Embolization Therapy: Principles and Clinical Applications, 1 Ed.

Abdominal Aorta Aneurysm Endoleaks

Ajita Deodhar • John A. Kaufman

The goal of endovascular repair of an abdominal aortic aneurysm (AAA) is complete exclusion of the aneurysm sac so that it is no longer exposed to arterial pressure, thereby eliminating the risk of rupture. An endoleak is defined as persistent pressurization of the aneurysm sac post endovascular stent graft repair of the aneurysm (EVAR). This usually manifests on imaging studies as aneurysmal sac growth, but not always as sac contrast opacification. Endoleaks are reported in 4% to 30% of patients post EVAR1,2 and carry the risk of subsequent increase in aneurysm sac size and rupture.1,2

Endoleaks may be early (within 30 days post procedure) or late (after 30 days), simple with only inflow, or complex with inflow and outflow. Endoleaks are classified into five types (Table 45.1). A type I endoleak originates at the stent graft seal site in the proximal aorta (IA), distal aorta (IB), or at the iliac occluder site in an aorto-uni-iliac graft (IC). Incomplete apposition of the stent graft with the aortic wall leads to direct communication between the aneurysm sac and the systemic arterial circulation. This type of endoleak is usually seen in the early postprocedural period. However, progressive remodeling of the aorta and aneurysm sac can lead to dilation of the aneurysm neck and iliac angulation, thereby affecting the longitudinal and lateral stability of the stent graft within the aorta.3 This can lead to delayed device migration, kinking (type I endoleak), or component separation (type III). Factors that predispose to early type IA endoleak include a short, angulated, and thrombus-containing neck. Similarly, tortuous and dilated iliac arteries can predispose to an early type IB endoleak.

Type II endoleaks are the most common type of endoleaks encountered after an EVAR, accounting for approximately 40% of all endoleaks.4 In this, there is retrograde flow through one or multiple aortic branches into the aneurysm sac. The inferior mesenteric artery, lumbar arteries, and median sacral and gonadal arteries are the usual culprits. Conceptually, most type II endoleaks are like arteriovenous malformations, with multiple inflow vessels that coalesce in a nidus (the aneurysm sac) and multiple outflow vessels.

Type III endoleaks occur with structural stent graft failure. This includes junctional separation of modular devices or disruption of the graft fabric. Type III endoleaks are fairly unusual, particularly in the early postprocedure setting. Component separation can result from the previously described aortic remodeling.

Type IV endoleaks are uncommon with current endograft technologies. These endoleaks, related to graft porosity, are usually seen during the immediate post stent graft deployment and resolve spontaneously within 48 hours.

Type V endoleak is enlargement of the AAA sac without identification of a distinct endoleak. With older devices that are now no longer in use (i.e., the first-generation Excluder; W. L. Gore & Associates, Flagstaff, Arizona), sac enlargement was due to ultrafiltration of blood by the graft material. This was not a true leak. With the newer devices, a type V endoleak is most likely due to an unidentified type I through III endoleak.


Routine surveillance of patients undergoing EVAR is essential. Surveillance modalities include radiography, ultrasound (US), computed tomography (CT), and magnetic resonance imaging (MRI). Endoleaks can also be diagnosed definitively on angiograms, but the invasive nature of the procedure precludes this from being a routine surveillance modality. In general, CT angiography (CTA) is the accepted standard of care.5 Hence, most patients undergoing EVAR undergo CTA at 1 month, 6 months, and 12 months followed by yearly examinations thereafter, provided there are no complications. Follow-up strategies are in evolution with a tendency toward decreased use of CT. The stent material (stainless steel, nitinol, or Elgiloy), strut thickness, and geometry can affect the choice of follow-up imaging modality.5 Patients with a nitinol endoprosthesis can undergo magnetic resonance angiography (MRA).6Abdominal US and radiography in combination are useful in patients who are unable to get a CT or magnetic resonance examination, usually due to high serum creatinine/low glomerular filtration rate. Abdominal radiographs are helpful in evaluating for stent graft migration, separation of modular components, or limb kinking. Color Doppler US can evaluate for aneurysm sac size as well as the presence of endoleaks, although the sensitivity is operator dependent and varies from 42% to 97%.7 Recent use of second-generation US contrast agents has improved the sensitivity,8 but these agents are not widely used.

CTA and MRA demonstrate endoleaks as contrast material within the aneurysm sac (Fig. 45.1). It is important to obtain delayed-phase images as an endoleak may not be visible on the arterial phase. On US, endoleaks manifest as Doppler signal within the aneurysm sac or to-and-fro waveform in the feeding vessel (Fig. 45.2).

Type I endoleaks are usually evident on the post stent graft deployment angiogram as continued sac opacification of contrast resulting from insufficient proximal or distal seal (Fig. 45.3). On CTA/MRA, they will be visualized as contrast adjacent to or in communication with the proximal or distal seal zone. With a type II endoleak, contrast is visualized within the aneurysm sac, usually at its periphery with retrograde opacification of the vessel responsible for the endoleak (Fig. 45.1). The location of the contrast suggests the vessel of origin: anterior opacification may be from the inferior mesenteric artery or a gonadal artery; posterior opacification may be from lumbar or median sacral arteries. Diagnostic angiograms typically demonstrate retrograde flow, filling the culprit vessel and opacifying the aneurysm sac (Fig. 45.4). The most common sources are collateral flow to the inferior mesenteric artery via an arc of Riolan (from the superior mesenteric artery), lumbar artery via the iliolumbar branch of the internal iliac artery, or the median sacral artery.

Endoleaks located around the stent graft and sparing the periphery of the sac on CTA/MRA may represent type III leaks. Angiography will demonstrate a leak through the graft material (Fig. 45.5). Abdominal CT, radiograph, or fluoroscopy can demonstrate separation of the modular components. Type IV and V leaks are diagnosis of exclusion. US, CT, and MRI all demonstrate sac size enlargement. In type IV, CTA or MRA will demonstrate contrast opacification of the aneurysm sac with no evident reason. In type V, there will be sac enlargement without contrast opacification of the sac on CTA or MRA. Diagnostic angiogram will confirm lack of a type I through III endoleak.


Endoleaks can lead to continued growth in the aneurysm sac size and increase the risk of rupture. Type I and type III endoleaks should always be considered for repair soon after diagnosis. Type II endoleaks can be watched provided there is no growth in sac size and the patient is asymptomatic. Treatment options included endovascular repair and open surgical repair in the case of failed endovascular attempts. A type I endoleak is repaired at the time of initial device placement using angioplasty balloon (e.g., Coda Balloon; Cook Medical, Inc., Bloomington, Indiana) to mold the existing stent graft for better fixation. If angioplasty fails, a balloon-expandable bare metal stent (Palmaz XL; Cordis, Bridgewater, New Jersey) or an overlapping stent graft (extension, cuff, or dedicated device such as the Zenith Renu from Cook Medical, Inc., Bloomington, Indiana) can be used (Fig. 45.3). The choice of a bare metal stent versus a stent graft depends largely on the proximity of visceral aortic branches such as the renal artery, superior mesenteric artery (SMA), or hypogastrics that may be occluded with a stent graft (Fig. 45.6). Endostaples (Aptus Endosystems, Inc., Sunnyvale, California) are available that can be used to approximate the endograft fabric to the wall of the aorta9 (Fig. 45.7). To prevent a type I endoleak resulting from a hypogastric artery when the endograft must be extended into the external iliac artery, prophylactic embolization of the hypogastric artery should be performed either with coils or a vascular plug. With this approach, it is important to ensure that the contralateral internal iliac artery is patent to avoid pelvic and buttock ischemia. Similarly, type III and type V endoleaks are repaired by placing a stent graft across the region of device disruption (type III) or entire stent graft (type V).

Whether and when to repair a type II endoleak is an area of debate. Small type II endoleaks often thrombose spontaneously, particularly over time.10 Hence, some interventionalists choose to follow these via imaging, whereas others will be more aggressive in repairing them as soon as they are diagnosed. Most interventionalists agree on treating type II leaks that are associated with an abdominal pain, increase in sac size, or with a large nonshrinking sac (>5.5 cm) to prevent rupture.

Type II endoleaks can be challenging to repair. The goal is to obliterate the patent lumen within the aneurysm sac as well as embolize the feeding vessels at their insertion from the aorta. Embolization of the feeding vessel without addressing the intra-aneurysmal component can lead to recruitment of other collateral vessels with continued flow to the sac. In essence, the aneurysm sac should be viewed like an arteriovenous malformation where it is of paramount importance to treat the nidus. Embolization can be achieved via a transarterial route, percutaneous translumbar or transcaval sac puncture, or perigraft access (see Fig. 45.6).6 With a transarterial approach, a microcatheter is used to catheterize the culprit artery via the collateral feeding it (e.g., IMA via the arc of Riolan) and placed within the sac. The sac is embolized first followed by the feeding artery. In a direct sac puncture, an angiographic catheter is inserted in the supine position. Then, the patient is placed in a prone position to identify the optimal site for puncture. The sac is punctured under fluoroscopy or cone-beam CT guidance, usually via a left translumbar approach. Alternatively, the access site can be selected based on landmarks from the angiogram as well as diagnostic CT. A transcaval approach can also be used safely, usually with intravascular US guidance. Needle placement within the sac is confirmed by blood return and subsequent contrast injection, which will also demonstrate the feeding and draining vessels. The endoleak cavity and vessels can then be embolized using stainless steel or platinum coils or liquid embolics such as N-butyl cyanoacrylate (NBCA) (TruFill; Cordis Neurovascular, Miami Lakes, Florida) “glue” or Onyx (Covidien, Irvine, California) (Fig. 45.6D). Recent studies11 have shown that the transarterial and translumbar approach are approximately equal and 70% to 75% effective in treating a type II endoleak.

Continued surveillance is recommended even after successful treatment of an endoleak as these can recur, particularly type I, due to sac remodeling and potential progression of the aneurysmal disease proximally and distally. In addition, continued sac expansion has been seen in the absence of an endoleak in patients who continue to smoke or are hyperlipidemic.12


Not all endoleaks can be prevented. However, appropriate patient selection and meticulous attention to planning before stent graft deployments can limit the incidence of endoleaks. Type I endoleaks can be avoided by selecting an appropriate seal zone. Manufacturers have different recommendations depending on the type of graft (e.g., minimum neck length). These instructions for use should be adhered to particularly when the operator has limited experience. Complete embolization of the internal iliac artery origin should be performed when extending the limbs into the external iliac artery. Some authors advocate pre-EVAR embolization of patent IMA13,14 and lumbar arteries to prevent type II endoleaks. However, there is insufficient data to recommend this practice routinely pre-EVAR. Type III endoleaks can be prevented by ensuring adequate overlap between modular components. Endoleaks occurring due to fabric defects are materials’ limitations that cannot be entirely prevented but are fortunately rare. However, aggressive angioplasty after deployment of the stent graft in a region of heavy calcification can lead to fabric disruption and a type III endoleak. Type IV endoleaks are device dependent and usually only seen during the EVAR procedure.

A new approach to endoleak prevention is the use of sac fillers (Nellix Endograft; Endologix, Irvine, California). This technology uses bags attached to the endograft that are filled with biostable polymer during placement of the stent graft. The polymer-filled bags are intended to fill the aneurysm sac, thereby eliminating types II, IV, and V endoleaks. This device is not approved for clinical use in the United States, although early results from international trials are promising.15


• Preoperative embolization to prevent a type II endoleak:

• Coil embolization of accessory renal artery, prominent lumbar arteries,13 and IMA

• Coil or vascular plug embolization of ipsilateral internal iliac artery in case the stent graft landing zone is in the external iliac artery

• Stent graft: Careful sizing, adequate component overlap, and an AAA neck with good anatomy (correct patient selection) can help prevent an endoleak.

• Endoleak diagnosis: Performing selective angiograms of the superior mesenteric and internal iliac arteries will increase the chance of diagnosing an endoleak.

• Endoleak treatment: The AAA sac (endoleak cavity) must be seen as the nidus of an arteriovenous malformation when embolizing the source of a type II endoleak. The AAA sac and the feeding vessels must be adequately embolized.

• Never assume that there is only one type of an endoleak or that an endoleak will not recur.

• Repeat interventions are common, so long-term surveillance is needed.


Endoleaks remain an unsolved problem in the treatment of AAA with stent grafts. As endograft technology improves, more attention will shift to the prevention of endoleaks. The treatment of endoleaks begins with careful analysis of cross-sectional imaging and detailed angiography to appropriately characterize the endoleak and determine AAA size. A full range of endograft and embolization techniques are required to successfully treat endoleaks.


 1. Veith FJ, Baum RA, Ohki T, et al. Nature and significance of endoleaks and endotension: summary of opinions expressed at an international conference. J Vasc Surg. 2002;35(5):1029–1035.

 2. van Marrewijk C, Buth J, Harris PL, et al. Significance of endoleaks after endovascular repair of abdominal aortic aneurysms: the EUROSTAR experience. J Vasc Surg. 2002;35(3):461–473.

 3. Rafii BY, Abilez OJ, Benharash P, et al. Lateral movement of endografts within the aneurysm sac is an indicator of stent-graft instability. J Endovasc Ther. 2008;15(3):335–343.

 4. Bashir MR, Ferral H, Jacobs C, et al. Endoleaks after endovascular abdominal aortic aneurysm repair: management strategies according to CT findings. AJR Am J Roentgenol. 2009;192(4):W178–W186.

 5. Stavropoulos SW, Charagundla SR. Imaging techniques for detection and management of endoleaks after endovascular aortic aneurysm repair. Radiology. 2007;243(3):641–655.

 6. Ayuso JR, de Caralt TM, Pages M, et al. MRA is useful as a follow-up technique after endovascular repair of aortic aneurysms with nitinol endoprostheses. J Magn Reson Imaging. 2004;20(5):803–810.

 7. Raman KG, Missig-Carroll N, Richardson T, et al. Color-flow duplex ultrasound scan versus computed tomographic scan in the surveillance of endovascular aneurysm repair. J Vasc Surg. 2003;38(4):645–651.

 8. Gilabert R, Bunesch L, Real MI, et al. Evaluation of abdominal aortic aneurysm after endovascular repair: prospective validation of contrast-enhanced US with a second-generation US contrast agent. Radiology. 2012;264(1):269–277.

 9. Deaton DH, Mehta M, Kasirajan K, et al. The phase I multicenter trial (STAPLE-1) of the Aptus endovascular repair system: results at 6 months and 1 year. J Vasc Surg. 2009;49(4):851–857.

10. White SB, Stavropoulos SW. Management of endoleaks following endovascular aneurysm repair. Semin Intervent Radiol. 2009;26(1):33–38.

11. Stavropoulos SW, Park J, Fairman R, et al. Type 2 endoleak embolization comparison: translumbar embolization versus modified transarterial embolization. J Vasc Interv Radiol. 2009;20(10):1299–1302.

12. Sarac TP, Gibbons C, Vargas L, et al. Long-term follow-up of type II endoleak embolization reveals the need for close surveillance. J Vasc Surg. 2012;55(1):33–40.

13. Axelrod DJ, Lookstein RA, Guller J, et al. Inferior mesenteric artery embolization before endovascular aneurysm repair: technique and initial results. J Vasc Interv Radiol. 2004;15(11):1263–1267.

14. Marchiori A, von Ristow A, Guimaraes M, et al. Predictive factors for the development of type II endoleaks. J Endovasc Ther. 2011;18(3):299–305.

15. Krievins DK, Holden A, Savlovskis J, et al. EVAR using the Nellix sac-anchoring endoprosthesis: treatment of favourable and adverse anatomy. Eur J Vasc Endovasc Surg. 2011;42(1):38–46.