Thoracic Endovascular Aortic Repair for Challenging Aortic Arch Diseases Using Fenestrated Stent Grafts From Zone 0

Transcription

1 ADULT CARDIAC Thoracic Endovascular Aortic Repair for Challenging Aortic Arch Diseases Using Fenestrated Stent Grafts From Zone 0 Yoshihiko Kurimoto, MD, PhD, Ryushi Maruyama, MD, PhD, Kousuke Ujihira, MD, Naritomo Nishioka, MD, Kousei Hasegawa, MD, Yutaka Iba, MD, Eiichiro Hatta, MD, PhD, Akira Yamada, MD, PhD, and Katsuhiko Nakanishi, MD, PhD Department of Cardiovascular Surgery, Teine Keijinkai Hospital, Sapporo, Japan Background. Although previous reports have described the repair of distal aortic arch aneurysms through debranching and chimney techniques, these methods invariably involve surgical management of the carotid artery. We report clinical results of thoracic endovascular aortic repair (TEVAR) using fenestrated stent grafts in the treatment of aortic arch aneurysms located less than 15 mm from the left common carotid artery. Methods. A semi-custom-made fenestrated stent graft designed to fit aortic arch tortuosity and preserve blood flow at least into the brachiocephalic and left common carotid arteries was placed from zone 0. Results. From 2007 through 2013, TEVAR from zone 0 was performed on 37 high-risk patients for open surgery (mean age 78.2 years). The mean length between the left common carotid artery and aortic aneurysm was 11.1 mm (range, 5 to 15 mm). The left subclavian artery was preserved for 26 patients (70.3%) through surgical reconstruction (n [ 19) and graft fenestration (n [ 7). The early mortality rate was 0%. Postoperative strokes and spinal cord ischemia occurred in 2 (5.4%) and 3 (8.1%) patients, respectively. Although type I endoleaks at discharge were noted in 12 (32.4%) patients, aneurysm enlargement was noted during follow-up in 6 (16.2%). Four patients (10.8%) underwent secondary interventions consisting of 3 coil embolization procedures; 2 re- TEVARs and 1 open conversion. There were no aortarelated late deaths. Survival and aorta-related event-free rates at 2 years were 86.3% and 88.8%, respectively. Conclusions. Thoracic endovascular aortic repair using fenestrated stent graft from zone 0 can be considered as one of therapeutic options for high-risk patients with aortic arch diseases. (Ann Thorac Surg 2015;100:24 33) Ó 2015 by The Society of Thoracic Surgeons Based on studies indicating encouraging early and mid-term results [1, 2], thoracic endovascular aortic repair (TEVAR) has been employed as a first line surgical treatment for descending thoracic aortic aneurysms. However, in these studies the proximal end of the first generation TAG device (W.L. Gore & Associates, Inc, Flagstaff, AZ) or Zenith TX2 device (Cook Medical, Inc, Bloomington, IN) did not sufficiently conform to the curvature of the distal aortic arch. Moreover, it was also observed that there was a lack of graft fixation against the lesser curvature of the aortic arch for aneurysms located in the distal aortic arch or proximal descending aorta. To overcome these problems, endovascular industries have recently developed improved versions of TAG and TX2, such as conformable TAG and TX2 Pro-form, Accepted for publication Jan 16, Presented at the Fiftieth Annual Meeting of The Society of Thoracic Surgeons, Orlando, FL, Jan 25 29, Address correspondence to Dr Kurimoto, Department of Cardiovascular Surgery, Teine Keijinkai Hospital, 1-40 Maeda 1 Jo 12 Chome, Teine-ku, Sapporo , Japan; kurimoto-spr@umin.ac.jp. which show enhanced conformability to the acutely angled distal aortic arch [3]. However, it remains difficult to further expand an indication of simple TEVAR for distal aortic arch lesions because additional management of supra-aortic branches, particularly the brachiocephalic (BCA) and the left common carotid artery (LCCA), is necessary to ensure sufficient length of a proximal neck. In Japan, the Najuta fenestrated stent graft (SG) (Kawasumi, Inc, Tokyo, Japan) has been commercially available for distal aortic arch aneurysms since June According to the instructions for use of this SG, a proximal neck length of more than 20 mm between the LCCA and the distal aortic arch aneurysm is recommended to achieve satisfactory clinical outcomes. In light of the improved results presently being observed through the use of other currently available non-fenestrated SGs, it is expected that the Najuta fenestrated SG should also be able to treat more challenging aortic arch pathologies. In this study, we report early and mid-term results of TEVAR using the fenestrated SG as a semi-custom handmade devise for aortic arch aneurysm, and describe its clinical usefulness and limitations. Ó 2015 by The Society of Thoracic Surgeons /$36.00 Published by Elsevier

2 Ann Thorac Surg KURIMOTO ET AL 2015;100:24 33 FENESTRATED STENT GRAFT FOR ARCH ANEURYSM 25 Fig 1. (A) Computed tomographic images regarding a proximal neck (PN) length. Although the PN length at the lesser curvature of the aortic arch is almost zero, this case was used in the present study because there is a PN length between the left common carotid artery and aortic arch aneurysm, a. The PN length at the greater and lesser curvature of the aortic arch is more than 15 mm. This type of aneurysm has been treated using a conventional non-fenestrated stent graft, which means that this case was not included in the present study, b. (B) Computed tomographic images after TEVAR. The case on the left, a, was treated using a fenestrated stent graft. The case on the right, b, which was not a subject in this study, was treated using a Valiant (Medtronic, Inc) modified to make a fenestration to preserve the left subclavian artery. ADULT CARDIAC Material and Methods Patient Selection From July 2007 through March 2013, patients with aortic arch aneurysms located less than 15-mm away from the LCCA (Figs 1A, 1B) and proximal and distal aortic neck diameter of less than 40 mm were considered possible candidates for TEVAR using a fenestrated SG. Of these patients, those whose logistic EuroSCORE-predicted risk of mortality was more than 10%, and who accepted the terms provided through informed consent regarding TEVAR using a fenestrated handmade SG were the subjects of this study. The present research has been reviewed and approved by the ethics committee in our institution. Although a patient s consent for this study was waived because of the retrospective one, all patients accepted the informed consent regarding TEVAR using a fenestrated handmade SG. Procedure Using a Fenestrated Stent Graft The fenestrated SG used in this study was a semi-custom and handmade device (referred to as the Yokoi HIJ graft) a prototype model of Najuta. The details of this type of SG and procedure have been previously reported [4]. Briefly, the pre-curved fenestrated SG was fabricated through selection from several types of three-dimensional curved stainless stent skeletons and a fenestrated expanded polytetrafluoroethylene graft (Fig 2) based on each patient s preoperative three-dimensional computed tomographic (CT) images. The common femoral artery was primarily used as the access vessel. A brachial-wire Fig 2. Illustration of a semi-custom-made fenestrated stent graft. The white arrow shows a fenestration made to preserve the left common carotid artery. Generally, both the second and proximal section of the third (white arrow) fenestrations preserve the brachiocephalic artery (BCA). A proximal fenestration was made to prevent the possibility of an unintentional distal migration of a stent graft covering BCA.

3 ADULT CARDIAC 26 KURIMOTO ET AL Ann Thorac Surg FENESTRATED STENT GRAFT FOR ARCH ANEURYSM 2015;100:24 33 Table 1. Patient Characteristics (n ¼ 37) Variable No. (%) Age (years) (range, 57 88) Male 29 (78.4) Hypertension 28 (75.7) Diabetes mellitus 8 (21.6) Dyslipidemia 17 (45.9) Smoker 7 (18.9) Cerebrovascular disease 8 (21.6) Coronary artery disease 11 (29.7) Ejection fraction < (5.4) CKD (creatinin >1.7 mg/dl) 9 (24.3) Previous cardiac surgery 9 (24.3) Previous TEVAR 4 (10.8) Previous AAA repair 7 (18.9) Logistic EuroSCORE (range, ) Etiology (%) Degenerative 31 (83.8) Chronic dissection 3 (8.1) Others 3 (8.1) Urgent (%) 1 (2.7) AAA ¼ abdominal aortic aneurysm; CKD ¼ chronic kidney disease; EuroSCORE ¼ European system for cardiac operative risk evaluation; TEVAR ¼ thoracic endovascular aortic repair. method was the preferred guiding technique used to stretch this pre-curved device (outer diameter 22 to 23 French) in the access route. The pre-curved fenestrated SG was deployed in the aortic arch under fluoroscopic guidance without additional circulatory support (ie, without adenosine-induced cardiac arrest or rapid pacing). Device rotation was automatically controlled by the pre-curved shape of the SG and the dedicated stabilizing mechanism. The proximal tip of the delivery system had a concavity for fixation, and the proximal end of the SG had a stainless steel hock. The proximal end of the SG was anchored to the tip, and the traction suture to shrink the proximal end was fixed at the distal end of the SG. The Table 2. Aneurysmal Configuration (n ¼ 37) Variable No. (%) Length from LCCA to TAA (mm) Length of proximal neck at great curvature (mm) Length of proximal neck at lesser curvature (mm) Radius of the aortic arch (mm) Aneurysmal size (mm) Aneurysmal length (mm) Aneurysmal location: Great curvature/lesser/anterior-great/ 4/2/4/13/14 Anterior-lesser/Great-lesser LCCA ¼ left common carotid artery; TAA ¼ thoracic aortic aneurysm. proximal end of the SG was shrunk by the traction suture, and pulled down to a lesser curvature. As the traction suture gradually straightened during the SG deployment, the proximal end of the SG gradually opened. Commercial endografts, TAG, Zenith TX2, and Valiant (Medtronic Inc, Santa Rosa, CA) were used as distal stent grafts when needed. All TEVARs in this study were performed by 1 endovascular surgeon. Management of the Left Subclavian Artery The left subclavian artery (LSA) in most cases was believed to be fully covered by the SG because of the pathology of aortic arch aneurysms. Except for emergency cases, preoperative anatomic evaluation of the vertebral-basilar artery including the circle of Willis patency was performed by three-dimensional CT or magnetic resonance angiography. The LSA revascularization was undertaken in selected patients. Selection criteria mandated that patients coronary circulation was supplied by the LSA through the left internal thoracic artery, their contralateral vertebral arteries exhibited inadequacies, and they were at high risk for spinal cord ischemia. When the left vertebral artery was not considered critical for cerebral perfusion, simple coverage of the the LSA by the SG was positively applied to the elderly Fig 3. A schema expressing the values used in Table 2. (Left) (1) Length of a proximal neck (PN) between the left common carotid artery and the aneurysm. (2) Length of the PN at the great curvature. (3) Length of the PN at the lesser curvature. (Right) (4) Radius of the lesser curvature at PN.

4 Ann Thorac Surg KURIMOTO ET AL 2015;100:24 33 FENESTRATED STENT GRAFT FOR ARCH ANEURYSM 27 Fig 4. (A) A schema showing the left subclavian artery (LSA) management techniques in this study. The LSA, located very close to the left common carotid artery (LCCA), was partially preserved in 7 cases by the same fenestration used to maintain blood flow into the LCCA. (B) A histogram outlining the distribution of a distal end of placed stent grafts. The 3 sections in black indicate the patients experiencing spinal cord ischemia complications. ADULT CARDIAC without risk factors for spinal cord ischemia. In situations in which anatomic evaluation could not determine the importance of LSA patency, an LSA balloon occlusion test was performed under local anesthesia (ie, patients were awake during the procedure) in the operating room prior to TEVAR. The details of the LSA occlusion test have been reported previously [5]. Management of the Spinal-Cord Protection Preoperative cerebrospinal fluid drainage was selectively used in a very limited number of patients who had previously undergone open or endovascular abdominal aortic aneurysm repair and whose descending thoracic aorta was entirely covered by the SG, and LSA and internal iliac artery were occluded or planned to be covered by the SG. Postoperatively, the cerebrospinal fluid drainage was selectively used at the onset of symptoms of anterior-spinal-cord ischemia. The drainage was set to a continuous pressure of 10 cm water to allow the cerebrospinal fluid to drain freely up to 30 ml/hour. Regardless of the drainage placement, mean arterial pressure was perioperatively maintained at greater than 90 mm Hg with the use of inotropic drugs when required. Transfusion was perioperatively considered for patients with a high risk of spinal cord ischemia when their blood hemoglobin levels were less than 10.0 mg/dl regardless of the amount of blood loss during surgery. Follow-Up and Statistical Analysis During the follow-up, patients received a CT scan with a contrast agent at 6 and 12 months after TEVAR, upon which a CT scan without a contrast agent was an option if patients aneurysms were not enlarged without endoleak. And then a CT scan was scheduled annually if the aneurysms were stable. Telephone clinical assessments and outpatient clinic records were abstracted. The mean follow-up was months (range, 1 to 63 months). All patients were able to have a follow-up after TEVAR. Late aortic

5 ADULT CARDIAC 28 KURIMOTO ET AL Ann Thorac Surg FENESTRATED STENT GRAFT FOR ARCH ANEURYSM 2015;100:24 33 Table 3. Early Results Variable No. (%) Thirty-day mortality (%) 0/37 (0) Success of head-vessel preservation 37/37 (100) through fenestrations (%) Perioperative stroke (%) 2/37 (5.4) Spinal cord ischemia (%) 3/37 (8.1) a Type I endoleak at discharge (%) 12/37 (32.4) b Other complications: AAA rupture 1 Renal & leg ischemia due to embolism 1 Iliac injury 1 a One patient who showed temporary paraparesis was included among these 3 patients. b All 12 patients with endoleak were discharged without any additional treatment. AAA ¼ abdominal aortic aneurysm. re-interventions were characterized according to whether they involved the previously treated segment of the aorta, including proximal and distal landing zones, or an unrelated segment such as the infrarenal aorta. Early outcomes were compared using standard univariate analysis including the Mann-Whitney U test for continuous data and Fisher exact test for categoric data. Late survivals were analyzed using the Kaplan-Meier survival technique. The level of significance was taken at a p value less than All calculations were performed using Statistical Package of Social Science (SPSS, Inc, Chicago, IL). Results A total of 37 patients (29 male, mean age years) with a mean aneurysm diameter of mm (range, 50 to 86 mm) and mean aneurysm length of mm (range, 40 to 303 mm) formed the study (Tables 1, 2). Mean length between the LCCA and aneurysm was mm (range, 5 to 15 mm) (Fig 3). Urgent repair was performed in 2.7% (1/37) of cases; one patient presented a rupture into the left pleural cavity. As seen in Table 1, etiologies consisted of degenerative aortic aneurysms in 31 patients (83.8%), chronic aortic dissections in 3 (8.1%), both degenerative aneurysms and dissections in 2 (5.6%), and a pseudoaneurysm in 1 (2.7%). All fenestrated SGs were placed from zone 0 and these fenestrations preserved antegrade blood flow into BCA and LCCA. Fenestrated stent grafts were not intended to preserve antegrade blood flow into the LSA. Among 30 cases in which LSA was completely covered by the SG, concomitant right axillo-left axillary artery bypass grafting with coil embolization of a LSA root was performed to preserve antegrade blood flow into the LSA in 19 (51.4%). Including 7 instances in which the LSA was partially preserved by a fenestration for LCCA due to their close proximity, the LSA was reconstructed in 70.3% (26 of 37) of the cases. The LSA was simply covered by the SG without any reconstruction in the remaining 11 cases (29.7%) (Fig 4A). A retrograde transfemoral approach was Table 4. Three cases of Spinal Cord Ischemia Complications After TEVAR Neurologic Status Late Result Iliac Artery Status TAA/AAA Surgery Length of Aneurysm LSA Status End of Distal Neck Age (year) Sex Case Delayed paraplegia a Late death at 13 months due to aspiration pneumonia 1 81 Male Th mm Reconstruction No history Rt. iliac occlusion 2 78 Male Th mm Covered No history Patent paraparesis Alive Late death at 5 months due to colon perforation Delayed paraparesis b (full recovery) Rt. Iliac occlusion 3 73 Male Th 6 60 mm Preservation Concomitant EVAR Covered (29.7%) TAA/AAA 10.8%/18.9% Male (78.4%) Median Th 8 This study a Paraplegia occurred 10 hours after TEVAR. b Paraparesis occurred 3 hours after TEVAR but recovered completely by 8 hours after onset. AAA ¼ abdominal aortic aneurysm; LSA ¼ left subclavian artery; TAA ¼ descending thoracic aortic aneurysm; TAA/AAA ¼ previous TAA or AAA operation; TEVAR ¼ thoracic endovascular aortic repair.

6 Ann Thorac Surg KURIMOTO ET AL 2015;100:24 33 FENESTRATED STENT GRAFT FOR ARCH ANEURYSM Table 5. Comparison Between the Two Groups With and Without Endoleaks Characteristics Endoleak (þ) Endoleak ( ) p Value Number of cases (%) 12 (32.4) 25 (67.6) Age (years) Female (%) 3 (25.0) 5 (20.0) Proximal stent graft diameter (mm) Length from LCCA to TAA (mm) Length of proximal neck at great curvature (mm) Length of proximal neck at lesser curvature (mm) Radius of the aortic arch (mm) Aneurysmal size (mm) Aneurysmal length (mm) Aneurysmal location: Great curvature/lesser/anterior-great/ Anterior-lesser/Great-lesser 1/0/2/3/6 3/2/2/10/ ADULT CARDIAC LCCA ¼ left common carotid artery; TAA ¼ thoracic aortic aneurysm. used in 35 cases. In 2 patients (5.4%) with a history of Y- shaped prosthetic vascular graft replacement for abdominal aortic aneurysms, a retroperitoneal trans-iliac vascular graft approach was used due to kinking of the Y-shaped graft leg. The mean number of SGs per case was (range, 1 to 4). The mean proximal device diameter was mm (range, 32 to 44 mm). A distal end of the placed SG was above T6 in 21.6% of the cases (8 of 37) and below T6 in 78.4% (Fig 4B). Early Outcomes The perioperative mortality rate was 0%. No aneurysm expansion or ruptures were detected perioperatively. The 30-day rates of death, stroke, and permanent paraplegia or paresis (including temporary) were 0%, 5.4%, and 5.4% (8.1), respectively (Tables 3, 4). The incidence of type Ia endoleaks at discharge was 32.4% (12 of 37). The average proximal neck length in the group with endoleaks was statistically shorter than the group without endoleaks (Table 5). Follow-Up and Mid-Term Outcomes The survival rate at 2 years after TEVAR was 86.3% 6.5% (Fig 5). The overall death rate was 16.2% (6 of 37). Six patients died during the follow-up due to aspiration pneumonia (2 patients), brain injury secondary to falling down (2), esophageal cancer (1), and panperitonitis secondary to sigmoid colon perforation (1). No aorta-related late death was detected. The incidence of endoleaks at follow-up (mean months, range 1 to 63 months) was the same as that at discharge (12 of 37, 32.4%). Among these patients, aneurysms in 6 (16.2%) were enlarged to more than 5 mm in size. Four patients underwent re-interventions, which consisted of coil embolization (3 patients), re-tevar (2), and open repair (1). Two patients required both re-tevar and a coil-embolization or open repair. We did not treat 2 patients due to their poor condition. Both patients died due to aspiration pneumonia, one at 13 months and the other at 27 months after TEVAR (Table 6). There was no incident of stroke, preserved head-vessel occlusion, or axillo-axillary bypass-graft occlusion during the followup periods. The event-free rate at 2 years after TEVAR was 88.8% 6.2% (Fig 5). Comment Minimally invasive endoluminal approaches to thoracic aortic pathology have received widespread acceptance. Expansion of TEVAR to pathology extending into the aortic arch often requires coverage of 1 or more supraaortic branches in order to obtain an adequate seal. Initially, open debranching procedures were applied to restore flow to supra-aortic arch vessels. Compared with conventional open repair, 1 benefit of the so-called debranching TEVAR for aortic arch aneurysms is that cardiopulmonary bypass and cerebral protection can be avoided. As this technique can be done using a commercially available SG and bypass grafting, which is a Fig 5. Survival and aorta-related event-free curves after thoracic endovascular aortic repair using a fenestrated stent graft. The black line and gray dotted line indicate overall survival rates and aortarelated event-free rates after discharge, respectively.

7 ADULT CARDIAC 30 KURIMOTO ET AL Ann Thorac Surg FENESTRATED STENT GRAFT FOR ARCH ANEURYSM 2015;100:24 33 Table 6. Thoracic Aorta-Related Events After TEVAR Using a Fenestrated Stent Graft Case Age (years) Sex Period After TEVAR Event Reason Treatment Result 1 84 Male 20 months TAA Enlargement LSA preservation Conservative Late death a 2 81 Male 13 months TAA Enlargement Short neck Conservative Late death a 3 57 Male 3/30 months TAA Enlargement PN dilatation re-tevar/open repair Alive 4 79 Male 10 months TAA Enlargement Short neck re-tevar & coil-emboli Alive 5 73 Female 7 months TAA Enlargement Short neck Coil-emboli Alive 6 88 Female 4 months TAA Enlargement Short neck Coil-emboli Alive a Patient died due to non-aorta-related reason (aspiration pneumonia). LSA ¼ left subclavian artery; PN ¼ proximal neck; TAA ¼ thoracic aortic aneurysm; TEVAR ¼ thoracic endovascular aortic repair. familiar procedure for vascular surgeons, debranching TEVAR has been widely accepted by many vascular surgeons and the feasibility of this procedure has been reported from many institutes [6 8]. While less invasive than traditional open repairs, hybrid surgical debranching procedures may still require sternotomy if BCA coverage is necessary. Moreover, even if sternotomy is not required surgical manipulation of the carotid artery cannot be avoided, which means that debranching TEVAR diminishes some of the benefits of endovascular surgery. Branched SGs may be an even less invasive alternative treatment for aortic arch aneurysms [9, 10]. Some endovascular industries are currently developing branched SGs but their use is still being investigated and they are not yet commercially available. One study has focused on simultaneous branch vessel stent placement through retrograde cannulation, or the so-called chimney technique, as a method of minimizing or eliminating the need for complex surgical reconstruction [11]. Technically, this procedural approach for treating aortic arch aneurysms appears to be similar to TEVAR using branched SGs. Chimney TEVAR is close to a totally endovascular treatment, which means it is theoretically less invasive. However, there are seemingly few aortic arch aneurysms that can be treated by SG placement or bare stents from the LCCA to the ascending aorta concomitant with TEVAR using currently available thoracic endografts. Namely, a gutter between the SG or bare stent placed through the LCCA and the main SG placed from the ascending aorta to the descending thoracic aorta usually causes a type Ia endoleak because the aortic arch aneurysm is located close to the LCCA. If a single chimney from the BCA is chosen for chimney TEVAR, the overall procedure becomes similar to debranching TEVAR because bypass grafting to the LCCA is necessary. In addition, the possible stroke incidence after the SG placement through the carotid artery is likely to be considerably high, as is stroke incidence after branched TEVAR, because of inevitable surgical management of the carotid artery [12, 13]. As mentioned above, in Japan a semi-custom-made precurved fenestrated stent graft (Najuta) has been commercially available since June Other than management of the LSA, this device has made it simpler to provide treatment for distal aortic arch aneurysms by TEVAR. Avoidance of surgical management of the LCCA may potentially reduce incidence of stroke during TEVAR, and SG placement from zone 0 can expand an indication of simple TEVAR for aortic arch aneurysms [4, 14]. According to the instructions for use, the length of the proximal neck from the LCCA should be more than 20 mm, similar to the length recommended for other currently available nonfenestrated SGs. However, as we mentioned in the Materials and Methods section, this device can make it possible to treat for aortic arch aneurysms located at the lesser Table 7. Early Results of Treatment for Aortic Arch Pathology Author Year Type of Repair Cases Early Mortality (%) Stroke (%) Survival at 1 Year (%) Di Eusanio et al [15] 2003 Open repair NP Okada et al [16] 2012 Open repair Iba et al [17] 2013 Open repair 1, Ferrero et al [6] 2012 Debranch TEVAR Clough et al [7] 2013 Debranch TEVAR NP Preventza et al [8] 2013 Debranch TEVAR Yang et al (Review) [13] 2012 Chimney TEVAR Our series 2015 Fenestrated TEVAR Chimney TEVAR ¼ TEVAR using chimney technique; Debranch TEVAR ¼ TEVAR after supraaortic trunks debranching; Fenestrated TEVAR ¼ TEVAR using a fenestrated stent graft; NP ¼ data not provided; TEVAR ¼ thoracic endovascular aortic repair.

8 Ann Thorac Surg KURIMOTO ET AL 2015;100:24 33 FENESTRATED STENT GRAFT FOR ARCH ANEURYSM 31 Fig 6. Computed tomographic images before and after thoracic endovascular aortic repair (TEVAR) for an extended thoracic aortic aneurysm. The (A) descending thoracic aortic aneurysm combined with the aortic arch aneurysm, that (C, D) ruptured into the left pleural cavity was treated by TEVAR using a fenestrated stent graft and conventional nonfenestrated stent grafts. The (B) brachiocephalic and left common carotid arteries were well preserved and the left subclavian artery was simply covered by a non-fenestrated stent graft without any reconstruction. ADULT CARDIAC curvature of the aortic arch that cannot be treated by a conventional non-fenestrated SG (Fig 1). In the present study we have reported clinical outcomes of simple TEVAR that challenged the pathology of aortic arch aneurysms located less than 15-mm away from the LCCA. Compared with previous reports of early mortality and stroke rates, fenestrated TEVAR seems to be safer than other surgical options (Table 7). On the other hand, our results also revealed a high incidence rate of type Ia endoleaks through a fenestration on the SG. A type Ia endoleak obviously increases the possibility of aortic aneurysm enlargement. After this challenging TEVAR, the aneurysm enlarged in 6 patients (16.2%) within 20 months. For the time being we will not expand beyond current indications of fenestrated TEVAR for challenging aortic arch aneurysms in patients whose logistic European system for cardiac operative risk evaluation-predicted risk of mortality was less than 10%, because clinical results of open repair for aortic arch aneurysms in the high-volume center have been promising [15 17]. In the endovascular treatment for distal aortic arch aneurysms, the SG has to be placed in a location more proximal to zone 2, which means the LSA is usually covered by the SG. The management of the LSA during TEVAR has been discussed for more than a decade. Physicians experienced with TEVAR have learned not only that LSA can be simply covered without revascularization in most cases [18] but also that simple coverage of LSA increases neurologic complications to some extent [19]. It is imperative to continue large volume statistical analysis regarding the influence of LSA coverage, but at present strategies of LSA management during TEVAR should be considered on an individual basis for each patient [20]. Endovascular aortic repair needs to be simple and less invasive. Toward that end a fenestrated stent graft might be useful toward the development of surgical strategies against extended thoracic aortic aneurysms (Fig 6). Based on our experience, many elderly have not complained of any inconveniences after a simple coverage of the LSA. The LSA surgical reconstruction rate of 51.4% in this study suggests that we have not ignored possible complications after simple coverage of the LSA. Based on our experience using the Najuta prototype, the high incidence rate of type Ia endoleak is seemingly due primarily to the length of each stainless stent; namely, there is an apparent lack of conformability. Stents of 25 mm in length are too long to fit the acutely angled lesser curvature of the aortic arch as a proximal neck or to seal a space around a fenestration. The LCCA fenestration size might have to be even smaller than that which we used in the present study. Further modification of the fenestrated stent graft should enhance its clinical results, making it a useful method for future treatment of aortic arch pathologies. Limitation of This Study There are a number of problematic issues in the current study. For example, there was no surgical control group and the series were conducted with small numbers of patients. Additionally, it is primarily a retrospective study, and all subjected TEVAR was performed by only 1 operator with extensive experience of endovascular aortic repair. Conclusion Thoracic endovascular aortic repair using a fenestrated SG from zone 0 can be considered as 1 of therapeutic options for high-risk patients with aortic arch pathologies. A fenestrated stent graft makes TEVAR simpler and safer,

9 ADULT CARDIAC 32 KURIMOTO ET AL Ann Thorac Surg FENESTRATED STENT GRAFT FOR ARCH ANEURYSM 2015;100:24 33 and might expand the possibilities for truly endovascular treatment of aortic arch pathologies if appropriate modifications can be done. The authors thank Dr Yoshihiko Yokoi for providing the handmade fenestrated stent grafts (Yokoi HIJ graft) used in this study. References 1. Bavaria JE, Appoo JJ, Makaroun MS, et al. Endovascular stent grafting versus open surgical repair of descending thoracic aortic aneurysm in low-risk patients: a multicenter comparative trial. J Thorac Cardiovasc Surg 2007;133: Desai ND, Burtch K, Moser W, et al. Long-term comparison of thoracic endovascular aortic repair (TEVAR) to open surgery for the treatment of thoracic aortic aneurysms. J Thorac Cardiovasc Surg 2012;144: Hsu HL, Chen CK, Chen PL, et al. The impact of bird-beak configuration on aortic remodeling of distal arch pathology after thoracic endovascular repair with the Zenith Pro- FRoem TX2 thoracic endograft. J Vasc Surg 2014;59: Yokoi Y, Azuma T, Yamazaki K. Advantage of a precurved fenestrated endograft for aortic arch disease: simplified arch aneurysm treatment in Japan 2010 and J Thorac Cardiovasc Surg 2013;145(3 Suppl):S Kurimoto Y, Ito T, Harada R, et al. Management of left subclavian artery in endovascular stent-grafting for distal aortic arch disease. Circ J 2008;72: Ferrero E, Ferri M, Viazzo A, et al. Is total debranching a safe procedure for extensive aortic-arch disease? A single experience of 27 cases. Eur J Cardiothorac Surg 2012;41: Clough RE, Lotfl S, Powell J, Lee A, Taylor PR. Hybrid aortic arch repair. Ann Cardiothorac Surg 2013;2: Preventza O, Bakaeen FG, Cervera RD, Coselli JS. Deployment of proximal thoracic endograft in zone 0 of the ascending aorta: treatment options and early outcomes for aortic arch aneurysm in a high-risk population. Eur J Cardiothorac Surg 2013;44: Inoue K, Sato M, Iwase T, et al. Clinical endovascular placement of branched graft for type B aortic dissection. J Thorac Cardiovasc Surg 1996;112: Chutter TA, Schneider DB, Reilly LM, Lobo EP, Messina LM. Modular branched stent graft for endovascular repair of aortic arch aneurysm and dissection. J Vasc Surg 2003;38: Ohrlander T, Sonesson B, Ivancev K, Resch T, Dias N, Malina M. The chimney graft: a technique for preserving or rescuing aortic branch vessels in stent-graft sealing zones. J Endovasc Ther 2008;15: Cires G, Noll RE Jr, Albuquerque FC Jr, Tonnessen BH, Sternbergh WC III. Endovascular debranching of the aortic arch during thoracic endograft repair. J Vasc Surg 2011;53: Yang J, Xiong J, Liu X, Jia X, Zhu Y, Guo W. Endovascular chimney technique of aortic arch pathologies: a systematic review. Ann Vasc Surg 2012;26: Kawaguchi S, Yokoi Y, Shimazaki T, Koide K, Matsumoto M, Shigematsu H. Thoracic endovascular aneurysm repair in Japan: experience with fenestrated stent grafts in the treatment of distal arch aneurysm. J Vasc Surg 2008;48(6 Suppl): 24S 9S. 15. Di Eusanio M, Wesselink RM, Morshuis WJ, Dossche KM, Schepens MA. Deep hypothermic circulatory arrest and antegrade selective cerebral perfusion during ascending aorta-hemiarch replacement: a retrospective comparative study. J Thorac Cardiovasc Surg 2003;125: Okada K, Omura A, Kano H, et al. Recent advancements of total aortic arch replacement. J Thorac Cardiovasc Surg 2012;144: Iba Y, Minatoya K, Matsuda H, et al. Contemporary open aortic arch repair with selective cerebral perfusion in the era of endovascular aortic repair. J Thorac Cardiovasc Surg 2013;145(3 Suppl):S Lee M, Lee do Y, Kim MD, et al. Selective coverage of the left subclavian artery without revascularization in patients with bilateral patent vertebrobasilar junctions during thoracic endovascular aortic repair. J Vasc Surg 2013;57: Holt PJ, Johnson C, Hinchliffe RJ, et al. Outcomes of the endovascular management of aortic arch aneurysm: implications for management of the left subclavian artery. J Vasc Surg 2010;51: Kurimoto Y, Kawaharada N, Ito T, et al. Less-invasive management of left subclavian artery in stent-grafting for distal aortic arch disease. Interact Cardiovasc Thorac Surg 2009;8: DISCUSSION DR MICHAEL FISCHBEIN (Stanford, CA): In your institution how do you decide between a total arch replacement versus an endovascular approach with the fenestrated graft? DR KURIMOTO: We perform the endovascular surgery for selected patients. In this study, subject patients had EuroSCORE [European system for cardiac operative risk evaluation] more than 10%. Like the patients in this study, high-risk patients are possible candidates of TEVAR [thoracic endovascular aortic repair] using a fenestrated stent graft. We generally select open repair for non-elderly patients with aortic arch aneurysm. DR ALI KHOYNEZHAD (Los Angeles, CA): I have a question about the high incidence of type Ia endoleaks in this cohort. We reviewed our experience with 184 TEVARs up to 2005, and published it in 2008 in JTCVS [Journal of Thoracic and Cardiovascular Surgery]. Type Ia endoleak was an indecent risk factor of mortality in multivariate analysis in this paper. The procedures you just described were zone 0 TEVARs. Is that correct? DR KURIMOTO: Yes. DR KHOYNEZHAD: In zone 0 cases, you have the entire ascending aorta, which is anywhere between 7 to 9 cm long, as your proximal landing zone. So how come, you felt you did not have adequate proximal landing zone and therefore type Ia endoleak? DR KURIMOTO: Actually, the proximal part of the stent graft on the ascending aorta is not the proximal sealing zone. Proximal part of the fenestrated stent graft usually works to stabilize the second part of the stent graft on proximal aortic arch and to reduce possible complications following challenging TEVAR like this study. A fenestrated stent graft has fenestrations to preserve supra-aortic vessels, which means the second stent part of this stent graft is the real proximal sealing zone. In another words, proximal aortic arch is a real proximal landing zone. In this study, high incident rate of type Ia endoleak occurs through fenestrations, not through proximal end of the stent graft. More than 10 mm between a fenestration and aneurysm as a proximal neck length should be necessary to treat aortic arch disease using a fenestrated stent graft.

10 Ann Thorac Surg KURIMOTO ET AL 2015;100:24 33 FENESTRATED STENT GRAFT FOR ARCH ANEURYSM DR KHOYNEZHAD: Based on this explanation, most of these endoleaks are from the fenestrations, and therefore type II endoleaks. Is that correct? DR KURIMOTO: No, I think it is type I, because the systemic systolic pressure is going to aneurysm through a fenestration. DR KHOYNEZHAD: I see, this is an experimental stent graft design. How about if we would lengthen the proximal seal zone of this stent graft by 2 to 3 cm proximal to the innominate artery fenestration. Do you think that would help reduce the endoleak rate? Or do you think the lack of radial force is the culprit for the high type Ia endoleak? In either way it seems like there are device design flaws that can be addressed. DR KURIMOTO: Proximal landing zone cannot be changed from proximal aortic arch because there are fenestrations on this stent graft to preserve supra-aortic arch vessels. Regarding radial force, yes, we are thinking about changing the radial force around proximal aortic arch, but I am afraid of type A dissection as the most serious complication in this technique. We have to continue to improve the device. DR WILSON SZETO (Philadelphia, PA): I have 2 questions. One, in terms of preoperative planning, I assume this is CT [computed tomographic] based. What is involved in your planning in terms of choosing the graft with the appropriate dimensions so that it accommodates your arch pathology? Number 2. To follow up on Ali s question, the endoleak rate appears to be high. The question is: Do you think this is a design issue of the endograft system that you can improve on, or do you think this is an inherent shortcoming of the ongoing debate between fenestrated versus branch endografting to address this type of pathology? DR KURIMOTO: We use CT images including axial image and the volume rendering image. And we measure the length, diameter of aorta and the direction of the supra-aortic arch vessels, of course considering aortic arch tortuosity. And the other question? DR SZETO: Fenestrated versus branch technology and the incidence of endoleak at 32%. DR KURIMOTO: Actually, simple TEVAR without surgical management of the left common carotid artery has limitations, like we showed in the presentation. IFU [instructions of us] of this device required more than 20 mm from the fenestration to aneurysm. This study revealed minimum length between a fenestration and aneurysm was more than 10 mm to treat aortic arch disease by less-invasive real endovascular treatment. The high incident rate of type Ia endoleak is simply due to patients selection. DR ERIC ROSELLI (Cleveland, OH): Just a couple of comments and a question. Although you described this as the treatment of zone 0, I do not believe that is accurate. I think the device is residing in zone 0, but the pathology you show is actually involving the distal arch and predominantly descending aortic disease. It would be helpful if you could tell me about the experience before you used the fenestrated devices, when you used a more standard commercially available device. Have you looked at the series of patients that you had with a 1 cm landing zone where you just brought the standard device up to the left common carotid artery, what kind of results did you have? Because I would argue that you could expect you might see about a 30% risk of type I endoleak in those patients without the need for this fenestrated device. Do you have that series to compare with this group? Ultimately, I think what we need are branched devices for treating the more proximal arch. Could you comment on that? DR KURIMOTO: Actually, I totally agree, the proximal landing zone less than 1 cm is not a candidate for endovascular therapy before the era of fenestrated stent graft. But in this study, we wanted to present limitation of TEVAR using a fenestrated stent graft because the patients were relatively high risk for open surgery. Absolutely our first choice is open repair for low-risk patients with aortic arch disease. However, we are also waiting for branched stent grafts for aortic arch aneurysm. Branched stent grafts will expand an indication of endovascular treatment for aortic arch diseases including aneurysm involving the left common carotid artery. If you agree that aneurysm including the left subclavian artery is one of pathology of aortic arch disease, we could present some data treated by totally endovascular treatment without any surgical management of the left common carotid artery for aortic arch diseases. This paper is simply a report of experiences using a fenestrated stent graft for challenging pathology, so we have no group compared with presented data. 33 ADULT CARDIAC