Endovascular treatment of arch and proximal thoracic aortic lesions
Article Outline
Objective
To analyze at one institution the endovascular treatment for aortic arch and proximal thoracic aortic lesions, categorize open arch reconstruction, and make preliminary recommendations based on pathology (dissection vs aneurysm), and anatomical extent of disease.
Methods
A retrospective review of aortic arch and descending thoracic aortic lesions managed with endovascular treatment between June 2002 and June 2007.
Results
Thirty-four patients received endovascular repair for aortic dissection (n = 28) and aneurysm (n = 6). Open supra-aortic transposition or debranching of the great vessels was performed in 14 cases of dissection (50%) and six cases (100%) of aneurysm. In 14 dissections, the entry tear was located in the distal aortic arch, enabling the left subclavian artery to be sealed without reconstruction. The procedures were successful in 33 patients (97.1%); one intraoperative death occurred. Type I endoleaks were found intraoperatively in eight cases. After management with balloon angioplasty and by extending the stent implantation, the endoleaks resolved in four cases and decreased in four cases. One patient with Stanford type A dissection died from an unknown cause 3 months after treatment. The overall survival rate was 94.1% (32/34), and all bypass grafts remained patent during the follow-up period.
Conclusions
Endovascular stent grafting is a safe and effective method for the treatment of aortic arch lesions. Transposition of the supra-aortic great vessels can be effectively combined with endovascular stent grafting to ensure both cerebral blood supply and enough landing area for the stent graft.
The conventional method for open aortic arch replacement using a prosthetic graft to reconstruct all three great vessels is often accompanied by complications, significant morbidity, and high mortality.1, 2, 3, 4 Although endovascular stent grafting has greatly enhanced the progress of endovascular repair of lesions originating from and confined to the descending aorta and abdominal aorta,5, 6, 7, 8, 9 the endovascular treatment of aortic arch lesions is still very difficult because it is necessary to reserve or reconstruct one or more of the supra-aortic great vessels. Thus, it is essential to find a safe and effective method for preserving the blood supply to the brain before endovascular repair of aortic arch lesions.
This retrospective analysis was designed to review the endovascular treatment of aortic arch and proximal aortic pathology, categorize open arch reconstruction, and make preliminary recommendations based on pathology (dissection vs aneurysm) and anatomical extent of disease. Extent of open arch revascularization was examined in relation to disease presentation (dissection vs aneurysm) and anatomic extent, and categorized into three groups: (1) ascending aorta (AO) based in-situ reconstruction; (2) extra-anatomic reconstruction (carotid to carotid or carotid to subclavian); and (3) no open debranching.
Methods
Patient data
From June 2002 to June 2007, 215 patients were treated at our institution with the stent graft for thoracic aorta disease. This study is a retrospective analysis of the 34 patients with aortic arch involvement. Patient data is presented in Table I, Table II. We included only patients with aortic arch lesions who received endovascular treatment in that period; thus, our series only accounts for 42% (34/81) of all aortic arch disease in that period. More than half of the patients did not receive endovascular treatment because there was no indication, or they were unwilling or could not afford to receive it, or they preferred open surgery. Diagnosis was confirmed by magnetic resonance angiography (MRA), enhanced computed tomography scan angiography (CTA), and angiography in all cases. Intravascular ultrasound was not performed on any of the patients. The distal extent of all disease processes in patients in this series was confined to the thoracic or abdominal aorta. Cases of aortic trauma resulting in pseudoaneurysm or dissection and cases involving the aortic valve or coronary artery orifices were excluded from this patient cohort. The indications for acute aortic arch involved Stanford type B dissection include pain not relieved after proper control of blood pressure, a very large dissecting aneurysm confirmed by imaging (including arteriography), a true lumen that is markedly compressed, and a suspicion of rupture. Additionally, a patient's insistence on undergoing emergency surgery after being told the prognosis of the disease is an indication. No patients in this series were receiving outpatient antihypertensive medications prior to admission. In this experience, if a patient presented to the emergency department hypertensive, endovascular aortic repair was delayed until antihypertensive medications were initiated and the blood pressure stabilized.
Table I. Clinical data of patients with aortic arch aneurysm (n = 6)
| Sex | |
| 5 | |
| 1 | |
| Age (y) | 57 |
| Time from onset to surgery | 5 months-10 years |
| Length of aneurysm (cm) | 16.4 |
| Diameter of aneurysm (cm) | 6.6 |
| Aortic arch between BA and LCCA involved | 2 cases |
| Aortic arch between LCCA and LSA involved | 3 cases |
| LSA involved | 1 case |
| Endovascular treatment combining supra-aortic transposition | 6 cases |
| One-stage procedure | 4 cases |
| Two-stage procedure | 2 cases |
Table II. Clinical data of patients with aortic arch dissection (n = 28)
| Sex | |
| 26 | |
| 2 | |
| Age (y) | 55 |
| Time from onset to surgery (days) | 2-26 |
| Sudden chest pain | All |
| Upper back pain | All |
| Hypertension | All |
| Acute dissection (< 14 days) | 18 |
| Chronic dissection (> 14 days) | 10 |
| Stanford type A (distal AO involvement) | 20 |
| 10 | |
| 10 | |
| Stanford type B (AO uninvolved, distal AA involved, entry tear beyond LSA) | 8 |
| Entry tear size (mm) | 10.6 |
| Maximal diameter of dissection (cm) | 7.2 |
| Endovascular treatment with supra-aortic transposition | 14 |
| 11 | |
| 3 | |
| Endovascular treatment alone | 14 |
Surgical technique
All 34 cases underwent endovascular stent grafting with endotracheal anesthesia. Among them, transposition of the supra-aortic vessels was performed on 20 cases, including 14 aortic dissections and six aortic aneurysms (Table III). Then, one-stage or two-staged endovascular stent grafting was performed. Expanded polytetrafluoroethylene (ePTFE) Gore-Tex grafts (W. L. Gore & Associates, Inc, Flagstaff, Ariz) were used for all bypasses. The entry tear was located in the aortic arch area in the patients with dissection. The dissection only involved the brachiocephalic trunk area (ie, distal ascending aorta) and most of the ascending aorta. The Stanford classification cannot be applied to aortic arch dissection; some authors refer to it as type “non-A, non-B”. However, these dissections were still classified as type A because the distal ascending aorta was involved. Bypass surgery could be performed between the ascending aorta and supra-aortic vessels because most of the ascending aorta was uninvolved.
Table III. Category and extent of supra-aortic debranching or reconstruction
| Ascending aortic based in-situ reconstruction (n = 4) | |
| 2 | |
| 1 | |
| 1 | |
| Extra-anatomic reconstruction (n = 16) | |
| 11 | |
| 5 | |
| No debranching or reconstruction | 14 |
When the intimal entry tear or aortic aneurysm was located in the distal aortic arch, one-stage sealing of the left subclavian artery (LSA) was carried out to ensure enough landing area for the stent graft. Preoperative vertebral artery angiography was performed. If the right vertebral artery (RVA) and circle of Willis in the brain could be observed clearly in the angiograph, LSA reconstruction was not performed. If there was more than 70% obstruction or stenosis in the RVA, LSA reconstruction was performed using one of two methods: a prosthetic bypass graft was placed between the left common carotid artery (LCCA) and the LSA, or the proximal end of the LSA was cut down and anastomosed to the LCCA.10
The right common femoral artery approach was used in all the cases to place the endovascular stent grafts. We employed several different types of commercially available stent grafts (Table IV). Stents from the same manufacturer were used when a patient required multiple stents. Because all six cases with aortic aneurysms involved the aortic arch, supra-aortic transposition of the great vessels was performed before either one- or two-stage endovascular stent grafting (Table I). In the first stage, supra-aortic transposition of the great vessels was performed. In the second stage, endovascular treatment was carried out.
Table IV. Types and number of endovascular stents
| Talent Medtronic (Medtronic, Inc, Minneapolis, Minn) | 39 |
| (36) | |
| (3) | |
| Zenith TX2 (Cook Medical, Bloomington, Ind) | 4 |
| Ancura II (Xianjian Science and Technology Co, Shenzhen City, Guangdong, China) | 2 |
The supra-aortic transposition of the great vessels was carried out on 14 patients with aortic dissections before the endovascular treatment (Table II). The one-stage procedure was performed for a very large dissecting aneurysm or true aneurysm (especially large diameter), severe pain, and high possibility of rupture, thus avoiding the rupture at the interval between two stages. In addition, the one-stage procedure can reduce expense, prevent a second anesthesia, and prevent readmission. Also, most patients choose a one-stage procedure. One- or two-stage stent grafting was performed to seal the entry tear (among them, three cases received two-stage endovascular treatment). In the other 14 cases of aortic dissections with the entry tear located in the distal aortic arch, intraoperative sealing of the LSA was carried out without reconstruction because the preoperative vertebral angiography showed excellent blood supply to the RVA. If an endoleak occurred after stent graft implantation, ballooning was carried out to make the stent graft closely adhere to the blood vessel wall. If the endoleak was caused by stent displacement or angulation, an aortic extending stent graft was implanted. If the endoleak was caused by reflux of the LSA, an occluder was implanted from the left brachial artery to seal the opening of the LSA. No special spinal cord protection was used during surgery. Intrathoracic drainage tubes were placed in patients who underwent thoracotomy for transposition of supra-aortic great vessels; no drainage tubes were used in other patients.
Results
Surgery was successful in 33 patients (97.1%). A representative surgery with graft is shown in the figure. One intraoperative death occurred in a patient who had a Stanford type A aortic dissection. In this case, a bypass graft was placed between the LCCA and the RCCA, after which, one-stage endovascular treatment was carried out and a stent graft was successfully deployed to seal the entry tear. A delayed type I endoleak was found and balloon angioplasty was conducted. During this procedure, the patient experienced cardiopulmonary arrest and died. Type I endoleaks were also found in eight other cases following stent grafting, and were managed intraoperatively. In one case, the LSA was sealed during the operation, and the endoleak was thought to be caused by reflux of the LSA. An occluder was implanted from the left brachial artery to seal the opening of the LSA. Postprocedure angiography indicated that the endoleak had disappeared. In three cases, the endoleak did not decrease despite management, and an aortic extending stent graft was implanted to exclude the endoleak. In four cases, the endoleaks decreased significantly after ballooning. Angiography showed that they were all delayed endoleaks (they appeared after three cardiac cycles in angiography) and were very small. No further treatments were performed. In four of 33 cases (12%) endoleaks were noted postoperatively. The endoleaks of these four cases were resolved without intervention during the follow-up period, one in 3 months, two in 6 months, and one in 12 months postoperatively. Aphasia appeared in one case on the first postoperative day. A CT scan revealed no detectable cerebral hemorrhage and cerebral infarction, and there was no evidence it was caused by the bypass procedure. The patient recovered after a week of active treatment.

Fig.
Stanford type A aortic dissection. A, Digital subtraction angiography (DSA) showing intimal entry tear in the aortic arch, involving left common carotid artery (LCCA). B, Previously placed Y-shaped bypass graft connected ascending aorta to LCCA and left subclavian artery. C, Stent graft was implanted to seal the intimal entry tear without endoleak and maintain patency of the graft.
No cases required immediate or late surgical conversion. The mean follow-up duration was 26.2 ± 4.5 months (range, 1-61). The follow-up was conducted by three-dimensional CTA examination at 1, 3, 6, and 12 months postoperatively and once a year thereafter. No stent migration was noted in any of the cases. There were no postoperative complications, such as infection, myocardial infarction, congestive heart failure, venous thrombosis, acute renal failure, or acute respiratory failure. No paraplegia was observed. Complete thrombus formation in the false lumen of the aortic dissection was found in 20 cases at the last follow-up, and partial thrombus formation was found in eight cases within a shorter follow-up period. Complete thrombosis was observed in all lumina of the aortic aneurysms at the 6-month postoperative follow-up. The organization and reshaping of thrombi were found by CT scan in the 1 to 1.5 cm space between the true and false lumina in the dissection cases within 12 months after surgery. The true lumina were significantly enlarged at 3 months postoperative and their size remained unchanged up to 12 months postoperative. The bypass prostheses were patent at the last follow-up in all patients who underwent the transposition of supra-aortic vessels; no stenosis or occlusion was noted. One patient with Stanford type A aortic dissection died from an unclear cause in the follow-up period (3 months after operation) while the other 32 cases survived.
Discussion
Aneurysms and dissections involving the aortic arch are at risk of progressive enlargement and eventual rupture, and the 5-year survival rate of thoracic aneurysms range from 15% to 55%.11, 12 The 5-year survival rate in cases involving the aortic arch is even lower.12 The conventional surgical treatment is open aortic arch replacement. Though surgical techniques have improved significantly, mortality is still as high as 25.6%12 due to intra- or postoperative cerebral stroke and related complications. In the last 10 years, endovascular stent grafting has become a safe and effective method for treating aortic aneurysms and dissections. However, because of the anatomy of the aortic arch, lesions in the aortic arch are still difficult to treat using endovascular methods. The most difficult problem is preserving the supra-aortic great vessels to ensure the blood supply to the head and neck while excluding the aneurysm or sealing the intimal tear at the same time. Most cases lack enough landing area in the proximal end for stent graft deployment. However, adequate landing area for the stent graft can be achieved by transposition of the great vessels in the aortic arch.
In 1998, Buth et al13 reported a case where endovascular methods were used to successfully treat an aortic arch aneurysm. A Y-shaped bypass graft connected the AO to the LSA and the LCCA. Then a stent graft was inserted from the third artificial vessel (anastomosed to the superior aspect of the AO) to the aortic arch to seal the aneurysm, as well as the LSA and the LCCA. In 2003, Dietl et al14 reported two cases of Stanford type A dissections that were treated by displacing the AO, but later the dissection extended to the aortic arch. They carried out Y-shaped bypass surgery with a prosthetic graft to connect the AO to the BA and the LCCA. Then the aortic arch lesion was treated using an endovascular stent graft. In 2006, Bergeron et al12 reported a group of cases treated with this hybrid surgical and endovascular procedure. In 15 cases, all of the great arteries of the aortic arch were transposed and partial transposition of the great vessels was performed in 10 cases. A stent graft was then placed to seal or exclude aortic dissections or aneurysms. The success rate of this technique was reported to be 92%. Similarly, in our series, surgery was 97% successful. In the same year, Saleh et al15 successfully treated 15 aortic aneurysms using transposition of the supra-aortic great vessels followed by endovascular stent graft deployment. Y-shaped prosthetic grafts were placed to bypass the AO to the BA and the LCCA. In cases where the RVA was occluded, bypass grafts were placed between the LCCA and the LSA. Two to 3 weeks later, stent grafts were inserted via the femoral or iliac artery to exclude the aneurysm. The success rate of these procedures was 100%. All stent grafts and bypass vessels were patent without endoleaks, stent displacement, or neurological sequelae in the early postoperative period. One patient died 2 months postoperatively because of a pulmonary complication.
Twenty-eight cases in our series were aortic dissections, including 18 acute cases which were classified as Stanford type A, all of them forming aortic dissecting aneurysms. In 11 of the acute cases, supra-aortic transposition of the great vessels was performed first because of the lack of landing area for the stent graft. After transposition, one-stage endovascular stent grafting was performed to seal the intimal entry tear. Each case in which endovascular stent grafting was performed after supra-aortic transposition of the great vessels was successful, not only effectively sealing or excluding the aortic arch lesions, but also preserving the blood supply to brain. Except for a mild cerebral accident in one case, there were no peri- or postoperative neurological complications, indicating that a combination of open supra-aortic transposition of the great vessels followed by endovascular stent grafting is an effective and safe method for treating aortic arch lesions. Intimal tears were located in the distal aortic arch in 14 aortic patients in which preoperative vertebral artery angiography showed sufficient blood supply by the RVA. Therefore, intraoperative sealing but no reconstruction of the LSA was carried out as we sealed the entry tear. In these cases, there was no postoperative brain or upper extremity ischemia, indicating that it is safe to seal the LSA in cases with sufficient blood supply by the RVA.
Type I endoleaks occurring after stent graft implantation can be treated by performing balloon angioplasty of the stent graft to make the stent tightly adhere to the blood vessel wall.8 In nine cases with type I endoleaks after stent graft implantation, balloon angioplasty was performed. If there was still a significant endoleak after ballooning, the causes were analyzed. If the stent graft was near the entry tear but did not completely seal it, an aortic extending stent graft was implanted to effectively seal the incompletely sealed entry tear. If the endoleak was caused by blood reflux of the LSA that was not sealed when the entry tear was sealed, an occluder was implanted to seal the opening of the LSA, thereby excluding the endoleak. One must pay special attention to the condition of arterial intima when managing endoleaks because balloon angioplasty can aggravate an intimal tear. In one case, the patient died of sudden cardiac arrest during angioplasty. We determined it was caused by aggravation of the intimal tear, inducing rupture of the aneurysm. Therefore, we suggest that endoleaks do not need to be completely excluded and extensive ballooning should be avoided. If the endoleak decreases significantly after proper ballooning, ie, a delayed endoleak with a volume less than 10%, no other management is needed. In most cases, however, endoleaks should be treated actively.
In our series, only one patient developed ischemic symptoms (aphasia) and recovered following a week of active treatment. No evidence indicating that it may have been the result of a small fragment of plaque released during the operation was found.
Conclusions
Our experience suggests that endovascular treatment of aortic arch lesions is safe and effective, with an acceptable risk of injury, and few complications. In cases without enough landing area for stent graft in the proximal end, complete or partial supra-aortic transposition of the great vessels can be performed to ensure both cerebral blood supply and enough landing area for the stent graft. This kind of combined procedure can be performed in one or two stages, depending on the location and type of lesion. As the follow-up of the cases in this group was relatively short (2-6 years), longer follow-up periods may be warranted in order to further demonstrate the efficacy of the procedure.
References
- . What is the best treatment for patients with acute type B aortic dissections–medical, surgical, or endovascular stent grafting?. Ann Thorac Surg. 2002;74:S1857–S1863
- . Natural history of aneurysms of the descending and thoracoabdominal aorta. J Card Surg. 1997;12(Suppl):279–284
- . The continuing dilemma concerning medical versus surgical management of patients with acute type B dissections. Semin Thorac Cardiovasc Surg. 1993;5:33–46
- . Endovascular treatment of descending thoracic aortic aneurysms and dissections. Surg Clin North Am. 1999;79:551–574
- Nonsurgical reconstruction of thoracic aortic dissection by stent graft placement. N Engl J Med. 1999;340:1539–1545
- Endovascular stent graft placement for the treatment of acute aortic dissection. N Engl J Med. 1999;340:1546–1552
- Endovascular stent graft placement in aortic dissection: a meta-analysis. Eur Heart J. 2006;27:489–498
- Midterm results of stent graft repair of acute and chronic aortic dissection with descending tear: the complication-specific approach. J Thorac Cardiovasc Surg. 2002;124:306–312
- Endovascular stent graft repair of abdominal and thoracic aortic aneurysms: a ten-year experience with 817 patients. Ann Surg. 2003;238:586–593
- . Stent graft repair in the aortic arch and descending thoracic aorta: a 4-year experience. J Vasc Surg. 2002;36:1121–1128
- Yearly rupture or dissection rates for thoracic aortic aneurysms: simple prediction based on size. Ann Thorac Surg. 2002;73:17–28
- Great vessel management for endovascular exclusion of aortic arch aneurysms and dissections. Eur J Vasc Endovasc Surg. 2006;32:38–45
- . Combined approach to stent graft treatment of an aortic arch aneurysm. J Endovasc Surg. 1998;5:329–332
- . Off-pump management of aortic arch aneurysm by using an endovascular thoracic stent graft. J Thorac Cardiovasc Surg. 2003;126:1181–1183
- . Combined surgical and endovascular treatment of aortic arch aneurysms. J Vasc Surg. 2006;44:460–466
Competition of interest: none.
PII: S0741-5214(08)00260-7
doi:10.1016/j.jvs.2008.02.010
© 2008 The Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.
