Stent-Graft Repair of the Thoracic Aorta
Evolving applications and results forecast a diminishing role for surgical reconstruction.
To view the figures for this article, please refer to the print version of our January/February 2003 issue, page 18.
Treatment of thoracic aortic disease represents a continuing formidable challenge. Aneurysms, aortic dissection, and other lesions (traumatic injury, penetrating ulcers, pseudoaneurysms, and intramural hematomas) occur with considerable frequency. The incidence of such conditions is almost surely higher than once estimated; this is especially true for thoracic aortic aneurysms (TAA).1-5
Until recently, the only effective treatment modality for thoracic aortic lesions was surgery, involving rather morbid and often complex operations. Surgical reconstruction carries substantial risks of serious complications and mortality, which have diminished only slightly in recent years in spite of significant advances in perioperative care and surgical techniques.6-8 As a result, the majority of patients are managed with benign neglect or observation only. Endovascular stent-graft repair is a truly innovative, less-invasive treatment modality based on the principle of segmental exclusion with an endoluminal prosthesis.9 The new procedure offers patients hope where none existed before.
SUMMARY OF PERSONAL EXPERIENCE
Ninety-one patients were referred for repair of thoracic aortic lesions over the 63-month period ending November 30, 2002. Nineteen (21%) were excluded from consideration of endovascular treatment because of small and/or diseased access iliac arteries (n=10), or because of extensive aneurysmal disease without identifiable landing zones for endograft attachment and seal (n=9). Cardiac or vascular surgeons referred most of the patients after ruling out operative intervention. Seventy-two patients were accepted for stent-graft intervention; 48 had TAA (aneurysms, pseudoaneurysms, and ulcers) and 24 had type B aortic dissection (AD). Aneurysms were considered for repair when one or both of the following were present: symptoms (pain), or a sac with a maximum diameter measuring >5.0 cm or more than twice the diameter of an adjacent nonaneurysmal segment of the thoracic aorta. Only three treated TAA patients had aneurysms <6.0 cm.
Indications for Intervention
Intervention in AD cases was indicated by one or more of the following findings: continuing pain despite ?optimal? medical management, evidence or suspicion of aortic branch and/or lower extremity ischemia, and hemothorax. Three patients with AD did not show a clinical imperative for intervention, but were found to have a markedly compressed true lumen on CT (true/false lumen ratio <0.1). Device design and implant strategy were based on evaluation of morphology by catheter angiography, intravascular ultrasound (IVUS), and 3-mm slice contrast-enhanced spiral computed tomography. Tables 1 and 2 summarize patient characteristics and morphology. Local anesthesia was used in six procedures, 39 patients received spinal, and 27 patients received general anesthesia.
Stent-graft placement for aneurysm exclusion was designed to cover the length of the lesion and extend (if possible) 20 mm proximally and distally. The strategy for treatment of AD involved endograft coverage of the entry site to re-direct blood flow down the true lumen exclusively; it also extended 20 mm in either direction if anatomically possible. Endoluminal access was achieved by single-groin femoral cutdown in 65 instances, and a temporary iliac conduit in seven. A diagnostic pigtail catheter was used in all cases: transfemoral in 19 and transbrachial in 53. The device used on all patients was Medtronic AVE’s (Santa Rosa, CA) Talent patient-specific self-expanding endograft, with a 4 mm-6 mm diameter oversize, and a proximal bare spring (Figure 1). Configuration (for 70 implanted patients) was single-segment in 35, two overlapping segments in 29, and three or more in six. Proximal endograft attachment was in the descending thoracic aorta (DTA) more than 2 cm distal to the origin of the left subclavian artery (LSA) in 18, within 2 cm of the LSA (the parasubclavian aorta) in 31, and proximal to the LSA in 21 (Figure 2). The endograft was placed over (occluding) the origin of the LSA (without revascularization) in 11 instances (Figure 3).
Ten patients had one (n=9) or two (n=1) preliminary adjunctive cervical operations designed to disconnect or revascularize the LSA and/or the left common carotid artery (LCCA), to enable stent-graft placement in the more proximal aortic arch (Figures 4-6). Adjunctive procedures to prevent spinal cord ischemia (ie, spinal fluid drainage) were not used in the course of this series.
Two procedures (2.8%) were aborted because of access failure. One patient (1.4%) died intraoperatively from access-related iliac artery rupture causing massive hemorrhage; a second patient suffered a similar rupture that was repaired successfully with an endoluminal graft. There were no additional 30-day mortalities, but one other death occurred at 60 days from a ruptured and infected thoraco-abdominal aneurysm with a juxtaceliac attachment-site endoleak. Two patients (2.8%) developed symptoms of spinal cord ischemia (one case of paraparesis and one of monoparesis); both resolved within 72 hours. There were no surgical conversions. Sixteen patients (22%) sustained nonfatal morbid adverse events within the first 30 days.
Five secondary endovascular procedures were performed for repair of attachment-site type I endoleak. In total, nine of 70 implanted patients (13%) were shown to have endoleak on the 30-day contrast CT scan; they all underwent further evaluation with catheter angiography, which demonstrated a type I endoleak in seven, and no contrast extravasation in two. Five of the seven patients with type I endoleak underwent secondary endovascular repair, consisting of placement of a proximal endograft extension in two and a distal extension in three. I was able to achieve secure exclusion with resolution of endoleak in all five patients. Endoleak in the other two could not be repaired because of the extent of their disease. The remaining two patients (with negative angiograms) continue to be observed; after 2 months, the CT-detected endoleaks resolved in one and persisted in the other. Both patients’ TAAs have remained stable, without further enlargement. Out of the 24 cases of AD, the false lumen became completely thrombosed after stent-graft placement in 16 of 21 (76%) with a patent channel preoperatively, and only partially thrombosed in four patients. Preoperative false-lumen thrombosis was noted in three patients.
Length of follow-up ranged from 1 to 61 months (mean=26 months); two patients were lost to follow-up and three withdrew from the study. Patient surveillance involved contrast/noncontrast CT scan and plain-film radiographs of the stent-graft device at 6-month intervals. Six deaths have occurred during follow-up so far—five were unrelated to the endograft or aortic pathology and one was due to unknown causes.
Several previous reports have demonstrated the feasibility of stent-graft repair in the thoracic aorta. Dake et al at Stanford have led the way, accumulating one of the most impressive series.9-13 With few exceptions, these efforts have focused on the treatment of TAA and type B AD. Endovascular exclusion of TAA can be a straightforward procedure both in concept and technical execution, especially for those aneurysms that begin more than 2 cm distal to the LSA. Aneurysms that involve the parasubclavian aorta (Figure 7) or extend into the arch itself offer a greater technical challenge. Lengthening the branchless DTA is an effective strategy to expand applicability of stent-graft repair with endograft attachment as proximal as the LCCA, and even up to the innominate artery. Coverage of the origin of the LSA has been reported to apparently be safe, and our own experience supports this.14 Caution should be exercised, however, and, at a minimum, angiographic patency of the contralateral vertebral artery should be ascertained beforehand.
Should vertebro-basilar or arm ischemia develop following stent-graft placement, it would be fairly simple for the surgeon to undertake secondary revascularization by cervical transposition or bypass. On the other hand, aneurysmal involvement or close proximity to the LSA pose a different kind of concern related to the potential for reflux endoleak. Preliminary transposition of the LSA is the preferred strategy for such cases, although my colleagues and I are now exploring the feasibility of simply coiling the LSA after endograft placement. 15,16
Acute aortic dissection is a potentially catastrophic condition (Figure 8), quite distinct from aneurysmal disease. A management consensus of sorts has evolved in which type B dissections are treated medically in the absence of imperatives for intervention (including continued severe pain, acute aneurysmal expansion of the false lumen, rupture, and visceral-branch or leg ischemia).8-10 Surgical treatment involves graft replacement of the segment containing the entry point to redirect blood flow exclusively down the true lumen. Surgeons are reluctant to recommend an aggressive management strategy because of the high rates of serious perioperative complications and mortality. Endovascular repair represents a more appealing, less-invasive approach, but significant questions remain concerning indications and best strategy.17 It would seem that in the absence of imperatives (as defined previously), it might be prudent to reserve stent-graft intervention for patients with significant aneurysmal dilatation, imminent rupture, aortic-branch or distal ischemia, or for those who have failed medical treatment. Compression of the true lumen (by the false lumen), as assessed by CT imaging, may be a good predictor of ischemic complications and the need for intervention.
OVERVIEW AND OBSERVATIONS
The following observations would seem warranted on the basis of the clinical experience reported here:
• Stent-graft repair of TAA and entry-site type B AD is feasible and can be achieved with very acceptable technical success.
• Perioperative morbidity and mortality compare favorably with historical surgical results.
• Medtronic’s Talent customized device has proved versatile for endovascular exclusion in a variety of morphologies; recent improvements of the delivery system allow for easy and precise deployment in the aortic arch.
• Stent-graft applicability can be expanded through adjunctive surgical techniques designed to transpose arch branches to “lengthen” the DTA.
• Clinical indications remain unclear, especially for treatment of AD.
• Very little is known about long-term durability of either the device or the repair after placement.
In all likelihood, endovascular techniques will prove to be the preferred treatment approach for thoracic aortic repair, with distinct advantages over surgical reconstruction.
Frank J. Criado, MD, is the Director of the Center for Vascular Intervention and Chief of the Division of Vascular Surgery at Union Memorial Hospital/MedStar Health in Baltimore. Dr. Criado may be reached at (410) 235-6565; firstname.lastname@example.org.
1. Bickerstaff LK, Pairolero PC, Hollier LH, et al. Thoracic aortic aneurysms: A population based study. Surgery. 1982;92:1103-1108.
2. Pressler V, McNamara JJ. Aneurysm of the thoracic aorta. J Thorac Cardiovasc Surg. 1985;89:50-54.
3. Pressler V, McNamara JJ. Thoracic aortic aneurysm: Natural history and treatment. J Thorac Cardiovasc Surg. 1980;79:489-498.
4. Sorenson HR, Olsen H. Ruptured and dissecting aneurysms of the aorta: Incidence and prospects of surgery. Acta Chir Scan. 1964;128: 644-650.
5. Pate JW, Richardson RJ, Eastridge CE. Acute aortic dissections. Ann Surg. 1976;42:395-404.
6. Svensson LG. Natural history of aneurysms of the descending and thoracoabdominal aorta. J Card Surg. 1997;12(suppl):279-284.
7. Glower DD, Fann JI, Speier RH, et al. Comparison of medical and surgical therapy for uncomplicated descending aortic dissection. Circulation. 1990;82:(suppl IV):39-46.
8. Miller DC. The continuing dilemma concerning medical versus surgical management of patients with acute type B dissections. Semin Thorac Cardiovasc Surg. 1993;5:33-46.
9. Fann JI, Miller DC. Endovascular treatment of descending thoracic aortic aneurysms and dissections. Surg Clin North Am. 1999;79:551-574.
10. Dake MD, Kato N, Mitchell RS, et al. Endovascular stent-graft placement for the treatment of acute aortic dissection. N Engl J Med. 1999;340:1546-15521
11. Dake MD, Kato N, Slonim SM, et al. Endovascular stent-graft placement to obliterate the entry tear: A new treatment for acute aortic dissection [abstract]. Circ J Am Heart Assoc. 1998;98(suppl I):67.
12. Dake MD, Miller DC, Mitchell RS, et al. The first generation of endovascular stent-grafts for patients with descending thoracic aortic aneurysms. J Thorac Cardiovasc Surg. 1998;116:689-704.
13. Fann JI, Dake MD, Semba CP, et al. Endovascular stent-grafting after arch aneurysm repair using the elephant trunk. Ann Thorac Surg. 1995;60:1102-1105.
14. Gorich J, Asquan Y, Siefarth H, et al. Initial experience with intentional stent-graft coverage of the subclavian artery during endovascular thoracic aortic repairs. J Endovasc Ther. 2002;9(suppl II):1139-1143.
15. Criado FJ, Barnatan MF, Rizk Y, et al. Technical strategies to expand stent-graft applicability in the aortic arch and proximal descending thoracic aorta. J Endovasc Ther. 2002;9(suppl II):1132-1138.
16. Criado FJ, Clark NS, Barnatan MF. Stent-graft repair in the aortic arch and descending thoracic aorta: A 4-year experience. J Vasc Surg. 2002;36:1121-1128.
17. Nienaber CA, Fattori R, Lund G, et al. Nonsurgical reconstruction of thoracic aortic dissection by stent-graft placement. N Engl J Med. 1999;340:1539-1545.