodynamic
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2014.08.004