Endovascular repair of abdominal aortic aneurysm

INTRODUCTION — Endovascular aneurysm repair (EVAR) is an important advance in the treatment of abdominal aortic aneurysm (AAA). EVAR is performed by inserting graft components that are folded and compressed within a delivery sheath through the lumen of an access vessel, usually the common femoral artery. Upon deployment, the endograft expands, contacting the aortic wall proximally and iliac vessels distally to exclude the aortic aneurysm sac from aortic blood flow and pressure (figure 1).

Compared with open AAA repair, EVAR is associated with a significant reduction in perioperative mortality, primarily because EVAR does not require operative exposure of the aorta or aortic clamping. Since the approval of endograft devices for use in the United States, there has been a 600 percent increase in the annual number of EVAR procedures performed, with EVAR accounting for nearly half of AAA repairs. Concurrent with the increased use of EVAR, a decrease in the incidence of ruptured AAA and associated morbidity and mortality has been reported in the United States, likely due to the ability to offer EVAR to patients who would not otherwise be candidates for open surgical repair [1,2].

Endovascular repair of abdominal aortic aneurysm is reviewed here. General issues regarding the management of abdominal aortic aneurysm, and the clinical features and diagnosis of this condition, are presented separately. (See "Management of asymptomatic abdominal aortic aneurysm".)

ANATOMIC CONSIDERATIONS — The abdominal aorta is the most common site of arterial aneurysm. The abdominal aorta is defined as aneurysmal when a localized dilation is identified and the diameter of the dilated region is increased more than 50 percent relative to normal aortic diameter [3]. The normal diameter of the aorta at the level of the renal arteries is approximately 2.0 cm (range 1.4 to 3.0 cm). An aortic diameter greater than 3.0 cm is considered aneurysmal for most individuals.

Aortoiliac anatomy — The abdominal aorta is a retroperitoneal structure that begins at the hiatus of the diaphragm and extends to its bifurcation into the common iliac arteries at the level of the fourth lumbar vertebra (figure 2). It lies slightly left of the midline to accommodate the inferior vena cava, which is in close apposition. The branches of the aorta (superior to inferior) include the left and right inferior phrenic arteries, left and right middle suprarenal arteries, the celiac axis, superior mesenteric artery, left and right renal arteries, possible accessory renal arteries, left and right gonadal arteries, inferior mesenteric artery, left and right common iliac artery, middle sacral artery, and the paired lumbar arteries (L1-L4).

The common iliac artery bifurcates into the external iliac and internal iliac arteries at the pelvic inlet (figure 3). The internal iliac artery, also termed the hypogastric artery, gives off anterior and posterior branches to the pelvic viscera and also supplies the musculature of the pelvis. The external iliac artery passes beneath the inguinal ligament to become the common femoral artery [4].

Aneurysm extent — Abdominal aortic aneurysms (AAAs) are described as infrarenal, juxtarenal (pararenal), or suprarenal depending upon the involvement of the renal or visceral vessels.

●Infrarenal – Aneurysm originates below the renal arteries

●Juxtarenal – Aneurysm originates at the level of the renal arteries

●Suprarenal – Aneurysm originates above the renal arteries

AAAs most often occur in the segment of aorta between the renal and inferior mesenteric arteries; approximately 5 percent involve the renal or visceral arteries (figure 4).

Up to 40 percent of AAAs are associated with iliac artery aneurysm(s). The normal size and diameter corresponding to aneurysm for the iliac artery is discussed elsewhere. (See "Iliac artery aneurysm", section on 'Definition of iliac aneurysm'.)

Although the majority of endovascular aneurysm repairs are performed on aneurysms affecting the infrarenal aorta and iliac arteries, endovascular repair of juxtarenal and suprarenal aneurysms has been performed using advanced endovascular techniques with specialized investigational endografts. (See 'Advanced devices and techniques' below.)

Measurement definitions — Several aortic measurements are important for determining the feasibility of endovascular aneurysm repair and for endograft sizing. The definitions of important terms are as follows [5]:

●Aortic neck diameter – The aortic diameter at the lowest renal artery.

●Aortic neck length – The distance from the lowest renal artery to the origin of the aneurysm.

●Aortic neck angulation – The angle formed between points connecting the lowest renal artery, the origin of the aneurysm, and the aortic bifurcation.

●Conical/reverse tapered aortic neck – A conical neck is present when the diameter of the aorta 15 mm below the lowest renal artery is ≥10 percent larger than the diameter of the aorta at the lowest renal artery.

●Infrarenal aortic length – The distance from the lowest renal artery to the aortic bifurcation.

Other measurements that are important for sizing endografts include the maximal common iliac artery diameter, minimum external iliac artery diameter, distance from the aortic neck to the iliac bifurcation, and maximal AAA sac diameter. The criteria defining anatomic suitability for endovascular aneurysm repair are discussed below. (See 'Anatomic suitability' below.)

Renal artery anomalies — Accessory renal arteries are present in up to 30 percent of the population and commonly originate from the lumbar aorta [6]. Exclusion of an accessory renal vessel by an endograft can lead to partial renal infarction.

Horseshoe kidney, a renal fusion abnormality, is associated with anomalous arterial supply with multiple renal arteries that may be derived from the aorta, or the iliac (common, external, or internal) or middle sacral arteries. Renal arteries often arise from the aneurysmal aortic segment. (See "Renal ectopic and fusion anomalies".)

AORTOILIAC IMAGING — Prior to consideration for endovascular aneurysm repair, aortoiliac imaging is needed to define the anatomy, determine the feasibility of endovascular repair, and choose the size and configuration of endograft components. Computed tomography (CT) is typically used for elective abdominal aortic aneurysm (AAA) repair, but under urgent or emergent circumstances, endograft feasibility and sizing can be determined intraoperatively using arteriography. (See "Surgical and endovascular repair of ruptured abdominal aortic aneurysm", section on 'Endovascular repair'.)

CT angiography of the abdomen and pelvis with ≤2.5 mm cuts with three-dimensional (3D) reconstruction is obtained. Although two-dimensional (2D) CT images can be used, measurement errors (aortic diameter, aortic length) can occur due to volume averaging. In addition, aortic diameter measurements will be overestimated if the aorta is angulated and the longitudinal axis is not perpendicular to the imaging plane. CT angiography with 3D reconstruction allows measurements to be made that are perpendicular to the true axis of the aorta (image 1). Centerline length measurements can also be obtained with 3D reconstruction (image 2). 3D length measurements are more accurate than 2D measurements and can improve graft sizing, particularly in patients with tortuous vessels [7].

Magnetic resonance (MR) angiography can be used for preoperative endograft planning preoperatively; however, gadolinium administration in the setting of renal dysfunction is a relative contraindication.

The use of digital subtraction arteriography (DSA) to obtain aortic measurements is limited by measurement errors due to parallax and magnification. Since the inner lumen is imaged, but not the wall of the aorta, DSA cannot evaluate the true lumen diameter, extent of thrombus, plaque, or degree of calcification. Errors in length measurement can occur when intraluminal catheters that follow the shortest distance around the curves of the aorta are used. Arteriography can be used in emergent situations to estimate the proximal aortic diameter and the diameter of the iliac graft landing zones when treating a ruptured AAA with EVAR, but it is otherwise not recommended as a routine pre-EVAR imaging modality. (See "Surgical and endovascular repair of ruptured abdominal aortic aneurysm", section on 'Endovascular repair'.)

Surface duplex ultrasound is not an adequate imaging modality for determining feasibility or planning EVAR. On the other hand, intravascular ultrasound (IVUS), which provides accurate intraoperative diameter and length measurements (based on branch vessel location), can be used as the sole imaging modality during EVAR and is a particularly attractive imaging alternative for patients with renal insufficiency. IVUS is best used during EVAR and not as a preoperative imaging modality given its invasive nature but has limited availability and requires significant skill and experience to perform and interpret.

INDICATIONS FOR REPAIR — Repair of abdominal aortic aneurysm (AAA) is indicated for patients with AAA who are symptomatic (tenderness or abdominal or back pain, evidence for embolization, rupture), have an AAA ≥5.5 cm, or an AAA that has expanded by more than 0.5 cm within a six-month interval. The clinical evaluation and management of AAA are discussed in detail elsewhere. (See "Clinical features and diagnosis of abdominal aortic aneurysm" and "Management of asymptomatic abdominal aortic aneurysm".)

Anatomic suitability — Anatomic suitability is the most important determinant for successful endovascular aneurysm repair (EVAR) in the long term. With early endograft designs, approximately 50 percent of patients were not candidates for EVAR because of the site, extent, or morphology of the aneurysm, or unsuitability of access vessels. The availability of devices that allow for shorter proximal seal zones and lower profile devices has expanded the use of EVAR to almost two thirds of patients with an infrarenal AAA.

To exclude blood flow from the aneurysm sac, the endograft must provide an adequate seal where the endograft contacts the arterial wall proximally at the aortic neck and distally in each of the iliac arteries, otherwise known as the landing zones. The security of the repair relies solely upon the radial force generated by the graft at the landing zones since endografts do not have any suture-mediated stability. Thus, certain anatomic criteria must be fulfilled to perform EVAR [8]. These criteria include measurements of the aortic neck, iliac, and femoral arteries (figure 5). Definitions for the measurement terms are described above, and the criteria for each are described in detail below. (See 'Measurement definitions' above.)

Characteristics of currently available endografts and potential advantages for different anatomic situations are given in the table (table 1). The specific criteria recommended for a specific device are given in the instructions for use that are published and packaged with each device used. Failure to follow device specifications increases the risk for complications. (See 'Perioperative morbidity and mortality' below.)

Aortic neck diameter — The required endograft diameter is determined by measuring the aortic neck diameter (eg, 20 mm) and adding an additional 15 to 20 percent of the aortic neck diameter (20 mm + 3 to 4 mm = 23 to 24 mm). Undersizing the diameter of the endograft will lead to an inadequate seal and failure to exclude the aneurysm. Oversizing the endograft 15 to 20 percent over the measured aortic neck diameter should provide sufficient radial force to prevent device migration. Commercially available devices have endograft diameters as large as 36 mm, which allow endovascular repair of aneurysms to a maximal aortic neck diameter of 32 mm. However, the durability of EVAR fixation in an aneurysmal aortic neck (ie, ≥30 mm) remains to be determined in large series with long-term follow-up [9,10]. One series evaluated the two-year clinical and morphologic outcomes of 188 patients two years after EVAR for AAA with large (≥28 mm) infrarenal neck [10]. Such large necks were associated with infrarenal aortic neck enlargement and a high risk of proximal type I endoleak and proximal neck-related reinterventions. The authors suggested that in this subgroup of patients, main body oversizing >15 percent and suprarenal sealing should be considered. However, oversizing the endograft can result in incomplete expansion of the endograft with infolding and inadequate seal and can be associated with intermediate and long-term neck expansion [11]. Oversizing can also lead to kinking of the device, which can form a nidus for thrombus formation or can also lead to endoleak (ie, blood outside the endograft (table 2)).

Aortic neck length — The aortic neck length should be at least 10 mm to provide an adequate proximal landing zone for endograft fixation.

Qualitative assessment of the proximal neck is also important. Ideally, the proximal aorta should be normal in appearance, without significant thrombus or calcification. Although not an absolute contraindication for EVAR, large amounts of thrombus or calcification will interfere with fixation of the graft and increase the risk for graft migration or type I endoleak (table 2). (See 'Handling immediate endoleak' below.)

Aortic neck angulation — Ideally, the aortic neck angle should be less than 60º. Angles that are greater lead to difficulties in implantation, kinking, endoleak, and the potential for distal device migration. Severe angulation (>60°) is generally considered to be a contraindication to EVAR, although investigational devices are under trial that are more conformable and may be useful in such anatomic situations. However, the ability to place a device in aneurysms with significant angulation at the neck is ultimately determined by the conformability of the specific device type and its delivery characteristics (table 1). The specific characteristics of individual devices are discussed in detail elsewhere. (See "Endovascular devices for abdominal aortic repair".)

Iliac artery and access vessel morphology — Suitable iliac artery morphology is also required for endograft placement. The iliac arteries should have a minimal amount of calcification and tortuosity, and no significant stenosis or mural thrombus should be present in the distal graft landing zones. The common iliac artery is the preferred distal attachment site, but the external iliac artery can also be used. When the external iliac artery is used for distal fixation (eg, common iliac artery aneurysm), the origin of the hypogastric artery (ie, internal iliac artery) is covered by the endograft.

While an attempt at hypogastric preservation should be made to maintain pelvic perfusion, hypogastric embolization may be required prior to endograft placement to prevent backbleeding into the aneurysm sac. This may not be necessary if there is severe stenosis at its origin since stent-graft coverage across its origin will allow it to thrombose and prevent endoleak formation. With the availability of approved iliac branch devices, which allow the preservation of one or both hypogastric arteries at the time of EVAR (image 3), hypogastric embolization is less often necessary to extend the EVAR seal zone. (See "Surgical and endovascular repair of iliac artery aneurysm", section on 'Associated abdominal aortic aneurysm'.)

A minimal external iliac artery diameter of 7 mm is needed to allow safe passage of the majority of endograft delivery sheaths. However, low-profile devices can be introduced through access vessels of <6 mm diameter. The common iliac artery diameter should measure between 8 and 22 mm, and the length of normal diameter common iliac artery into which the limbs of the endograft will be fixed should measure at least 15 to 20 mm to achieve an adequate seal.

Focal narrowing and mild angulation can be overcome with standard guidewire and catheter techniques, while diffuse narrowing or significant calcification is more problematic. If the device fails to pass, an iliac conduit can be created [12]. (See 'Iliac conduits' below.)

Aneurysmal iliac arteries (>22 mm) are treated by excluding them. The management of iliac artery aneurysm in conjunction with abdominal aortic aneurysm is discussed elsewhere. (See "Surgical and endovascular repair of iliac artery aneurysm", section on 'Associated abdominal aortic aneurysm'.)

Other considerations — The inferior mesenteric artery is often occluded by thrombus in patients with an abdominal aortic aneurysm, and in these cases, excluding the inferior mesenteric artery with an endograft will be of no consequence. However, if the inferior mesenteric artery is patent but associated with significant stenosis of the superior mesenteric artery, the inferior mesenteric artery may provide important collateral blood flow to the intestine. Under this circumstance, covering the origin of a patent inferior mesenteric artery with an endograft may lead to intestinal ischemia [12].

A visceral hybrid procedure combines endovascular exclusion of the aneurysm with open visceral revascularization, and is more commonly used for the endovascular repair of thoracoabdominal aortic aneurysm (TAA), but in some circumstances may be needed in patients with AAA who have visceral artery disease. Prior to endovascular repair, retrograde revascularization of the visceral and renal arteries is performed (surgical debranching) using an open approach, allowing subsequent coverage of the origins of these vessels during endovascular stent-graft placement [13-15].

Contraindications — Endovascular repair of AAA is contraindicated in patients who do not meet the anatomic criteria required to place any of the available endografts (table 1). Adverse anatomic features include suprarenal or juxtarenal AAA, small-caliber vessels, circumferential aortic calcification, and extensive tortuosity. Depending upon the location of the main and accessory renal arteries, endovascular repair may also be contraindicated for the management of AAA associated with horseshoe kidney. A variety of next-generation devices are being developed to treat suprarenal and juxtarenal abdominal aortic aneurysms. (See 'Anatomic suitability' above and 'Endografts' below and "Endovascular devices for abdominal aortic repair".)

A relative contraindication to EVAR is the inability to comply with the required post-EVAR surveillance. (See 'Endograft surveillance' below.)

Whether younger patients (<60 years of age) who are not at high risk for open surgery should undergo open surgical repair versus EVAR remains controversial. Surveillance over an extended period of time exposes the patient to greater levels of cumulative radiation, and EVAR does not completely eliminate the risk of future aortic rupture. Guidelines from major medical and surgical societies emphasize an individualized approach when choosing endovascular repair, taking into account the patient's age and risk factors for perioperative morbidity and mortality [16-18]. The decision for open surgical repair versus endovascular repair is reviewed elsewhere. (See "Management of asymptomatic abdominal aortic aneurysm", section on 'Open versus endovascular aneurysm repair' and 'Preoperative risk assessment' below.)

ENDOGRAFTS — Endovascular repair of abdominal aortic aneurysm is accomplished using a fabric-covered stent, termed an endograft or stent-graft. The first endovascular aneurysm repair (EVAR) was performed in 1987 by Nicholas Volodos in Kiev [19]. Juan Parodi popularized stent-grafting following his initial clinical experience in 1991 and was the driving force for commercializing stent-graft systems [20]. Stent-grafts were approved for clinical use in the United States in 1999.

Adoption of stent-graft technology by vascular surgeons has been rapid. EVAR has become a preferred approach across all ages for the repair of infrarenal abdominal aortic aneurysm (AAA) due to decreased perioperative morbidity and mortality [1,2,21].

Stent-graft design — Eight stent-graft systems are currently available in the United States. These are listed below and discussed in detail elsewhere. (See "Endovascular devices for abdominal aortic repair".)

●Zenith (see "Endovascular devices for abdominal aortic repair", section on 'Zenith Flex')

●Excluder (see "Endovascular devices for abdominal aortic repair", section on 'Excluder')

●AFX2 (see "Endovascular devices for abdominal aortic repair", section on 'AFX2')

●Alto (see "Endovascular devices for abdominal aortic repair", section on 'Alto')

●Endurant (see "Endovascular devices for abdominal aortic repair", section on 'Endurant')

●Aorfix (see "Endovascular devices for abdominal aortic repair", section on 'Aorfix')

●Incraft (see "Endovascular devices for abdominal aortic repair", section on 'Incraft')

New endograft designs are continually being tested to enhance performance. Targeted improvements have focused upon lower device and delivery profiles, more accurate deployment, improved fixation systems, and flexibility in managing challenging anatomy. These improvements, along with increased operator experience, have led to improvements in the short-term and long-term results of endovascular aneurysm repair and have expanded the application of endovascular repair to many whose aortic anatomy was previously deemed unsuitable. Careful preoperative sizing and planning, along with strict adherence to device-specific instructions for use, lead to the best outcomes. Other devices that have received approval for use in other countries or are at the clinical trial or preapproval stage in the United States are discussed separately. (See "Endovascular devices for abdominal aortic repair", section on 'Withdrawn/investigational devices'.)

Endovascular grafts for infrarenal aneurysm repair share a bifurcated, modular design. Although there are variations from device to device, three components (delivery system, main body, and iliac extensions) are common to all endograft device systems and are briefly described below.

●Delivery system – The endograft is typically delivered through the femoral artery, either percutaneously or by direct surgical cutdown. If the femoral artery is too small to accommodate the delivery system, access can be obtained by suturing a synthetic graft to the iliac artery (ie, iliac conduit) through a retroperitoneal low abdominal incision. The size of the delivery system varies depending upon the device diameter.

●Main body device – The main body device is usually a bifurcated graft, but unibody grafts are available for use in special circumstances (image 4). Bifurcated-to-unilateral graft conversion kits are also available and effectively turn a bifurcated main body graft into a unibody graft by occluding one of the two proximal iliac limbs. The length of the limb(s) of the main body device varies. Two-component bifurcated grafts have one short and one long iliac limb. Three-component bifurcated devices have two short limbs. Endovascular grafts rely primarily upon tension to maintain the positioning of the main body device. Main body fixation is classified as infrarenal or suprarenal. Devices with infrarenal fixation may have barbs or hooks on the outer aspect of the devices, whereas suprarenal fixation is accomplished with a portion of uncovered graft extending superior to the renal arteries. Devices with suprarenal fixation may also have barbs and hooks.

●Iliac extensions – One or more iliac extension devices are required to complete the endograft construction. Following the deployment of the main body of a two-component system, the attached iliac limb is pulled down and deployed into the ipsilateral iliac artery. A separate iliac device is introduced into the contralateral iliac artery to complete the endograft. For a three-component device, both of the iliac limbs are introduced separately. Additional iliac extensions may be needed to obtain a proper seal in the iliac artery distally. (See 'Endograft placement' below.)

The degree of structural support (nitinol or stainless-steel stents) throughout the graft varies from device to device. The support structure may be internal or external to the graft fabric, which is typically made of polyester (eg, Dacron) or polytetrafluoroethylene (PTFE). Proponents of designs that have less metallic support structure claim the device is better able to adapt to changes in aneurysm configuration over time. On the other hand, some physicians feel that fully supported endografts are less prone to kinking and subsequent thrombosis.

Advanced devices and techniques — When aneurysmal disease is more extensive, involving the visceral vessels proximally or associated with common or hypogastric artery aneurysms, the complexity of the required endovascular or open repair increases. Fenestrated and branched graft technology is under investigation to manage more challenging anatomy without the need for surgical debranching. The early results using these endografts have been promising, with high rates of successful exclusion of juxtarenal and thoracoabdominal aneurysms, but with an increased risk of visceral artery or stent-branch occlusion. The devices and techniques for more advanced endovascular repair are briefly reviewed below and are discussed in more detail elsewhere. (See "Endovascular devices for abdominal aortic repair", section on 'Advanced devices' and "Complications of endovascular abdominal aortic repair", section on 'Ischemic complications'.)

●Fenestrated – Fenestrated endografts have openings in the fabric of the endograft, which allow flow into the visceral arteries. Stent-grafts with fenestrations at the renal arteries can be used when the proximal aortic neck is short (ie, <10 mm) [22,23].

●Branched – Branched grafts have a separate small graft sutured to the basic endovascular graft for deployment into a vessel to preserve flow into it. Branched grafts have been designed to accommodate the hypogastric (ie, internal iliac) and renal arteries.

●Parallel grafting – The parallel (ie, "chimney," "periscope," "snorkel") endografting technique is used to preserve perfusion to branch vessels when an endograft needs to be placed in a suprarenal position to gain additional neck length [24]. Peripheral stents are placed into the branch vessel(s) before the aortic endograft is completely deployed. The branch vessel stent is deployed alongside the endograft (parallel position between the inside of the aortic wall and outside the endograft) [25,26]. Although this technique maintains flow into the visceral vessels, complications related to device insertion and type I endoleak (eg, gutter leak between the visceral stent graft and the aortic graft) (table 2) continue to be a problem. In the absence of available off-the-shelf fenestrated or branched grafts in the United States, chimney grafts remain a feasible endovascular option for high-risk patients who have more extensive abdominal aortic aneurysms, particularly in emergency situations.

Choice of graft — In general, there are no clear advantages of one stent-graft design over another. Some endografts may fare better in specific anatomic situations, such as low-profile devices for patients with small iliac vessels, or, for those with a short aortic neck, devices suited to a less than 15 mm neck length. The overall performance among the available devices is similar, and the available data confirm uniformly low complication rates. (See "Endovascular devices for abdominal aortic repair".)

The choice of a particular device design is based upon multiple factors, including patient anatomy, operator preference, and cost. An endograft system that can handle all types of AAA, including those with angulated or tortuous anatomy, has yet to be achieved.

Bifurcated grafts are most often chosen but are not appropriate for patients with unilateral severe iliac stenosis or occlusion. Under this circumstance, unibody (nonbifurcated) grafts, also known as aorto-uni-iliac (AUI) devices, can be used (image 4). AUI devices are used when contralateral iliac access or gate cannulation is difficult or impossible, and for the treatment of some ruptured aneurysms for expeditious control of hemorrhage. After deployment of an AUI device, a plug will need to be inserted into the contralateral iliac artery if it remains patent to prevent retrograde flow of blood into the aneurysm sac. To provide adequate perfusion to the contralateral lower extremity, a femoro-femoral crossover bypass may be needed.

Whether to use a graft with suprarenal or infrarenal fixation remains debated. Suprarenal fixation may provide more durable proximal fixation of the stent-graft when anatomic features are not optimal, such as with a short aortic neck, aortic angulation, conical aortic neck, or circumferential mural thrombus or calcification. However, the placement of stents in the region of the visceral vessels has raised concerns for embolization both at the time of the endograft placement and over time. Suprarenal fixation may lead to a higher incidence of small renal infarcts, but most are not clinically evident. However, in the face of preexisting renal insufficiency, suprarenal fixation should be used with caution. (See "Complications of endovascular abdominal aortic repair", section on 'Renal ischemia'.)

Aortic endografts are prefabricated in various diameters and lengths. Once the graft design has been chosen, the particular device components that are used are determined by the measurements and configuration of the aneurysm being treated. The various manufacturers provide templates to assist the surgeon in choosing graft components based upon aneurysm measurements. (See 'Measurement definitions' above.)

Cost — There is some variability in the cost of EVAR, partially based upon the choice of the endograft and number of device components needed to complete the repair, as well as institutional agreements with device manufacturers. Various other components, such as ipsilateral and contralateral limbs, limb extensions, and other devices needed to correct any endoleaks, add cost per component used. Negotiated institutional prices for the grafts may vary from the list price, and costs related to the surgical suite, operating personnel, anesthesia services, and hospital costs based on the type of hospital bed and length of hospital stay are also variable. Other costs are incurred following discharge from the hospital for ongoing surveillance and re-intervention due to endoleak, when needed.

Compared with open repair of AAA, EVAR is more costly [27-34]. In an analysis of the EVAR 1 trial, the mean costs for the initial EVAR and open repair were £13,019 ($20,271 US) and £11,842 ($18,438 US), respectively [29,33]. The difference in lifetime cost was £3519 ($5479 US) higher with EVAR, but there was only a very small difference in quality-adjusted life-years (QALYs) of -0.032 (95% CI -0.117 to 0.096) in favor of open repair. In EVAR 2, the mean cost difference was £10,596 ($16,499 US) higher for EVAR compared with open repair. A later cost analysis found that EVAR was not cost effective in the long term compared with open repair among trials conducted in Europe [27]. However, EVAR did appear to be cost effective based on the Open versus Endovascular Repair Veterans Affairs (VA) Cooperative Study (OVER), which was conducted in the USA [35]. In a later analysis, endovascular implant costs constituted 38 percent of initial hospitalization costs. Mean device costs were $13,600 to $14,400 (USD), and initial hospitalization costs were $34,800 to $38,900 (USD). Total health care costs at two years were not significantly different among device systems: $72,400 to $78,200 (USD) for the endovascular group and $75,600 to $82,100 (USD) in the open repair group.

PREOPERATIVE RISK ASSESSMENT — Although endovascular aneurysm repair (EVAR) is associated with lower perioperative morbidity and mortality compared with open surgical repair, there is a small risk that the endovascular repair may need to be converted to an open repair, and thus, patients should be evaluated and prepared as if undergoing an open surgical repair. Coronary artery disease (CAD) is the leading cause of early and late mortality following AAA repair, and other comorbidities such as chronic obstructive pulmonary disease (COPD) and renal insufficiency also increase perioperative morbidity and mortality. We agree with the Society for Vascular Surgery and other societies that recommend a comprehensive assessment of medical comorbidities prior to EVAR including cardiac, pulmonary, and renal evaluation, also taking into account hypertension and patient age as relevant risk factors for morbidity and mortality [36-38]. The evaluation of cardiopulmonary risk and risk management strategies are discussed in detail elsewhere. (See "Evaluation of cardiac risk prior to noncardiac surgery" and "Management of cardiac risk for noncardiac surgery" and "Evaluation of perioperative pulmonary risk" and "Strategies to reduce postoperative pulmonary complications in adults".)

PREPARATION

Thromboprophylaxis — Patients undergoing AAA repair (endovascular or open surgery) are considered to be at moderate to high risk for deep venous thrombosis, and thromboprophylaxis is indicated. (See "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients".)

In a study of 193 patients undergoing AAA repair, the incidence of thromboembolism was lower for endovascular aneurysm repair compared with open surgery; however, the incidence of deep vein thrombosis following EVAR was 5.3 percent in spite of pharmacologic thromboprophylaxis [39]. Delayed initiation of pharmacologic prophylaxis was associated with a trend toward an increased incidence of deep venous thrombosis.

Antibiotic prophylaxis — Prior to endograft placement, antibiotic prophylaxis is recommended within 30 minutes of the skin incision. Appropriate antibiotics are given in the table (table 3) [40]. Antibiotics are discontinued within 24 hours given the lack of added benefit beyond that time frame. (See "Antimicrobial prophylaxis for prevention of surgical site infection in adults", section on 'Vascular surgery'.)

Prevention of contrast-induced nephropathy — EVAR increases the risk of renal complications, primarily due to the administration of intravenous contrast agents, but potentially also related to the dislodgement of embolic debris by manipulation of catheters and wires near the renal arteries, or impingement of the renal ostia by the graft or suprarenal fixation portion of the stent-graft. (See "Complications of endovascular abdominal aortic repair", section on 'Intravenous contrast complications'.)

When EVAR will be performed in a patient with preexisting renal insufficiency, strategies to reduce the risk for contrast-induced nephropathy should be used [36]. These strategies are discussed in detail elsewhere. (See "Prevention of contrast-associated acute kidney injury related to angiography".)

Prophylactic renal artery stenting — The need to deploy a stent-graft near the origin of stenotic renal arteries is not uncommon. Although grafts with suprarenal fixation do not seem to affect renal function in patients with normal renal arteries, it is not clear if this holds true in the setting of severe renal artery stenosis and preexisting renal insufficiency. Whether prophylactic renal artery stenting should be performed in this setting is unclear. A review from the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP, 2011 to 2014) database evaluated the outcomes of 281 patients who underwent concurrent renal artery angioplasty and stenting (RAAS) during EVAR. Even after controlling for risk factors that might contribute to postoperative renal dysfunction, concurrent RAAS was significantly associated with adverse renal outcomes after EVAR for infrarenal AAA [41]. If a device with suprarenal fixation is needed, the possibility that the renal artery flow will be compromised due to the suprarenal struts needs to be weighed against the risks and potential benefits of prophylactic renal artery stenting. (See "Complications of endovascular abdominal aortic repair" and "Complications of endovascular abdominal aortic repair", section on 'Renal ischemia'.)

Hypogastric embolization — Hypogastric (ie, internal iliac artery) embolization may be needed to prevent type II endoleak (table 2) during endovascular repair of aortic aneurysms involving the distal common iliac artery and/or hypogastric artery (ie, internal iliac artery). However, given the availability of approved branched iliac devices (figure 6), the need for hypogastric embolization should be limited to a very select group of patients. The endovascular management of patients with iliac artery aneurysm in conjunction with abdominal aortic aneurysm repair is discussed elsewhere. When unilateral hypogastric embolization is needed, it is often performed preoperatively; however, it can also be performed just prior to endovascular stent-graft placement. Patients with bilateral iliac artery aneurysms generally undergo a staged approach. Hypogastric embolization increases the proportion of patients who will have anatomy that is suitable for endovascular aneurysm repair. (See "Surgical and endovascular repair of iliac artery aneurysm" and "Complications of endovascular abdominal aortic repair", section on 'Pelvic ischemia'.)

Prophylaxis for postimplantation syndrome — Between 13 and 60 percent of patients experience a transient acute flu-like inflammatory syndrome following aortic endograft placement that can delay the usually quick recovery following EVAR. (See "Complications of endovascular abdominal aortic repair", section on 'Postimplantation syndrome'.)

Whether or not to provide specific preoperative prophylaxis to prevent this syndrome was addressed in the PreOperative Methylprednisolone in EndoVascular Aortic Repair (POMEVAR) trial, which randomly assigned 153 patients to receive 30 mg/kg intravenous infusion of methylprednisone or placebo prior to endovascular aortic repair [42]. Markers for inflammation (maximal plasma interleukin, C-reactive protein, interleukin 8, soluble tumor necrosis factor) were lower in the methylprednisone group. Postoperative narcotic requirements were significantly reduced, and fulfillment of discharge criteria occurred one day sooner for the treated versus placebo group (two versus three days postoperatively), with no differences in perioperative medical or surgical morbidity. Larger studies are needed to confirm these promising results and to identify optimal dosing, potential longer-term adverse effects (eg, wound complications, endoleak), as well as the subset of patients who might benefit (or be harmed) from such prophylaxis. Comparisons of prophylactic with symptomatic-only treatment for postimplantation syndrome are also needed. As such, we do not provide such prophylactic treatment.

ENDOGRAFT PLACEMENT — Endovascular aneurysm repair (EVAR) is performed by inserting folded graft components into the aorta to construct an endograft, which expands upon deployment, contacting the aortic wall proximally and the iliac vessels distally to exclude the aortic aneurysm sac from aortic blood flow and pressure.

The chosen device and its components should be present in the operating room before the start of the procedure, and additional device components, guidewires, and sheaths should be immediately available to manage any technical issues that might arise.

Centers treating AAA using EVAR should ideally be equipped with a dedicated endovascular operating suite in which conversion to open repair can be performed efficiently, if needed. EVAR can be performed using portable digital subtraction fluoroscopic imaging or fixed imaging systems. Although fixed imaging systems provide better image quality and may include concomitant computed tomography (CT) and three-dimensional imaging, newer-generation portable fluoroscopic systems provide adequate detail for routine EVAR.

Once the patient is anesthetized, endovascular aneurysm repair is accomplished through an orderly sequence that includes gaining vascular access, placement of arterial guidewires and sheaths, imaging to confirm aortoiliac anatomy, main body deployment, gate cannulation (bifurcated graft), iliac limb deployment, graft ballooning, and completion imaging. The general technique for routine endovascular repair is described below, and typical variations needed for troubleshooting common problems. The complications associated with EVAR are discussed separately. (See "Complications of endovascular abdominal aortic repair".)

Anesthesia — EVAR can be performed under general anesthesia, regional anesthesia, total intravenous anesthesia (TIVA), or local anesthesia with conscious sedation. The type of anesthesia used is often one of surgeon preference, but multiple studies have shown a benefit to limiting the use of general anesthesia, when possible [43-47]. Local/regional anesthesia is associated with shorter operating times and shorter hospital stays with no apparent differences in complications or perioperative mortality compared with general anesthesia [43]. If the cooperation level of the patient is anticipated to be poor, general anesthesia may be preferred to limit patient movement to allow accurate graft positioning. (See "Anesthesia for endovascular aortic repair".)

Vascular access — Bilateral femoral access is needed to place endografts. Endovascular repair of AAA can be performed via surgical femoral cutdown or percutaneously. The cutdown approach is similar to that used for femoral embolectomy or femoral graft placement; however, in patients with a limited amount of atherosclerotic disease and no evidence for femoral aneurysm, exposure can be limited to the area of the common femoral artery that will be punctured. The incision can be longitudinal or transverse, but if patch repair of the common femoral artery is anticipated, a longitudinal incision is preferred.

Open access can be more challenging in obese patients or in patients with prior groin surgery. Arteriotomy repair with a percutaneous suture device is feasible, even after the use of large-bore introducers [8,48-50]. In determining whether an artery is suitable for use with a percutaneous arterial closure device, several factors are important, including the size of the femoral artery, prior groin surgery or use of a closure device, significant arterial occlusive disease, and common femoral aneurysm.

Once the common femoral artery is accessed, deployment of the sheath into the anterior common femoral artery is critical. More proximal punctures through the inguinal ligament may cause problems with proper knot tying, and the proper application of pressure to achieve hemostasis may be more difficult. More distal access (eg, superficial femoral artery) can lead to vessel thrombosis.

Percutaneous access — Percutaneous access uses standard guidewire access and then uses progressive dilators to place the endograft sheath rather than direct exposure of the access vessel. With the increased availability of a later-generation large-bore closure devices, percutaneous access has become the norm at many institutions.

Once the endograft is in place, closure of the defect in the vessel is generally accomplished using specialized vascular closure devices [50-53]. A fascia suturing technique that places sutures around the sheath into the cribriform fascia after extending the skin puncture incision has also been described [54].

A systematic review identified four trials [50-53]. In the largest of these, the percutaneous endovascular aortic aneurysm repair (PEVAR) trial, 151 patients with suitable iliofemoral anatomy for a percutaneous technique were randomly assigned to open femoral exposure or a percutaneous approach using the Perclose ProGlide or Prostar XL devices. In another trial, 100 patients were randomly assigned to percutaneous or open cutdown. A meta-analysis found no differences between open femoral access and percutaneous access with respect to mortality, aneurysm exclusion, major complications, and bleeding complications, but operating time was significantly reduced [55]. A separate review that included these and other observational studies also found a shorter procedure time, lower complication rate, and similar technical success rates for percutaneous compared with open access [56,57].

The reported success rates of percutaneous EVAR is between 90 and 100 percent [8,48]. The variability appears to be related to the size of the sheath, degree of femoral artery calcification, and operator experience with the technique [50,58]. Sheath size and obesity do not appear to have a significant impact. The fascia suturing technique was found in one trial to provide a secure closure more quickly with less cost than using a closure device [52].

In the PEVAR trial, procedural technical success was 98 percent for the open femoral approach and 94 and 88 percent in the ProGlide and Prostar groups, respectively. At one month, primary treatment success rates (composite procedure technical success, major adverse events, and vascular complications) were 78, 88, and 78 percent, for the open, ProGlide, and Prostar groups, respectively. Differences between the devices led to a conclusion of noninferiority for the ProGlide device, but not the Prostar device compared with open femoral access. The percutaneous approach was associated with shorter times to hemostasis and procedure completion, less blood loss, less groin pain, and improved quality of life. Outcomes persisted to six months with no reported aneurysm rupture, conversion to open repair, device migration, or stent graft occlusion. The results of this trial confirm observed outcomes at several institutions where percutaneous EVAR has been used in daily practice using a variety of endografts of different profiles.

Iliac conduits — Small-diameter access vessels increase the technical difficulty of EVAR, particularly in the setting of vessel calcification and tortuosity. Severe vascular stenoses or occlusions of the iliac arteries (TASC C and D lesions) increase the risk of a major iliac vessel complication and are an independent predictor of procedural failure [59]. Iliac rupture is a potentially fatal complication that appears to be more common in women, likely related to the generally smaller caliber of iliac artery [60]. However, given the availability of low-profile endografts, the need for an iliac conduit has become uncommon in contemporary practice.

Small-caliber iliac vessels can be handled in a stepwise fashion, but the need for advanced interventions for iliac access is uncommon given the availability of later generation low-profile devices. The initial approach is stretching the artery using graduated dilators or focal balloon angioplasty, as needed. If these maneuvers are not successful, the stent-graft can be delivered through an iliac conduit, which is a graft (eg, Dacron, expanded polytetrafluoroethylene [ePTFE]) that is sutured to the common iliac artery or distal aorta. The need for an iliac conduit should be anticipated based upon preoperative imaging. After providing adequate exposure of the pelvic vasculature through a low retroperitoneal incision, the iliac conduit is sutured into place and the sheath inserted into the conduit (picture 1). To provide a less acute angle to pass the endograft, the iliac conduit can be passed through a subcutaneous tunnel under the inguinal ligament into the groin region before placing the sheath.

An alternative technique is termed an internal endoconduit. A covered stent is deployed within the iliac artery, and the endovascular sheath is subsequently passed from a common femoral artery access site; an adequately sized femoral artery is a prerequisite for this technique. Any disruption of the iliac artery during passage of the sheath is contained by the covered stent [61]. Unexplained hemodynamic instability may indicate avulsion of the iliac artery. This technique can be useful in patients with a hostile retroperitoneum or colostomy, which can complicate the placement of an open iliac conduit.

Anticoagulation — Prior to device insertion, systemic anticoagulation is initiated, typically using a weight-based intravenous heparin dose of 80 to 100 units/kg. The goal is an activated coagulation time (ACT) of 200 seconds or greater. After device deployment, anticoagulation may be reversed at the surgeon's discretion. We do not routinely reverse anticoagulation unless indicated due to bleeding or possibly endoleak that fails to respond to routine measures. (See 'Handling immediate endoleak' below and "Endoleak following endovascular aortic repair", section on 'Alterations to antithrombotic therapies'.)

Measures to prevent endoleak — Endoleak is a term that describes the presence of persistent flow of blood into the aneurysm sac after endograft placement (table 2 and figure 7). (See "Endoleak following endovascular aortic repair".)

Type I endoleak — Type I endoleak (figure 7) leads to sac enlargement and potential rupture. Adherence to the instructions for use (IFU) for each device manufacturer appears to be associated with a lower incidence of sac enlargement on follow-up. In a review of over 10,000 patients undergoing EVAR, 42 percent of patients had anatomy that met the most conservative definition of device IFU, and only 69 percent met the most liberal definition of the device IFU [5]. The overall rate of AAA sac enlargement was 41 percent at five years post-EVAR. Independent predictors of sac enlargement included endoleak, age ≥80 years, aortic neck diameter ≥28 mm, aortic neck angle >60°, and common iliac artery diameter >20 mm [5]. This suggests that compliance with the IFU can decrease the incidence of endoleaks leading to sac enlargement.

Type II endoleak — Preemptive measures to prevent type II endoleak (figure 7) appear effective, but given the natural history of type II endoleak, routine use of such measures is not suggested [62,63]. We agree with vascular society guidelines that advocate selective intervention post-EVAR [18,37]. (See "Endoleak following endovascular aortic repair", section on 'Type II endoleak'.)

Methods to prevent type II endoleak include embolization of aortic side branches or instillation of glue or coils into the aneurysm sac [64-66]. Intraprocedural glue or coil embolization of the aneurysm sac has been performed successfully, but with variable results [67-73].

Side branch embolization can be performed preoperatively or intraoperatively and is effective at reducing the incidence of type II endoleak. In a review that identified five studies, the pooled rate of type II endoleak was significantly lower following inferior mesenteric artery embolization compared with no embolization (19.9 versus 41.4 percent) [62]. In the only randomized trial, 106 patients at risk for type II endoleak were assigned to EVAR with or without preemptive inferior mesenteric artery embolization [67]. The incidence of type II endoleak was significantly reduced in the embolization group (absolute risk reduction 24.5 percent, 95% CI 6.2 to 40.5) and the incidence of aneurysm sac expansion was also reduced at 12 months follow-up (3.8 versus 17 percent).

The Sac COil embolization for Prevention of Endoleak (SCOPE1) trial randomly assigned 94 patients at high risk for type II endoleak to endovascular aneurysm repair with and without aneurysm sac coil embolization during aneurysm repair [72]. High risk was defined as: a patent inferior mesenteric artery with a luminal diameter at its origin >3 mm; or at least three pairs of patent lumbar arteries; or two pairs of patent lumbar arteries and a median sacral artery or an accessory renal artery. Preemptive sac coil embolization reduced the rate of endoleak at one (4.3 versus 34 percent), six (4.3 versus 42 percent), and 12 months (14.3 versus 41 percent), but the rate of endoleak was not significantly different at 24 months (10 versus 25 percent). The overall reintervention rate was reduced for preemptive coil embolization, but reintervention for sac enlargement due to type II endoleak at 24 months was similar. The criterion for reintervention in this trial was 5 mm sac enlargement. (See "Endoleak following endovascular aortic repair", section on 'Criteria for treatment'.)

Preemptive embolization can reduce the incidence of type II endoleak, but given the benign natural history of type II endoleak, it remains unclear whether the potential risks associated with the added intervention can be justified. Risks include the additional time and intravenous contrast required to perform the procedure and the potential for procedural complications (eg, visceral ischemia, distal embolization) [62].

Type III endoleak — Type III endoleak (figure 7) also leads to sac enlargement and potential rupture. Certain grafts have been associated with a higher incidence of type III endoleak. In a single institution review, patients receiving certain endografts (Powerlink or AFX) were included in group 1 while patients receiving other endografts were included in group 2. The rate of major adverse events involving aneurysm treatment were significantly more frequent in group 1 compared with group 2 (38 of 83 versus 5 of 68). Among group 1 endografts, AAA diameter >6.5 cm was significantly associated with development of a type III endoleak (odds ratio 11.16, 95% CI 2.17-57.27) [74]. Thus, type III endoleak can potentially be prevented through judicious device selection. Ongoing monitoring is important for all endografts.

Graft deployment — Once vascular access is established and landmarks for positioning the device are obtained with aortography, the main device is positioned with particular attention paid to the location of the opening for the contralateral iliac limb ("contralateral gate"). The aortic neck is imaged; a slight degree of craniocaudal and left anterior oblique angulation may improve imaging of the renal ostia. With the proximal radiopaque markers of the graft positioned appropriately below the lowest renal artery, the body of the graft is deployed to the level of the contralateral gate.

A guidewire is advanced through the contralateral access site into the contralateral gate. Gate cannulation is confirmed by placing a pigtail catheter over the guidewire into the main body of the graft, removing the guidewire and confirming that the pigtail catheter rotates freely within the main body of the graft; if it does not, the catheter is assumed to be in the aneurysm sac. Once the contralateral guidewire is positioned within the main body of the endograft, the deployment of the endograft at the neck of the aneurysm is completed followed by deployment of the contralateral, then ipsilateral iliac artery limbs (depending on the type of graft). Once the endograft components are in place, the attachment sites and endograft junctions are gently angioplastied with a compliant or semi-compliant balloon.

Completion aortography is performed to evaluate the patency of the renal arteries and evaluate for endoleak. Guidewire access is maintained throughout the procedure but is particularly important when removing the main graft body device sheath since disruption of the access vessels by an oversized sheath may not become apparent until after sheath has been removed.

Troubleshooting

Gate cannulation — Deployment of the main body of the device with the gate low in the aneurysm sac or below the aortic bifurcation can lead to a situation in which the contralateral gate does not open up when the device is deployed. This is referred to as "jailing the contralateral gate." In some cases, pushing the device upward will allow the gate to flare open. Many currently available endovascular graft devices allow recapture of the graft to reorient the gate opening. If the gate of the graft cannot be moved more superiorly, conversion to an aortouniiliac (AUI) configuration or open conversion may be needed.

Even with correctly positioned grafts, it is not uncommon to encounter difficulties with gate cannulation, particularly with large aneurysm sacs that have minimal mural thrombus. Catheters with reverse angulation (eg, Van Schie catheter) can help engage the gate. If various types of catheters do not help the wire pass through the gate, a wire can be passed up and over the graft flow divider from the ipsilateral to contralateral femoral access, or alternatively from an access site in the arm antegrade down through the gate. The antegrade wire can then be snared and pulled through the contralateral femoral access sheath.

Handling immediate endoleak — Endoleak is a term that describes the presence of persistent flow of blood into the aneurysm sac after device placement, indicating failure to completely exclude the aneurysm from the aortic circulation [75]. The types of endoleak and their management are discussed in detail separately (table 2 and figure 7). (See "Endoleak following endovascular aortic repair".)

Immediate type I and type III endoleak identified at the time of endograft placement are repaired and generally respond to additional ballooning or the placement of additional endograft components. Open conversion is almost never required. Type II and type IV endoleak do not require any specific intervention when identified at the time of endograft placement.

POSTOPERATIVE CARE — Following endovascular repair of abdominal aortic aneurysm, patients can be transferred to a regular floor once they have recovered from anesthesia. The patient can drink clear fluids and advance to a regular diet, as tolerated. Treatment of acute postoperative pain consists mainly of nonsteroidal anti-inflammatory drugs (NSAIDs) and/or opioids. (See "Approach to the management of acute pain in adults".)

Ambulation is resumed on the first postoperative day. The majority of patients can be discharged home within 24 hours following EVAR, provided no complications have occurred. Fluid therapy is continued to minimize the risk of contrast nephropathy, particularly in patients with preoperative renal insufficiency. (See 'Prevention of contrast-induced nephropathy' above.)

The peripheral pulse exam should be assessed at regular intervals and compared with the baseline preoperative examination. Any abnormalities should prompt duplex evaluation for potential endograft limb complications. (See "Complications of endovascular abdominal aortic repair", section on 'Extremity ischemia'.)

Resumption of antithrombotic therapy — Following successful EVAR, patients should resume their usual medications, including aspirin, which is recommended for secondary prevention in patients with peripheral artery disease. (See "Prevention of cardiovascular disease events in those with established disease (secondary prevention) or at very high risk", section on 'Antiplatelet therapy'.)

Whether antithrombotic therapies (antiplatelet agents, warfarin) increase the incidence for endoleak following EVAR is not well studied. (See "Endoleak following endovascular aortic repair", section on 'Alterations to antithrombotic therapies'.)

Endograft surveillance — Failure of aortic endografts is well documented and can lead to continued aneurysm expansion and potential rupture. Thus, it is mandatory for all patients to undergo routine surveillance to assure the integrity of the repair. The principle concerns are endoleak, aneurysm sac enlargement, migration of the stents at the aortic and iliac landing zones, and separation of the device components. Abdominal plain films are an economical and quick way to evaluate the integrity of the metallic structure of the graft and endograft alignment and can be obtained prior to discharge from the hospital. For ongoing surveillance, we suggest that surveillance should be performed at one month post-repair and then yearly thereafter for uncomplicated repairs. Although all the device clinical trials have mandated surveillance at six months postrepair, one review suggested that imaging at this interval does not reveal new findings or change patient management [76]. Thus, we do not obtain a six-month study when the one-month follow-up surveillance shows nothing concerning.

In a review of United States Medicare claims, surveillance information for nearly 8000 patients showed that 50 percent had incomplete surveillance [77]. When comparing those with and without complete surveillance, there were no significant differences in aneurysm-related mortality (0.3 versus 0.6 percent). Those with incomplete surveillance had a lower incidence of minor or major reintervention (1.4 versus 10 percent) and a lower incidence of late rupture (0.7 versus 1.4 percent). However, given the limitations of database reviews, which do not offer complete clinical information such as the presence of endoleak, such findings do not justify a change in clinical practice.

The most commonly used imaging modalities for ongoing endograft surveillance are contrast-enhanced computed tomographic (CT) arteriography and duplex ultrasonography (DU). Other surveillance modalities include magnetic resonance (MR) imaging and aneurysm sac pressure measurements. In a systematic review and meta-analysis, the highest endoleak detection rates were for surveillance approaches that used combined imaging modalities [78]. Most studies reported detection rates of patient-important outcomes at 1, 6, 12, 24, 36, 48, and 60 months. There were insufficient data to inform the best strategy for other EVAR complications in follow-up (eg, limb ischemia, rupture, renal complications).

CT angiography with delayed images is the most widely used modality for follow-up after EVAR. It is accurate for maximal diameter measurement and for the detection of endoleak (table 2 and figure 7) and other device-related complications [79-84]. However, CT angiography is costly, and repeated radiation exposure is associated with an increased lifetime cancer risk [79]. Repeated administration of intravenous contrast may also contribute to a progressive decline in renal function that has been observed following EVAR [85,86].

The guidelines for the management of abdominal aortic aneurysm (AAA) from the Society for Vascular Surgery advocate CT angiography and color DU at one month post EVAR. In the absence of an endoleak or sac enlargement, the guidelines recommend contrast-enhanced CT or DU imaging at 12 months after EVAR. Imaging at six months is no longer routinely recommended unless a device-related abnormality is identified at the one-month imaging study after EVAR [36,38]. If an endoleak or aneurysm enlargement is not documented during the first year after EVAR, DU is an alternative to CT angiography for ongoing postoperative surveillance. Alterations to the surveillance schedule when endoleak occurs are reviewed separately. (See "Endoleak following endovascular aortic repair", section on 'Diagnosis' and "Endoleak following endovascular aortic repair", section on 'Alterations to surveillance'.)

Unenhanced CT scan combined with DU is an acceptable alternative to avoid intravenous contrast load and repeated exposure to radiation associated with repeated CT angiography. DU needs to be performed by a skilled technician in an appropriately credentialed laboratory. Several studies have established the efficacy of DU and its ability to detect endoleaks and evaluate sac expansion [87-99]. However, some speculate a lower sensitivity of unenhanced ultrasound for diagnosis of endoleak, which has led to increasing interest in contrast-enhanced ultrasound imaging [82]. In one systematic review, contrast-enhanced (CE) ultrasound was nearly as sensitive as CT angiography [78]. The pooled sensitivities and specificities for CE ultrasound for detecting any type of endoleak are 96 to 98 percent, and 85 to 88 percent, respectively, using contrast-enhanced CT angiography as the diagnostic standard [78,96,97]. In comparing CE ultrasound to unenhanced ultrasound for all types of endoleak, CE ultrasound had a greater sensitivity but lower specificity than unenhanced ultrasound; however, for only type I and type III leaks, no differences were found [97]. This appears to suggest that the greater sensitivity of CE ultrasound compared with unenhanced ultrasound is related to identifying type 2 endoleak, for which the need for intervention is less certain.

Although MR imaging had a higher rate of endoleak detection in one systematic review when compared with CT angiography [78], MR imaging is not a standard modality for EVAR surveillance, but it can be used in specific situations where CT angiography is contraindicated [36,37]. The advantage of MR imaging is the lack of exposure to ionizing radiation. Disadvantages are its lack of wide availability and difficulty evaluating device integrity due to artifact. The placement of stent-grafts made of nitinol does not preclude MR imaging, though MR imaging is contraindicated for stainless-steel-based grafts (eg, Cook, Zenith) (table 1).

Direct measurement of the pressure within the aneurysm sac after EVAR can be performed and is a reliable indicator of intra-sac pressure but is invasive. Noninvasive aneurysm sac pressure measurements using implantable wireless pressure sensing systems have been developed, but accuracy may be limited by the presence of mural thrombus [100]. For type II endoleaks (table 2) of unclear clinical significance, the sensor may assist in therapeutic management. However, remote pressure sensing does not provide any information about device integrity, and these devices are unlikely to be used as stand-alone surveillance after EVAR but may complement other low-risk modalities such as DU imaging.

Digital subtraction arteriography is reserved for the evaluation of specific problems such as decreased limb flow, graft limb thrombosis, documented endoleak, or to measure aneurysm sac pressure when an enlarging aneurysm is identified in the absence of an endoleak [101].

During surveillance of endovascular repair, intraprosthetic thrombotic deposits are a common finding; however, their clinical significance is debated. In a systematic review, five studies involving 808 patients were analyzed [102]. Thrombus detection at any time during follow-up occurred in 20.8 percent of patients. Intraprosthetic deposits were generally detected during the first year after repair. Intraprosthetic mural thrombus was not significantly associated with thromboembolism during follow-up. Risk factors include polyester graft material (odds ratio [OR] 2.34, 95% CI 1.53-3.58) and aorto-uni-iliac configuration of the endograft (OR 3.27, 95% CI 1.66-6.44). Other studies have suggested that intraprosthetic thrombus formation may also be associated with long main bodies and large necks.

PERIOPERATIVE MORBIDITY AND MORTALITY — The overall complication rate for endovascular aneurysm repair, including access-related issues, is approximately 10 percent. The complications associated with endovascular aneurysm repair are discussed in detail separately. (See "Complications of endovascular abdominal aortic repair".)

Device-related complications may include vascular injury as a result of device deployment and endoleak. Following successful endograft repair, the aneurysm sac will eventually thrombose, and the sac of most aneurysms progressively shrinks. The endograft repair may respond to the changing configuration of the aneurysm, which can lead to late endograft complications such as angulation, kinking, thrombosis, or migration. Thus, lifetime endograft surveillance is required. (See 'Endograft surveillance' above.)

The short-term technical success rate for endovascular aneurysm repair has improved with increasing experience with the technique [18]. Technical failures may be due to a vascular complication or inability to resolve a type I leak. Open conversion at the time of repair is uncommon and is required in fewer than 2 percent of patients [103-105]. Thirty-day technical success rates range from 77 to 100 percent [106,107].

A systematic review identified four trials that included 1532 patients who were considered suitable candidates for endovascular or open repair of nonruptured abdominal aortic aneurysms larger than 5.0 cm in diameter [107]. The 30-day all-cause mortality was significantly lower with endovascular repair (1.6 versus 4.8 percent). The short-term survival advantage of endovascular repair appears to be much greater when endovascular repair is limited to patients at highest risk for open surgery. This was illustrated in a report of 454 consecutive patients who underwent elective repair (206 endovascular and 248 open surgeries) of an abdominal aortic aneurysm [108]. The overall 30-day mortality rates did not significantly differ for endografting and surgery (2.4 and 4.8 percent, respectively). However, among patients at highest surgical risk (American Society of Anesthesiologists [ASA] class IV), the 30-day mortality rates were lower for endovascular compared with open repair (4.7 versus 19.2 percent).

Using the United States Medicare database, long-term survival was evaluated in 22,830 matched pairs of patients who underwent elective repair with an open or endovascular technique between 2001 and 2004 [105]. There was a significantly lower rate of perioperative mortality with endovascular repair (1.2 versus 4.8 percent); a more pronounced benefit was seen with increasing age. However, as seen with other studies, overall mortality at three to four years following repair was nearly identical in the two groups.

Over the long term, however, it has not been definitively established that endovascular repair is superior to open surgical repair, even among patients at highest risk for surgery. Although a 3 percent reduction in aneurysm-related mortality persisted throughout the follow-up period in both the DREAM and EVAR1 trials, the initial reduction in all-cause mortality was eliminated within one to two years with equivalent overall survival in both treatment groups. A later analysis from the Medicare database of 4529 patients treated between 2003 and 2007 and followed over a median of 2.4 years found similar results in the perioperative period [109]. Aneurysm-related mortality and perioperative (30 day) mortality were reduced in the endovascular group. Among patients who survived >30 days, no difference in all-cause mortality was seen (hazard ratio [HR] 1.01, 95% CI 0.84-1.22). However, in this study, the survival advantage conferred with endovascular repair was maintained throughout the study period such that the all-cause mortality risk for open repair relative to endovascular repair remained elevated (HR 1.24, 95% CI 1.05-1.47). The average age of the patients who underwent endovascular repair was not significantly different compared with those treated with open repair (75 versus 76). (See "Management of asymptomatic abdominal aortic aneurysm", section on 'Open versus endovascular aneurysm repair'.)

Long-term follow-up data reinforce the concept that the initial survival advantage with EVAR may not be maintained beyond the perioperative period compared to open repair [110]. EVAR does not seem to increase overall life expectancy in patients ineligible for open repair but can reduce aneurysm-related mortality. Moreover, a meta-analysis of individual patient data from EVAR-1, DREAM, OVER, and ACE trials compared the outcomes of endovascular or open repair for AAA over five years and confirmed the early survival advantage in the EVAR group and its subsequent erosion [111]. Over five years, patients of marginal fitness had no early survival advantage from EVAR compared with open repair. Aneurysm-related mortality and patients with low ankle-brachial pressure index contributed to the erosion of the early survival advantage for the EVAR group.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Aortic and other peripheral aneurysms".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Endovascular surgery (The Basics)")

●Beyond the Basics topic (see "Patient education: Abdominal aortic aneurysm (Beyond the Basics)").

SUMMARY AND RECOMMENDATIONS

●Abdominal aortic aneurysm – The abdominal aorta is defined as aneurysmal when a localized dilation is identified, and the diameter of that region is increased more than 50 percent relative to normal aortic diameter. For most individuals, an aortic diameter greater than 3.0 cm is generally considered aneurysmal. (See 'Anatomic considerations' above.)

●Endovascular aneurysm repair – Endovascular repair of abdominal aortic aneurysm (EVAR) represents a widely available alternative to open surgical repair. However, the precise role of EVAR continues to be defined. Guidelines from major medical and surgical societies recommend an individualized approach to the patient when choosing between open and endovascular repair, taking into account the patient's age, risk factors for perioperative morbidity and mortality, anatomic factors, and the experience of the surgeon. (See 'Indications for repair' above.)

●Anatomic suitability – The preprocedural evaluation of patients undergoing endovascular repair requires medical risk assessment and a careful quantitative and qualitative evaluation of aortoiliac anatomy to determine suitability for endovascular repair. With early endograft designs, approximately 50 percent of patients were not candidates for EVAR because of the site, extent, or morphology of the aneurysm, or unsuitability of access vessels. The availability of devices that allow a shorter proximal seal zone and lower-profile devices have expanded the use of EVAR to almost two thirds of patients with an infrarenal abdominal aortic aneurysm (AAA). (See 'Aortoiliac imaging' above and 'Anatomic suitability' above.)

●Postprocedure surveillance – Endograft imaging is mandatory for the remainder of the patient's life to evaluate endograft integrity and positioning. We suggest a combination of plain abdominal radiographs, which can evaluate the metal stent structure; computed tomographic (CT) angiography (or alternatively unenhanced CT) and/or duplex ultrasonography (with or without contrast) to assess the graft at 1 and 12 months postoperatively, and annually thereafter. (See 'Postoperative care' above.)

●Endografts – Available endovascular grafts for infrarenal aortic repair share a bifurcated, modular design (table 1). There are no clear advantages of one design over another. The endovascular graft is constructed by the sequential delivery and deployment of device components in vivo. (See 'Endografts' above and 'Endograft placement' above.)

●Perioperative morbidity and mortality

•The overall complication rate for endovascular aneurysm repair, including access-related issues, is approximately 10 percent. Technical failures may be due to complications or the inability to resolve a type I leak (table 2), but these can usually be managed using adjunctive endovascular procedures. Open conversion may be required but is uncommon, occurring in fewer than 2 percent of patients at the time of repair. Device-related complications include vascular injury as a result of device deployment and endoleak. (See 'Perioperative morbidity and mortality' above and "Complications of endovascular abdominal aortic repair" and "Endoleak following endovascular aortic repair".)

•Short-term mortality rates for endovascular therapy compare favorably with open surgical repair in randomized trials and large observational studies. The benefit is greatest for patients at highest surgical risk in whom the short-term mortality for endovascular repair is significantly lower compared with open repair. It has not been definitively established that endovascular repair is superior to open surgical repair in the long term, even among patients at highest risk for surgery. (See 'Perioperative morbidity and mortality' above and "Management of asymptomatic abdominal aortic aneurysm", section on 'Open versus endovascular aneurysm repair'.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Emile R Mohler, III, MD (deceased), who contributed to an earlier version of this topic review.