Aneurysms of the descending thoracic aorta (DTA) affect an estimated 3 to 4 per 100,000 adults. The surgical treatment of DTA aneurysms began in the 1950s through pioneering work by DeBakey, Cooley, and others. Over the years, a number of advances have been made in the surgical techniques and perioperative care of patients with DTA aneurysms. Cerebrospinal fluid (CSF) drainage is now routinely applied, and there has been a shift away from the clamp-and-sew approach to the routine use of left-heart bypass, total cardiopulmonary bypass, or hypothermic circulatory arrest. This progress has enabled patients to undergo surgical resection with improved outcomes and morbidity. Nevertheless, many patients with DTA aneurysms are denied open surgical repair because of older age and multiple comorbidities. The concept of using an endovascular stent graft in patients with thoracic aortic disease emerged out of the desire to avoid the hazards of open surgery in this high-risk population.
Initially introduced by Parodi and colleagues for the treatment of abdominal aortic aneurysms (AAAs), endovascular stent-graft technology has been applied for DTA aneurysms for nearly 2 decades. Led by favorable outcomes with stent-graft repair of abdominal aneurysms, the group at Stanford University first applied stent-graft technology for the treatment of DTA aneurysms in the early 1990s. Thirteen high-risk nonoperable patients with DTA aneurysms were treated with custom-designed stent grafts, each constructed with self-expanding Gianturco Z stents (Cook Medical, Bloomington, Ind.) placed with Dacron (DuPont, Wilmington, Del.) grafts. Deployed through peripheral vascular access, the grafts enabled the exclusion and depressurization of the aneurysmal sac without the need for thoracotomy and aortic cross-clamping. Placement of these stents was successful in all patients, with thrombosis of the aneurysm surrounding the stent occurring in 12 of the 13 patients. With favorable early results, the study was then extended to the treatment of 103 additional patients with DTA aneurysms, many of whom were deemed unsuitable for conventional open surgical repair. Highly satisfactory results were achieved.
The first-generation “homemade” devices were limited because of the inflexible delivery system, making it difficult to navigate a tortuous aorta or to achieve secure fixation across an angled aortic arch. Years of experience with the endovascular repair of DTA aneurysms and AAAs have led to refinements in stent-graft technology and the commercial production of stent grafts. The Gore thoracic aortic graft (TAG) thoracic endoprosthesis (W.L. Gore and Associates, Newark, Del.) became the first thoracic stent graft to be approved by the U.S. Food and Drug Administration (FDA) for the treatment of DTA aneurysms in 2005. Approval subsequently followed for the Talent system (Medtronic Vascular, Santa Rosa, Calif.) and the TX2 Zenith system (Cook Medical). These grafts are manufactured with polytetrafluoroethylene (PTFE) or polyester for the graft material and nitinol or stainless steel for the stent material. Available DTA stent grafts feature less stiff and lower-profile delivery systems (20- to 28-Fr access), less porous graft material, greater flexibility, and a wider variety of available sizes.
Endovascular approaches have emerged as less invasive treatment options for patients with DTA aneurysms. They are well tolerated, even in elderly patients, and are associated with both shorter stays in the intensive care unit and decreased blood transfusion requirements. Perioperative mortality and paraplegia risk compare favorably with open surgical repair rates.
An aneurysm of the DTA is defined as a localized or diffuse dilatation, with a diameter at least 50% greater than the adjacent normal-sized aorta. In an average-height older man with a normal distal aortic arch diameter of 2.8 cm, dilatation of the proximal DTA of 5.6 cm or greater is defined as aneurysmal. The most common risk factors for aneurysmal degeneration of the aorta include smoking, chronic obstructive pulmonary disease, hypertension, atherosclerosis, bicuspid aortic valve, and genetic disorders.
In an asymptomatic patient with a DTA aneurysm, the risk of rupture or dissection increases to more than 40% per year when the aorta grows larger than 7 cm in diameter. Therefore intervention of an asymptomatic DTA aneurysm before it reaches 6 cm preempts most ruptures and dissections. For asymptomatic patients who are in excellent health and anatomically suitable for endovascular repair, DTA aneurysms 5.5 cm or larger in diameter and eccentric saccular aneurysms greater than 2 cm in width are potential candidates for repair. In contrast to asymptomatic aneurysms, aneurysms associated with symptoms should be treated regardless of size, if there are no contraindications, because symptoms are believed to be indications of impending rupture. As a result of compression or erosion into surrounding structures, DTA aneurysms may lead to back pain, abdominal pain, hoarseness, dysphagia, dyspnea, hemoptysis, or hematemesis. Spinal cord compression or thrombosis of spinal arteries may result in paraparesis or paraplegia. Embolization of atheromatous debris could lead to distal ischemia involving the viscera, kidneys, or lower extremities. Acute rupture leads to cardiovascular collapse.
Available thoracic aortic stent grafts are only approved by the FDA for the treatment of DTA aneurysms. However, they are commonly applied “off label” for the management of other thoracic aortic pathology, including penetrating atherosclerotic ulcer (PAU), intramural hematoma (IMH), aortic dissection, and traumatic aortic transaction. A PAU represents rupture of an atherosclerotic plaque, with penetration into the internal elastic lamina of the aorta. Although slow growing, PAU may lead to saccular aneurysm formation, severe pain, and aortic rupture. An IMH is thought to result from the spontaneous rupture of the aortic vasa vasorum, with hemorrhage into the aortic media. IMH may be associated with pain and can lead to an intimal tear and aortic dissection. Enlarging or symptomatic PAU and IMH represent appropriate indications for DTA stent-graft repair. The endovascular management of aortic dissection and traumatic aortic disruption are discussed in other chapters of this book.
Cardiac evaluation. A noninvasive myocardial stress test should be performed, such as exercise treadmill testing, dobutamine stress echo cardiography, or a persantine thallium study. Coronary angiography may be warranted if significant coronary artery disease is identified.
Carotid artery duplex imaging should be obtained.
Imaging the vascular anatomy. Computed tomography (CT) angiography with the possibility of three-dimensional (3D) reconstruction provides excellent visualization of the aorta, including aneurysm diameter, thrombus characteristics, branch vessel anatomy, and extent of calcification. Magnetic resonance imaging (MRI) is also capable of providing angiographic images and 3D reconstruction. Although it avoids ionizing radiation, MRI cannot be used in emergency circumstances or for patients with implanted pacemakers. Thin-slice CT angiography of the thorax, abdomen, and pelvis with distal arterial runoff and 3D reconstruction of the aorta should be obtained ( Fig. 20-1 ).
Assessment of aortic landing zones. Appropriate length and diameter of the aorta are required proximal and distal to the aneurysm for DTA stent-graft repair ( Fig. 20-2 ). The landing zones should be long enough to allow safe deployment between the aneurysm and the brachiocephalic arteries proximally and the celiac artery distally. In general, proximal and distal landing zones must be 2 cm in length and free of thrombus and calcification to achieve adequate seal and prevent endoleak. Available devices are 22 to 46 mm in diameter, and because stent grafts are oversized by 10% to 20%, landing zone diameters must be larger than 19 mm and smaller than 43 mm in diameter.
Assessment of iliofemoral vascular access. Size, tortuosity, and calcification of the iliofemoral vasculature must be examined preoperatively to determine the need for an iliac artery conduit.
Pitfalls and Danger Points
Avoiding stroke. The risk of stroke after DTA stent-graft repair was noted to be as high as 8% in early reports and was thought to be related to the manipulation of catheters and sheaths in and around the aortic arch and its branches. Older stent-graft delivery systems required the handling of large, stiff sheaths and dilators. Newer deployment systems only require a guidewire to pass through the arch and therefore are associated with less manipulation, with recent reports documenting a stroke risk of 3% to 4%. Coverage of the left subclavian artery may increase the risk of posterior circulation strokes, particularly if the right vertebral artery is absent or stenotic. However, the majority of strokes are atheroembolic. Therefore arch wire and catheter manipulations should be kept to a minimum.
Avoiding paraplegia. Spinal cord ischemic injury has been reported at a rate of 2% to 4% after DTA stent-graft repair. The risk of paraplegia or paraparesis is greatest if collateral blood supply to the spinal cord has been compromised during previous aortic intervention, such as through an open or endovascular abdominal aortic aneurysm repair or by the presence of unilateral or bilateral occlusion of the internal iliac artery. Likewise, the risk of paraplegia increases if extensive thoracic aortic coverage or coverage of left subclavian artery is planned for the current DTA intervention. Because large patent intercostal arteries cannot be reimplanted during stent-graft repair, care should be taken to avoid excessive distal coverage of the DTA, and the patency of as many intercostal arteries as possible should be maintained. The left subclavian artery supplies the superior portion of the anterior spinal artery via the vertebral artery and may therefore represent an important collateral source of spinal perfusion, especially in patients with prior abdominal aortic aneurysm repair. Some reports have suggested coverage of the left subclavian artery increases the risk of paraplegia during DTA stent-graft repair and have recommended routine preoperative revascularization with carotid-subclavian bypass, but others have found this not to be the case. If concerns are raised about compromised spinal cord collateral perfusion, or if long aortic coverage is required, a CSF drain may be placed immediately before surgery. The drain is left in position for 48 hours. After the procedure, the patient’s blood pressure is raised to 140- to 160-mm Hg systolic to encourage collateral blood flow to the spinal cord. If delayed neurologic injury arises after surgery, elevating systemic blood pressure and instituting CSF drainage can occasionally reverse neurologic deficits. For patients with concomitant abdominal aortic aneurysm and DTA aneurysm, the repair is staged and the larger aneurysm treated first, with the assumption that collateral circulation to the spinal cord will develop during the interval period.
Avoiding arm ischemia. A significant proportion of DTA aneurysms lie close to the left subclavian artery. Therefore stent-graft coverage across the subclavian origin is sometimes necessary to ensure adequate fixation of the device at the proximal landing zone and to exclude the aneurysm. Preemptive revascularization of the left subclavian artery before covering its origin is required in the setting of a dominant left vertebral artery, previous coronary artery bypass graft surgery with a patent left internal mammary artery graft, and functioning arteriovenous fistula in the left upper extremity. In the absence of these absolute indications, controversy exists regarding whether routine preoperative revascularization of the left subclavian artery is required. Some centers have reported complications associated with coverage of the left subclavian artery, such as left arm claudication or vertebrobasilar insufficiency, and advocate revascularization of the left subclavian artery before DTA stent-graft repair. The authors have observed that only 2% of patients develop postoperative left upper extremity ischemia, and thus advocate an expectant approach. Subclavian artery revascularization can be performed expediently for the treatment of left upper extremity ischemia. To identify patients who require a left carotid-subclavian bypass, preprocedural carotid and vertebral duplex ultrasound imaging and CT angiography can be used to evaluate the patency, size, and location of the vertebral arteries; to rule out an aortic arch origin of the left vertebral artery; and to document an intact circle of Willis.
Avoiding proximal type I endoleaks. A proximal type I endoleak is the most common type of endoleak seen after DTA stent-graft repair and represents a lack of apposition between the graft and the aortic wall, leading to active arterial flow within the aneurysmal sac. A type I endoleak is more common in the thoracic aorta than in the abdominal aorta because of the short and frequently angled attachment zone in the aortic arch. Achieving proximal anchorage of at least 2 cm is necessary to avoid a proximal type I endoleak. Although type I endoleaks occasionally seal spontaneously, the failure of sealing usually results in persistent sac pressurization, leaving the patient at risk of future aneurysmal rupture. Therefore identification and management of a proximal type I endoleak at the time of the initial stent-graft repair is critical. Proximal type I endoleaks can be visualized with aortography after stent-graft deployment. A proximal type I endoleak can often be corrected with angioplasty using a large-diameter, compliant balloon. Balloon angioplasty helps ensure that the stent graft is fully expanded and improves contact with the aortic wall. Should the endoleak persist despite angioplasty, consideration should be given to placement of a stent-graft extension to lengthen the seal zone. This may require coverage of the left subclavian artery and stent-graft deployment at the origin of the left common carotid artery to increase the length of the proximal landing zone. Alternatively, aortic arch debranching techniques with carotid-carotid bypass can be considered to further lengthen the proximal landing zone ( Figs. 20-3 and 20-4 ). In certain circumstances a proximal type I endoleak can be treated with the deployment of a balloon-mounted bare metal stent to the proximal portion of the stent graft to enhance apposition of the stent graft to the aortic wall.
Avoiding distal type I endoleak. A distal landing zone of at least 2 cm is required to avoid a distal type I endoleak at the time of DTA stent-graft repair. Should a distal type I endoleak be identified after stent-graft deployment, balloon angioplasty may be applied as an initial step to ensure that the stent graft is fully expanded distally and to improve graft contact with the aortic wall. A persistent distal type I endoleak may necessitate the distal deployment of a stent-graft extension or a balloon-mounted bare metal stent to lengthen the seal zone or improve graft contact with the aortic wall. If the endoleak persists and the distal landing zone extends to the origin of the celiac artery, either visceral debranching of the celiac artery can be performed via laparotomy or the celiac artery can be embolized and covered with a distal graft extension ( Figs. 20-5 and 20-6 ). This latter approach, without celiac artery revascularization, increases the risk of visceral ischemia.
Avoiding proximal type II endoleaks. Retrograde filling of the excluded aorta from a patent subclavian artery may predispose patients to type II endoleaks.19 These endoleaks can be treated at a delayed interval should they persist.
Avoiding visceral ischemia. Purposeful coverage of the celiac artery during DTA stent-graft repair should be avoided to reduce the risk of visceral complications. If the distal sealing zone proximal to the celiac artery is of insufficient length and sacrificing the celiac artery is anticipated, angiographic evaluation should first be performed to assess the collateral circulation between the celiac artery and the superior mesenteric artery. Temporary balloon occlusion of the celiac artery and selective angiography of the superior mesenteric artery determine whether sufficient collateral supply is supplied by the superior mesenteric artery via the gastroduodenal artery. Coil embolization of the celiac artery can then be performed, allowing a distal stent-graft extension to be deployed up to the origin of the superior mesenteric artery. Experience with coverage of the celiac artery to extend the distal landing zone remains limited.
Access Vessel Tortuosity and Size
Adequate vascular access is critical for DTA stent-graft deployment, and preoperative imaging must include an assessment of the iliofemoral vasculature. Current DTA stent-graft devices require a large-caliber delivery system that ranges from 20 to 24 Fr (0.7-0.8 cm) in outer diameter. Therefore the presence of small, tortuous, or calcified iliac vessels may prove to be hazardous in advancing the sheaths necessary for stent-graft deployment, and such patients are at increased risk of iliac artery rupture. Tortuous iliac vessels force the relatively stiff delivery catheter to assume a variety of angles as it negotiates the curves, increasing friction and decreasing the “pushability” of the device at each of these bends ( Fig. 20-7 ). Excessive force can lead to iliac artery injury. Fortunately, iliac artery tortuosity can be overcome with the use of superstiff guidewires such as the Amplatz SuperStiff or Meier guidewire (Boston Scientific, Natick, Mass.) or Lunderquist wires (Cook Medical). These wires straighten the iliac system and enable improved tracking of the device into position.