Rupture of abdominal aortic aneurysms (AAAs) accounts for over 15,000 deaths per year in the United States and is the 10th leading cause of death in men older than 55 years of age (Fig. 56-1).¹ Out-of-hospital AAA rupture is associated with a mortality of 80% to 90%. Most deaths are preventable by the early diagnosis and treatment of AAA, and as a result, more than 60,000 surgical or endovascular procedures for AAA are performed annually in the United States. Despite the excellent results of open surgical repair for the prevention of aneurysm rupture, this procedure is associated with significant morbidity and mortality, especially in high-risk patients. Endovascular aneurysm repair (EVAR), which was developed as a less invasive alternative to open surgical repair, is gaining widespread acceptance and is now performed in approximately 80% of elective aneurysm repairs. This chapter reviews the epidemiology, technical aspects, outcomes, and complications of EVAR for the treatment asymptomatic and symptomatic AAA.

The incidence of AAA increases with age. Approximately 2.6% of men and 0.5% of women between 45 and 54 years of age have an AAA, and by 75 to 84 years of age, 19.8% of men and 5.2% of women have an AAA.¹ Men are affected 4 to 6 times more frequently than women.² AAA is also more common in Caucasian patients than in black, Hispanic, or Asian patients.³

Approximately 15% of patients with AAA have a family history of AAA.⁴ Because the pathogenesis of AAA is thought to be multifactorial, including disorders of enzymes regulating connective tissue homeostasis, approximately 5% of patients with an AAA have a concomitant thoracic aneurysm and approximately 15% have either a femoral artery aneurysm or a popliteal aneurysm.⁵ Synchronous peripheral aneurysms are more common in men, while synchronous thoracic aneurysms may be more common in women.⁶ Approximately 10% of AAAs are juxtarenal, and iliac artery involvement occurs in up to 22% of AAAs.

Patients with AAAs often have significant medical and cardiac comorbidities. Coronary artery disease is found in 40% to 70% of patients with AAAs, with evidence of prior myocardial infarction in up to 46% of patients.⁷ Cerebrovascular disease is present in 25% of patients, claudication is present in 28%,⁷ hypertension is present in 55%, and chronic obstructive lung disease is present in approximately 20% to 25%. A history of blunt abdominal trauma is occasionally found; this can lead to both true aneurysms and pseudoaneurysms of the abdominal aorta. Most studies have shown that AAAs are approximately 6 to 7 times more common in smokers, whereas elevated high-density lipoprotein cholesterol (HDL-C) may be protective.¹ Furthermore, 23% of patients with a AAA have some form of cancer.⁷ Consistent with these broad comorbidities, approximately 67% of patients with AAAs die of a cardiovascular etiology.

The aneurysm growth rate, wall tension, and risk of rupture are dependent on aneurysm diameter. Aneurysms less than 5.5 cm in diameter grow at a median rate of 0.2 to 0.3 cm per year, whereas those greater than 5 to 6 cm expand at rates up to 3 cm per year.⁸ The strongest predictor of AAA rupture is aneurysm size. Aneurysms between 5.5 and 5.9 cm in diameter have an estimated annual risk of rupture of 9.4%, those between 6.0 and 6.9 cm in diameter have an estimated annual risk of rupture of 10.2%, and those greater than 7 cm have an estimated annual rate of rupture of 32.5%.⁹ Eccentric or saccular aneurysms have considerably increased wall stress and are thought to have greater rupture risk than fusiform AAA. Patients with chronic obstructive pulmonary disease or asthma may also have a higher risk of rupture, possibly related to allergic inflammation and/or the concomitant use of bronchodilators and steroids.¹⁰ Patient age and family history of AAA are not consistent independent predictors of rupture.

True aortic aneurysms result from the dilation of the intima, media, and adventitial layers of the aorta. Relatively uniform and concentric enlargement results in fusiform aneurysm formation. Saccular aneurysms result from eccentric enlargement of the aorta. As previously noted, saccular aneurysms have less uniform wall stress distribution and thus an increased risk of rupture.

Although traditionally attributed to atherosclerosis, AAAs more accurately can be attributed to a degenerative process in the arterial wall. Most AAAs result from the imbalance of connective tissue formation and degradation. Some congenital disorders such as Marfan syndrome and Ehlers-Danlos syndrome predispose to aneurysm development. Inflammatory aortitis is a rare cause of AAA formation. Pseudoaneurysms generally result from abdominal trauma or may occur at a graft anastomosis following previous aortic surgery.

Inflammatory AAAs are a subset (5%-10%) of AAAs, which commonly present as abdominal or flank pain, fever, and constitutional symptoms.¹¹ On computed tomography (CT), these AAAs usually have a “halo” sign with inflammatory adventitial tissue predominately anterior and lateral to the AAA. Due to the extensive inflammation and fibrosis of surrounding structures, including the duodenum, vena cava, left renal vein, and ureters, an endovascular approach may be preferred.¹²

Mycotic AAAs are fortunately rare. Most commonly, Salmonella species or Staphylococcus aureus are the etiologic bacteria, although fungal species or other bacteria are sometimes causative. Infected AAAs pose a significant risk to the patient and present many therapeutic challenges. The traditional surgical approach is ligation of the aorta, resection of the infected aneurysm, and extra-anatomic bypass. Antibiotic-soaked grafts or homografts may play a role. There have been case reports and small series involving the endovascular treatment of mycotic thoracic and AAAs. Although successful outcomes have been demonstrated in a few cases, the potential for infection of the endoprosthesis raises significant concerns regarding this treatment approach.¹³ Aortoenteric fistulae from AAAs have also been successfully excluded using an endograft.¹⁴

Although the basic understanding of the inflammatory and dynamic nature of AAAs has significantly advanced, there is as of yet no effective medical treatment for limiting the expansion of AAA. Based on the premise that aneurysmal growth is dependent in part on the δP/δt of the aortic pulsation, β-blockers have been studied in randomized trials but have not shown any reduction in AAA growth rate or mortality.¹⁵ Angiotensin-converting enzyme (ACE) inhibitors have been shown to reduce the rate of aneurysm expansion in animal models. Some observational studies have suggested an association between ACE inhibitors and a reduced risk of aneurysm rupture,¹⁶ whereas other studies have suggested accelerated aneurysm expansion with ACE inhibitors.¹⁷ Two randomized trials of ACE inhibitors or angiotensin receptor blockers in small aneurysms are currently examining this question.^18,19

Efforts to reduce the activity of the matrix metalloproteinase (MMP) enzymes with doxycycline in patients with AAA have provided mixed results. Small pilot studies demonstrated that doxycycline reduces the levels of MMP-9 and C-reactive protein (CRP).^20,21 However, a large clinical trial of doxycycline did not show any benefit in reduction of aneurysm expansion.²² Similarly, statin medications have been studied for a possible benefit in stabilization of aneurysm size by pleiotropic effects on reduction of MMP activity. Although observational studies have suggested a possible benefit of statin therapy in reducing aneurysm expansion, meta-analyses have not found a consistent benefit.²³

Despite the current lack of an adequate medical treatment for slowing the rate of AAA expansion, all patients with AAA should receive appropriate medical therapy for cardiovascular risk reduction, including smoking cessation, statin therapy, and antiplatelet therapy. Smoking cessation is associated with improved mortality after AAA repair, and smoking cessation may slow the expansion of a preexisting aneurysm.²⁴ Treatment with a statin is associated with improved long-term survival after AAA repair, likely due to a reduction in cardiovascular mortality.^25,26 Antiplatelet therapy may limit thrombus formation within the aneurysm sac, and aspirin therapy is recommended for all patients with AAA for reduction of other cardiovascular events.

The only proven therapy for reduction in mortality from AAA rupture is mechanical aneurysm exclusion or sealing. Elective open AAA repair is associated with a mortality of 2.7% to 7% and may be higher depending on patient comorbidities. Thus, the clinician is faced with balancing the interventional risk of AAA repair with that of continued observation and threat of rupture.

Several trials have been performed to determine the size of an AAA that optimizes the trade-off between risk of AAA rupture and perioperative mortality. These studies have suggested that aneurysms less than 5.5 cm that grow at rates of less than 0.7 cm per 6 months or 1 cm per 12 months may be safely observed. Based on the results of the United Kingdom Small Aneurysm Trial and the Aneurysm Detection and Management Trial, surgery can be reserved for aneurysms larger than 5.5 cm without increasing overall mortality or operative risk.^27,28 However, these studies are prefaced on routine AAA surveillance and close patient follow-up and were performed with open surgical repair rather than widespread availability of endovascular techniques for exclusion of the aneurysm. However, one trial (the Positive Impact of Endovascular Options for Treating Aneurysms Early [PIVOTAL]) found no difference in outcomes between endovascular repair and ultrasound surveillance among patients with aneurysms measuring 4 to 5 cm.²⁹

Despite the equivalent overall survival between surveillance and early repair for aneurysms less than 5.5 cm, caution must be used in extending these criteria to underrepresented study groups. Because women have smaller abdominal aortas than men, the criteria for diagnosis of an AAA as well as for intervention for an AAA may not be accurately applied to both genders. Some studies have suggested that a 5.0-cm diameter AAA in women has an equivalent rupture risk to that of a 6.0-cm diameter AAA in men.³⁰ Therefore, the application of a single measurement that is not gender specific or referenced to the patient’s height may not be entirely valid.

Patients with symptomatic AAAs usually present with abdominal pain that may radiate to the flank or groin, abdominal tenderness to palpation, low back pain, or hemodynamic instability related to the retroperitoneal hemorrhage. In addition to the risk of rupture, aneurysms can be associated with a significant risk of embolic complications. Due to the presence of thrombus or atherosclerotic material within the aneurysm sac, patients may present with evidence of atheroemboli (ie, livedo reticularis), acute renal failure, “blue toe syndrome,” or symptoms of lower extremity arterial insufficiency. Enlargement of the AAA can also cause a mass effect and obstructive symptoms. Obstructive uropathy can occur and is thought to be more common in inflammatory AAAs.

Contained or frank rupture of an AAA demands emergency intervention. Classically, there is severe pain, abdominal tenderness, and shock. The pain is generally sudden in onset and constant, and it may diffusely involve the abdomen or radiate to the flanks, groin, or legs. Occasionally, a ruptured AAA presents as a gastrointestinal hemorrhage or aortocaval fistula. Expeditious diagnosis with an imaging study (usually CT) or direct transfer to the interventional suite or operating room is necessary to reduce the attendant morbidity and mortality of a ruptured AAA.

In summary, the criteria for intervention on an AAA include an aneurysm diameter greater than 5.5 cm in men (perhaps 5.0 cm in women), rapid expansion of an aneurysm (>0.5-0.7 cm/6 months or 1 cm/12 months), or symptomatic aneurysm with either distal embolization from an AAA, aneurysmal mass effect, threatened rupture, or frank rupture.

The goal of EVAR is to prevent aneurysm rupture and reduce the morbidity and mortality associated with exclusion of the AAA. Successful EVAR exclusion is associated with a progressive reduction in aneurysmal volume at a rate of approximately 3 mL per month and reduction in aneurysm diameter of 4 to 6 mm per year.³¹

Four randomized trials have compared the outcomes of EVAR to open AAA repair (Table 56-1).^32-35 Each of these trials enrolled patients with asymptomatic AAA who were candidates for endovascular or open repair. All 4 studies reported lower 30-day mortality with EVAR, with 30-day mortality rates ranging from 0.5% to 1.2% for EVAR and 1.3% to 4.7% for open surgery. A subsequent meta-analysis found that EVAR is associated with significantly lower perioperative mortality compared with open surgical repair.³⁶ The long-term mortality in each of the studies did not demonstrate a benefit of EVAR over open AAA repair.^37,38 These findings may be a result of several factors, most significantly the competing risk of other comorbidities. However, concern has also been raised regarding the long-term implications of placing an endograft, as studies with longer-term follow-up have demonstrated an increased rate of late (>5 years) need for re-intervention and late graft failure. As a result, longer-term surveillance studies remain necessary to confirm the safety and long-term effectiveness of new devices.

Approximately 20% to 40% of patients (approximately 10%-15% per year) undergoing EVAR require subsequent re-intervention for a variety of indications, including endoleak, stent migration, graft limb occlusion, or infection.³⁹ Late conversion of EVAR to open repair is necessary in approximately 2% to 3% of cases, although the frequency of late conversion may be higher in patients with larger AAAs. Open conversion often requires a more extensive operation in a higher risk patient population and thus carries a 10% to 20% mortality rate. The incidence of late rupture of an AAA after EVAR is thought to be 0.5% at 3 to 4 years and is perhaps more frequent after use of a tube graft. Rupture after EVAR carries a mortality rate similar to native AAA rupture.⁴⁰

Despite improvements in identification and protocols for immediate treatment, ruptured AAA remains associated with an in-hospital mortality rate of 30% to 50%. Because EVAR is less invasive and may be performed using local anesthesia, the application of EVAR to acutely ill patients with ruptured AAA may reduce in-hospital mortality and minimize complications. A frequently used technique that may also improve outcomes is the immediate percutaneous placement of an aortic occlusion balloon; placement of such a balloon provides time for resuscitation efforts while a definitive decision can be made regarding EVAR versus open surgical repair.⁴¹

A number of observational studies have suggested a mortality benefit of EVAR over open surgery for patients with ruptured AAA. These studies have estimated mortality rates with EVAR of 16% to 30% and of 34% to 44% for open surgery.⁴² However, criticisms of such studies include the inherent selection bias of EVAR for less critically ill patients and patients with anatomically favorable characteristics.

Compared to the observational literature, randomized studies have not demonstrated a benefit of EVAR over open surgery for ruptured AAA. The largest of these studies was the multicenter IMPROVE trial, which was conducted in the United Kingdom and Canada.^43,44 In that study, 613 patients with a presumed ruptured AAA were randomized to a strategy of EVAR versus open surgical repair. At 30 days and 1 year, there was no difference in mortality between the 2 groups. However, several aspects of the study design should be considered. Most importantly, randomization was performed prior to imaging, and as a result, only 64% of patients were felt to have anatomy suitable for EVAR. Additionally, 8.9% of patients were found based on imaging and other evaluation to have a diagnosis other than ruptured AAA. In an as-treated analysis, patients treated with EVAR had numerically lower mortality (25% vs 38%). An additional analysis found that patients treated with local anesthesia had significantly lower 30-day mortality.⁴⁵

Based on the available data, rapid imaging can help identify patients with ruptured AAA who are anatomically appropriate for EVAR, and such patients should be approached with an “endovascular first” strategy using local anesthesia whenever feasible. This approach may help minimize the hemodynamic stress associated with anesthesia and also minimize the postoperative complications of open surgical repair. Recent analyses using large registry data have suggested that the majority of benefit for EVAR in ruptured AAA is in low- to moderate-risk patients, whereas open surgery and EVAR have a similar benefit in high-risk ruptures.⁴⁶ However, several barriers remain to implementation of such a strategy outside of specialized centers, including the ability to have EVAR available for unexpected anatomy and rapid availability of an endovascular suite.

Aneurysm morphology and the status of the iliac and femoral access vessels dictate whether a patient is a candidate for EVAR. Historically, only 50% to 60% of patients with infrarenal AAAs had suitable anatomy for an endograft. Limitations of EVAR include the length of the infrarenal aortic neck; the diameter of the aneurysm neck; the location of the mesenteric arteries, particularly the celiac axis and superior mesenteric artery; and the proximal neck angulation. Distally, anatomic concerns include the size of the iliofemoral arteries, the presence or absence of calcification, the tortuosity of these arteries, and whether there is aneurysmal dilation of the iliac arteries. Successful exclusion of a common iliac artery aneurysm may require embolization of the internal iliac artery and extension of the endograft into the external iliac artery. In recent years, the development of newer endografts and branched and fenestrated endografts has addressed many of these technical limitations in the endovascular treatment of AAA, and as a result, approximately 80% of AAAs are treated with EVAR.

Aneurysm Neck

The aneurysm neck is the length from the most inferior (main) renal artery to the onset of the aneurysm. Most endografts that are commercially available in the United States are designed for suprarenal fixation, but an adequate seal zone is necessary to exclude the aneurysm below the renal arteries. The maximal diameter of the nonaneurysmal aortic neck that can be sealed is endograft dependent. The device should be oversized by 10% to 20%. The minimal neck length tolerated is also device specific, although generally for infrarenal fixation, the neck length should ideally be at least 15 mm, especially if angulated (Fig. 56-2). An exception to this is the Ovation device from TriVascular (Santa Rosa, CA), which, due to its novel sealing mechanism, is indicated for EVAR regardless of aneurysm neck length. The angle between the suprarenal aorta and the infrarenal neck (α) and the angle of the flow axis between the infrarenal neck and the body of the aneurysm (β), in which the flow axis is the axis (line) between the distal neck/proximal aneurysm to the aortic bifurcation, should also be measured, because extreme neck angulation may limit successful aneurysm exclusion (Fig. 56-3). Calcification and thrombus of the proximal neck should also be noted, because both of these incrementally add to the morphologic complexity of EVAR.

In summary, the neck length should ideally be greater than 15 mm, the neck diameter should be less than 32 mm, α or β should be greater than 150°, and the degree of calcification or thrombus greater than 2 mm thick should be less than 25% of the neck circumference. The risk of adverse outcome after EVAR is directly related to the proximal aortic neck angulation. While modern endografts have in large part addressed the anatomic limitations, studies investigating the on-label versus off-label use of endografts have found that off-label use is associated with a higher risk of long-term complications and need for re-intervention.⁴⁷

Arterial Branches of the Abdominal Aorta

To anticipate and prevent type II endoleak, it is also important to assess for branch arteries off the aneurysm. The lack of patent lumbar arteries, inferior mesenteric artery, or accessory renal artery off the aneurysm sac should reduce the chance of type II endoleak. However, in current practice, these branches are excluded even if patent, and pre-intervention embolization is not generally considered necessary.

Iliac Arteries

The anatomy of the iliac arteries is important with regard to the deliverability and proper sealing of endografts. Preoperative assessment should include measurements of the diameter of the common and external iliac arteries and length of the iliac landing zone. The external iliac artery generally needs to be at least 6 to 6.5 mm in diameter to allow passage of the majority of commercially available devices. The iliac arteries should also be evaluated for the presence of ectasia, stenosis, calcification, and tortuosity. Aortoiliac tortuosity and calcification affect device deliverability and are associated with a more complex EVAR procedure. Maintenance of patency of at least 1 of the internal iliac (hypogastric) arteries is important to preserve pelvic blood flow and to prevent buttock claudication and intestinal ischemia (see below).

In 1999, the AneuRx device (Medtronic, Santa Rosa, CA) and the AnCure device (Guidant Endovascular Solutions, Menlo Park, CA) were approved by the US Food and Drug Administration (FDA) for clinical use and harkened in the era of endovascular aortic aneurysm repair. The AneuRx device was a modular, bifurcated stent graft with thin, polyester graft material fully supported by a nitinol stent framework.^48,49 This device was later supplanted by newer stent graft designs from Medtronic (Talent, Endurant) and is no longer in clinical use in the United States. The AnCure stent graft was a unibody, bifurcated polyester graft that was not fully supported by stent. More closely approximating the traditional surgical bifurcated graft, the AnCure device had a nitinol stent ring at the proximal and distal attachment sites with active fixation barbs.⁵⁰ While early studies demonstrated efficacy of this device, excessive perioperative complications with AnCure led to its withdrawal from the market in 2003.

Several EVAR devices have since been approved by the FDA for the treatment of infrarenal AAA. Each device has unique features that may provide benefit in certain anatomic substrates (Table 56-2). Many of these devices incorporate a suprarenal bare stent (with or without attachment hooks/barbs) to provide secure proximal fixation and to reduce the risk of device migration. A wide array of device diameters and lengths along with aortic and iliac extender cuffs are available to customize these devices for complex and varied patient anatomy. Over the years, devices with larger aortic and iliac diameters have been approved to expand the percentage of patients who can be treated with EVAR. More recently, a fenestrated stent graft was FDA approved to facilitate treatment of patients with juxtarenal aneurysms.⁵¹ Branch grafts have also become available to help maintain patency of the internal iliac artery when there is extension of aneurysm to the iliac bifurcation.⁵² Continued evolution of device design has allowed for lower profile, more flexible delivery systems that can be delivered through the iliac arteries more easily, thereby reducing the risk of iliac rupture or dissection. Due to the advent of these lower profile devices, EVAR can now be performed percutaneously in a large percentage of cases.

The Excluder stent graft (Gore and Associates, Flagstaff, AZ) was FDA approved in 2002.⁵³ The Excluder device is a modular, bifurcated endoprosthesis with expanded polytetrafluoroethylene (ePTFE) graft material fully supported by nitinol stent rings. This device does not have suprarenal fixation but has an ePTFE sealing cuff and proximal nitinol anchors. A lower permeability graft material was introduced in 2004. The C3 delivery system was introduced in 2010 for more precise and controlled deployment of the device. The latest version of the Excluder is available in aortic diameters up to 35 mm and iliac limb diameters up to 27 mm. The 23- and 26-mm devices can be delivered through a 16-Fr sheath.

The Zenith stent graft (Cook Medical, Bloomington, IN) was approved by the FDA in 2003.⁵⁴ The Zenith device is a modular, 3-piece, bifurcated endoprosthesis with woven polyester graft material fully supported by stainless steel stents. This device is distinguished by bare metal stent suprarenal fixation and proximal retention hooks. The newest version (Zenith Flex) incorporates a hydrophilic delivery sheath with trigger-wire release mechanism for precise placement. Main body diameters up to 36 mm and iliac diameters up to 24 mm are available. Smaller devices (up to 26 mm) can be delivered via an 18-Fr delivery system. An aorto-uni-iliac version of this device (Renu Ancillary Graft Converter) is available and may facilitate secondary endovascular intervention in patients who have received prior endovascular repair. The Zenith Fenestrated EVAR device was more recently approved by the FDA for treatment of challenging neck anatomy that is unsuitable for a conventional, nonfenestrated graft. The Zenith Fenestrated device incorporates up to 3 holes (fenestrations) and cutouts from the proximal margin of the graft material to maintain patency of the renal arteries and superior mesenteric artery when sealing with the stent graft across the juxtarenal segment of the aorta.

The Powerlink stent graft (Endologix, Irvine, CA) was approved by the FDA in 2004.⁵⁵ This is a unibody, bifurcated endoprosthesis with ePTFE graft material fully supported by a cobalt-chromium stent. The most recent iteration of this device is the AFX endovascular system. The AFX device is differentiated from other EVAR devices by its unique delivery mechanism, unibody design, and method of fixation. Fixation is not achieved by the usual mechanism of suprarenal bare metal stent and attachment hooks but rather is achieved by buttressing the bifurcation of the AFX device at the native aortoiliac bifurcation. Maintaining the native aortoiliac bifurcation has the advantage of preserving options for future contralateral access for the treatment of lower extremity occlusive disease. The AFX device is available in aortic diameters up to 32 mm and iliac diameters up to 20 mm. The main body can be delivered via a 17-Fr delivery sheath.

The Talent stent graft (Medtronic) was approved by the FDA in 2008.⁵⁶ The Talent device is a modular, bifurcated design consisting of polyester graft material fully supported by nitinol stent rings and full-length connecting bar. There is a bare metal proximal stent for suprarenal fixation. This device was supplanted by the next-generation stent graft from Medtronic, the Endurant, which was FDA approved in 2010.^57,58 The Endurant is also a 2-piece, modular, bifurcation stent graft with proximal bare metal stent and anchor pins from suprarenal fixation. The most recent iteration of this platform is the Endurant II stent graft system, which is available in 2-piece and 3-piece modular configurations (Endurant IIs). The Endurant II device is available in aortic diameters up to 36 mm and iliac diameters up to 28 mm. Main body diameters up to 28 mm can be delivered via an 18-Fr delivery system. An aorto-uni-iliac version of the device is also available.