- Email: [email protected]
Radiological Evaluation of Abdominal Endovascular Aortic Aneurysm Repair
Canadian Association of Radiologists Journal xx (2015) 1e14 www.carjonline.org
Vascular and Interventional Radiology / Radiologie vasculaire et radiologie d’intervention
Radiological Evaluation of Abdominal Endovascular Aortic Aneurysm Repair Avnesh S. Thakor, PhD, FRCRa, James Tanner, MB, BChirb, Shao J. Ong, MBBS, PhDb, Ynyr Hughes-Roberts, FRCRb, Shahzad Ilyas, FRCRb, Claire Cousins, FRCRb, Teik C. See, FRCRb, Darren Klass, MD, FRCRc, Andrew P. Winterbottom, FRCRb,* a
Department of Radiology, Stanford University, Stanford University Medical Center, Palo Alto, California, USA Department of Interventional Radiology, Addenbrooke’s Hospital, University of Cambridge, Cambridge, United Kingdom c Department of Interventional Radiology, Vancouver General Hospital, University of British Columbia, Vancouver, British Columbia, Canada b
Abstract Endovascular aortic aneurysm repair (EVAR) is an alternative to open surgical repair of aortic aneurysms offering lower perioperative mortality and morbidity. As experience increases, clinicians are undertaking complex repairs with hostile aortic anatomy using branched or fenestrated devices or extra components such as chimneys to ensure perfusion to visceral branch vessels whilst excluding the aneurysm. Defining the success of EVAR depends on both clinical and radiographic criteria, but ultimately depends on complete exclusion of the aneurysm from the circulation. Aortic stent grafts are monitored using a combination of imaging modalities including computed tomography angiography (CTA), ultrasonography, magnetic resonance imaging, plain films, and nuclear medicine studies. This article describes when and how to evaluate aortic stent grafts using each of these modalities along with the characteristic features of several of the main stent grafts currently used in clinical practice. The commonly encountered complications from EVAR are also discussed and how they can be detected using each imaging modality. As the radiation burden from serial follow up CTA imaging is now becoming a concern, different follow-up imaging strategies are proposed depending on the complexity of the repair and based on the relative merits and disadvantages of each imaging modality. Resume La reparation endovasculaire d’un anevrisme de l’aorte (REVA) est une solution de rechange qui presente des taux de mortalite et de mesure qu’ils gagnent en morbidite perioperatoires moins eleves que le traitement des anevrismes de l’aorte par chirurgie effractive. A experience, les cliniciens entreprennent des reparations complexes dans l’environnement anatomique difficile de l’aorte a l’aide d’endoprotheses munies de fen^etres ou de branches, auxquelles peut s’ajouter une cheminee, pour assurer la perfusion des branches viscerales et l’exclusion de l’anevrisme. La reussite d’une REVA est definie par des criteres a la fois cliniques et radiographiques, mais repose en definitive sur sa capacite a completement exclure l’anevrisme de la circulation. Les endoprotheses aortiques font l’objet d’une surveillance en combinant plusieurs modalites d’imagerie, notamment l’angiographie par tomodensitometrie (ATDM), l’echographie, l’imagerie par resonance magnetique (IRM), les images radiologiques sans produits de contraste et les examens de medecine nucleaire. Le present article decrit quand et comment chaque modalite se pr^ete a l’evaluation des endoprotheses, ainsi que les caracteristiques propres aux principales endoprotheses en usage dans la pratique clinique. Il aborde egalement les complications courantes de la REVA et la fac¸on de les detecter a l’aide de chacune des modalites d’imagerie. Enfin, compte tenu des inquietudes que suscite maintenant la realisation periodique d’ATDM de suivi sur le plan de la radioexposition, il propose diverses strategies de suivi qui varient selon la complexite de la reparation et selon les avantages et les inconvenients relatifs de chaque modalite. Ó 2015 Canadian Association of Radiologists. All rights reserved. Key Words: Endovascular aortic aneurysm repair; Fenestrated; Surveillance; Aortic aneurysm; Computed tomography; Ultrasound; Magnetic resonance imaging; X-ray; Nuclear medicine
* Address for Correspondence: Andrew P. Winterbottom, FRCR, Department of Interventional Radiology, Addenbrooke’s Hospital, University of Cambridge, Cambridge CB2 0QQ, United Kingdom.
E-mail address: [email protected] (A. P. Winterbottom).
0846-5371/$ - see front matter Ó 2015 Canadian Association of Radiologists. All rights reserved. http://dx.doi.org/10.1016/j.carj.2014.12.003
2
A. S. Thakor et al. / Canadian Association of Radiologists Journal xx (2015) 1e14
Endovascular aortic aneurysm repair (EVAR) was introduced in 1991, as an alternative to open surgical repair of aortic aneurysms, offering lower perioperative mortality and morbidity [1]. The repair involves placing a fabric covered stent graft into the aneurysmal segment of aorta to bypass the aneurysm thereby excluding it from systemic arterial blood pressure. The stent graft scaffold can be made of stainless steel, nitinol, or elgiloy (an alloy of cobalt, chromium, and nickel) and the graft fabric made of either polyester (eg, polyethylene terephthalate) or plastic (eg, polytetrafluoroethylene). Defining the success of EVAR depends on both clinical and radiographic criteria, but ultimately depends on complete exclusion of the aneurysm from the circulation. Abdominal Aortic Stent Graft Design Aortic stent grafts are modular in construction, made up of a bifurcated main body and 1 or 2 iliac limbs. During standard EVAR, a bifurcated stent graft is placed below the renal arteries, which has a main body that then extends into each iliac artery with a tube graft limb. When the bifurcation is narrow or when there is no suitable iliac landing zone (ie, due to an aneurysmal iliac artery), an aorto uni-iliac (AUI) stent graft can be used with a surgical femorofemoral cross over graft to maintain perfusion to the excluded leg. In cases where there is an insufficient seal at either the proximal or distal landing zone, extra components can be added (ie, proximal extensions/cuffs or covered stents) to create tighter apposition between the stent graft and the aortic wall. The construction of the aortic stent grafts consist of a combination of an exostent (ie, the stent structure is located outside the graft material) and endostent (ie, the stent structure is located inside the graft material). The AFX (Endologix, Irvine, CA) stent graft is a complete endostent with only the very top and bottom of the graft material adherent to the stent, thereby allowing the graft material to expand into irregular portions of the landing zones; this is an important feature of this graft as it can mimic an endoleak on follow up imaging (Figure 1). Most stent grafts have a bare metal stent to allow for suprarenal proximal fixation (Figure 2, B-D), often with barbs to prevent migration. Three exceptions are the C3 Excluder (Gore Medical, Flagstaff, AZ) (Figure 2A), Aorfix (Lombard Medical, Irvine, CA) (Figure 2, E and F), and Anaconda (Inchinnan, Renfrewshire, Scotland) stent grafts, which have infrarenal fixation. The shape of the top of the stent graft material also varies between devices; straight (Cook Zenith, Medtronic Endurant [Medtronic, Dublin, Ireland], Endologix AFX) (Figure 2, BD), castellated (Gore C3 Excluder) (Figure 2A), or fish mouth (Lombard Aorfix and Vascutek Anaconda) (Figure 2, E and F). The Ovation (TriVascular, Santa Rosa, CA) stent graft combines a conventional type suprarenal stent with active barb fixation along with a proximal sealing ring that is inflated using a polymer. The Nellix (Endologix) is a newer device, which uses a stent graft with surrounding inflatable endobags, which are filled with polymer that mould to the surrounding aneurysm sac (Figure 3).
Standard infrarenal stent grafts require a uniform infrarenal neck length between 10-15 mm to ensure an adequate seal between the stent graft and the aortic wall, however, neck lengths as short as 4 mm can be treated successfully [2]. An insufficient neck, which is too short, too angulated (ie, greater than 60 angle between the suprarenal aorta and proximal neck) [3] or too large in diameter (ie, juxta- or suprarenal aortic aneurysms), may require fenestrated or branched aortic stent grafts. Grafts for fenestrated endovascular aortic aneurysm repair have scallops (incomplete sections of the top of the stent graft material) and/or fenestrations (holes in the stent graft material) in their main body through, which additional covered stents can be deployed into visceral arteries (ie, celiac, superior mesenteric, and renal arteries) to maintain patency. Fenestrated stent grafts are made by Cook (Zenith) (Figure 2, G and H) and Vascutek (Anaconda). Other techniques include deploying chimneys or parallel grafts, which include snorkels and periscopes (for more information, please see the review by Wilson et al [4]) (Figure 4). Snorkels are branches that arise above the proximal end of the endograft to supply branches below the proximal endograft whereas periscopes are branches that arise below the distal end of the endograft to supply branches above the distal endograft. There are also sandwich grafts, which can run between 2 endograft components. When there is an unsuitable distal landing zone in the common iliac artery (ie, short or significant aneurysmal disease of the common iliac artery), the stent graft limb can be extended into the normal calibre external iliac artery, with prior embolization of the ipsilateral internal iliac artery (IIA) to prevent an endoleak (Figure 5, A and B). In these cases, flow in the contralateral IIA should be preserved to supply the pelvic organs and hence proper evaluation of the planning computed tomography angiography (CTA) should be undertaken to identify any IIA disease. Branched iliac devices allow the limb to extend and seal in a normal calibre external iliac artery whilst the branch portion of the device preserves flow into the IIA (Figure 5, C and D). Complications from Endovascular Aortic Aneurysm Repair The most common complications from EVAR are endoleaks. There are 5 types of endoleaks based on the source/ origin of blood flow into the aneurysm sac [5]: Type I endoleak e caused by blood flowing between the stent graft and the native arterial wall at either the proximal aortic (Type Ia) or distal (Type Ib) attachment site. A Type Ic endoleak occurs with incomplete occlusion of the contralateral iliac artery in AUI stent grafts. Type I endoleaks are most commonly seen at the time of stent graft deployment. Late development of Type I endoleaks (ie, delayed endoleaks) can be due to progressive aneurysmal disease involving the infrarenal neck (Type Ia; Figure 6, A and B) or iliac landing zone (Type Ib; Figure 6, C and D).
Radiological evaluation of abdominal EVAR / Canadian Association of Radiologists Journal xx (2015) 1e14
3
Figure 1. Axial arterial phase computed tomography images showing (A) the normal appearance of contrast medium lying outside the stent (black star) but inside the graft material of the Endologix AFX stent graft as opposed to (B) the normal appearance of contrast medium contained by both graft material and stent in a Medtronic Endurant stent graft.
Type II endoleak e retrograde flow of blood through vessels, which communicate with the aneurysm sac, most typically the lumbar and inferior mesenteric artery (IMA) (Figure 7). These are the most common type of endoleak encountered in clinical practice [6] and can be divided into simple, with only 1 patent retrograde branch (Type IIa), or complex, with 2 or more patent retrograde branches (Type IIb). Type III endoleak e structural failure of the stent graft as a result of junctional separation between stent graft components in modular devices (Type IIIa) (Figure 8) or as a result of a defect/tear within the graft fabric (Type IIIb).
Type IV endoleak e although now rare, this represent graft wall porosity and are identified immediately at the time of EVAR, most commonly due to the patient’s anticoagulated state (Figure 9). Type V endoleak (also known as endotension) e the continued expansion of the aneurysm sac without direct radiologic evidence of a leak [6]. Other complications include the stent graft becoming kinked or compressed, which may cause lower limb claudication, especially if 1 of the stent graft limbs becomes occluded. Kinking of either the main body or a limb of the stent graft can occur at the time of insertion due to tortuous
Figure 2. Volume rendered computed tomography images showing the proximal fixation of the aortic stent graft in relation to the renal arteries. (A) Gore C3 Excluder with no suprarenal fixation and castellated edge to the graft. (B) Medtronic Endurant, (C) Endologix AFX, and (D) Cook Zenith aortic stent grafts all with similar suprarenal fixation stent lying across the renal artery origins with a straight top edge to the stent graft. (E) Anteroposterior (AP) and (F) lateral views of the Lombard Aorfix stent graft with no suprarenal fixation and fish mouth configuration to the proximal edge of the stent graft. (G) AP and (H) lateral view of a Cook fenestrated stent graft with supraceliac bare stent fixation, scalloped top edge to the stent graft accommodating the celiac trunk, and 3 fenestrations for the superior mesenteric artery and renal arteries.
4
A. S. Thakor et al. / Canadian Association of Radiologists Journal xx (2015) 1e14
Figure 3. Axial computed tomography (CT) images showing an Endologix Nellix aortic aneurysm sealing system with 2 stents eccentrically surrounded by polymer filled endobags sealing the aneurysm lumen. (A) noncontrast CT showing intermediate density polymer bags (black stars) and (B) arterial phase showing luminal enhancement of the stent.
angulated anatomy or can be seen later due to changes in anatomy as the aneurysm sac reduces in size. In some cases, kinking can cause the modular components of the stent graft
to disconnect thereby resulting in a Type III endoleak. Remodelling of the aneurysm sac or improper sizing of the stent graft can also result in device migration over time, which may result in a late Type IA endoleak. Follow-Up of Endovascular Aortic Stent Grafts Lifelong surveillance is recommended for EVAR since endoleaks can develop at any time after stent graft placement, with cases being reported as late as 7 years after EVAR [7]. Remodelling of the aneurysmal sac along with progressive aneurysm disease involving the proximal and distal landing zones can result in 3-dimensional anatomical changes, which can jeopardize the integrity and function of endovascular stent grafts [8]. Decrease in the size of the aneurysmal sac is a surrogate marker of clinical success following EVAR [9]. Long-term durability of the metallic and fabric components is also a concern with reported cases of stent fracture and holes in the fabric [10]. Although CTA is the gold standard to monitor patients following EVAR, the cumulative radiation dose from repeated follow-up scans is an issue [11]. Hence, alternative ways of monitoring these patients are sought, including ultrasound (US), magnetic resonance (MR) imaging, and/or plain radiographs. Computed Tomography Angiography
Figure 4. Volume rendered computed tomography image showing a chimney stent graft (white arrow) within the SMA, which lies between the aortic wall and the second abdominal aortic stent graft (the celiac axis was intentionally covered as it was already occluded from long standing thrombus).
CTA is able to assess aneurysm size, graft stenosis, thrombosis, kinking, migration, detachment of modular components, visceral or peripheral arterial occlusion or embolization (resulting in organ ischemia), and endoleaks [11]. Despite the associated ionizing radiation dose and contrast nephrotoxicity, the reproducibility and spatial resolution of CTA have made it the preferred method of surveillance. A recent survey of UK practice (81 centers) demonstrated CTA to be the modality of choice for EVAR follow-up in the first year with a trend towards using US
Radiological evaluation of abdominal EVAR / Canadian Association of Radiologists Journal xx (2015) 1e14
5
Figure 5. Contrast-enhanced (A) volume rendered and (B) axial computed tomography (CT) images showing an internal iliac artery embolization with an Amplatzer plug (white arrowhead). Contrast-enhanced (C) volume rendered and (D) axial CT images showing an internal iliac artery branch graft (white arrow).
subsequently [12], reflecting the concern regarding cumulative radiation dose and resource allocation. Many of the responders in this survey stated that multiple contrast medium enhanced CT phases were used, each adding to the cumulative radiation dose. Traditionally, most centers use a triphasic CTA protocol (unenhanced, arterial and delayed phases). The unenhanced phase is used to differentiate small endoleaks in post-contrast medium images from aortic wall calcification, intra-thrombus calcification and metallic stents. The arterial phase is usually acquired following automated bolus tracking and requires an intraarterial threshold higher than 200 Hounsfield units; it is used to assess the integrity of the graft, the patency of visceral and distal vessels, the detection of endoleaks and to exclude any thrombus. CTA has excellent performance for the detection of endoleak, with sensitivity and specificity of approximately 92% and 90%, respectively [13]. The delayed phase is acquired 60-120 seconds post IV contrast medium administration and helps to identify slow flowing endoleaks that were inconclusive on the arterial phase as it allows time for contrast medium to pool within the aneurysm sac [14] (Figure 10). These type II endoleaks are less important as rarely their discovery leads to reintervention unless there is aneurysm sac enlargement, which can be detected with US [15].
To decrease radiation burden, many centers forego the delayed phase as most endoleaks can be detected on the arterial phase. Only when there is concern of continued aneurysm expansion (ie, endotension), then delayed phase images are performed to exclude an occult endoleak. Dual energy CT may aid in decreasing the radiation dose using complex reconstruction algorithms to provide a virtual unenhanced scan [16]. Studies have shown that the unenhanced images only need to be obtained at the first CT post-EVAR and these can be used as the unenhanced controls for subsequent studies without affecting endoleak detection rates [17]. However, with the more recent trend to use US as the primary EVAR follow-up modality, CT is often used as the problem solving tool and therefore a triphasic study (unenhanced, arterial, and delayed phase) is both useful and justified given the consequences of endoleak and the prior radiation dose saving provided by US. The stent graft position should be analysed for kinks and position, especially in relation to the planned proximal and distal landing zones (Figure 11), and this is best evaluated on the scout/localizer view, or when possible, a maximum intensity projection or curvilinear reformat [18]. However, CT topograms are inferior to plain abdominal radiographs for the assessment of endograft wire fractures and positional
6
A. S. Thakor et al. / Canadian Association of Radiologists Journal xx (2015) 1e14
Figure 6. Type Ia endoleak with contrast medium between the right lateral wall of the aorta and the stent graft (white arrow) due to inferior migration of the stent graft in the neck of the aneurysm on (A) axial computed tomography (CT) and (B) multiplanar reconstruction. Left aorto uni-iliac stent graft with Type Ib endoleak with contrast medium between the distal end of iliac limb and the left common iliac artery wall (white arrow) due to dilatation of the common iliac artery on (C) axial CT and (D) multiplanar reconstruction.
changes (please see section on plain radiographs). Comparison should be made to the operative fluoroscopic images taken at the time of stent graft insertion. The stent overlap at
component junctions should also be scrutinized to look for migration. Perfusion of the visceral vessels is paramount and the size and perfusion of each kidney should be carefully
Figure 7. Type II endoleak (black star) supplied by the inferior mesenteric artery (white arrow) seen on (A) axial computed tomography and (B) volume rendered images, which shows the direct route from the superior mesenteric artery (SMA) via the middle and left colic arteries to fill the aneurysm sac via retrograde flow in the inferior mesenteric artery (distal branches of the SMA and renal arteries have been removed).
Radiological evaluation of abdominal EVAR / Canadian Association of Radiologists Journal xx (2015) 1e14
7
Figure 8. Type III endoleak (black star) seen on (A) axial computed tomography and (B) sagittal curved reformat images that demonstrates limb dislocation.
evaluated on the enhanced images, especially if an accessory renal artery was covered at the time of the EVAR. The thickness and geometry of the suprarenal struts may generate artifacts, which may hinder interpretation of the aorta at this
Figure 9. An arteriogram showing a Type IV endoleak (white arrow) due to graft material porosity during stent graft insertion.
level [19]; however, newer stent grafts made from nitinol and elgiloy cause only minor streak artifacts [20]. The aneurysm sac should then be measured and compared to the pre-EVAR CTA to ensure it is not expanding; often no change in the size of the aneurysm sac occurs during the first year following EVAR. An increase in size should alert the clinician to look for an endoleak, which can be identified as an enhancing/ high attenuation area in the aneurysmal sac. This should then be cross-referenced with the unenhanced CT to confirm the absence of calcification or high-density thrombus in the same location. Type II endoleaks are common and often require no intervention if the sac is stable or decreasing in size. The iliac stent graft limbs should then be evaluated to ensure that there has been no stent graft migration and no in-stent thrombosis [11] (Figure 12). Partial stent graft thrombosis is reported to complicate EVAR in 3%-19% of cases [21]. Occlusion within the stent graft is best seen on axial and 3D reconstructions [22]. Mycotic pseudoaneurysms are a rare complication of EVAR (0.6%-3%), which carry a high mortality rate (25%88%) [23]. CTA findings are nonspecific and include perigraft thickening and fat stranding, which may however represent the normal host response to foreign material, and perigraft air, which can also be seen immediately after EVAR [11,23]; further assessment with a nuclear medicine white blood cell scintigram may help clarify the etiology of these findings. Colonic ischemia following EVAR of infrarenal abdominal aneurysms is a rare but recognized complication (1.5%-3%) and occurs more commonly in patients who have stenosis of the superior mesenteric artery (SMA), as the IMA will be occluded by the stent graft [24]. Assessment of the mesenteric vessels on the planning CTA is therefore paramount as surgical revascularization may need to be performed as part of a combined hybrid repair. Aortic dissection caused by injury from the introduction of the stent graft delivery system has also been reported [25].
8
A. S. Thakor et al. / Canadian Association of Radiologists Journal xx (2015) 1e14
Figure 10. Computed tomography axial (A) unenhanced, (B) arterial phase, and (C) portal venous phase images showing a Type II endoleak, which is only visible in the portal venous phase.
Ultrasound US is widely available and inexpensive, lacks ionizing radiation, and is dynamic. US assessment requires no nephrotoxic contrast medium, unlike CTA [26], which is important because a significant percentage of patients with aortic aneurysmal disease have some degree of renal impairment. Furthermore, renal creatinine clearance has also
been shown to decrease by an average of 10% within the first year after EVAR [26]. Disadvantages of US include its operator dependence, patient preparation (which generally requires fasting for 6 hours prior to examination to minimize bowel gas interference), and difficulties in assessing obese patients. Ultrasound operators are required to be familiar with the different types of stent grafts used within their institution to
Figure 11. Computed tomography (A) axial and (B) sagittal images of an aorto uni-iliac aortic stent graft with subsequent change in configuration following aneurysm shrinkage resulting in a kink between the body and limb seen on corresponding computed tomography (C) axial and (D) sagittal images.
Radiological evaluation of abdominal EVAR / Canadian Association of Radiologists Journal xx (2015) 1e14
9
Figure 12. Axial computed tomography images showing (A) concentric partial limb thrombosis and (B) complete limb occlusion due to thrombosis.
appreciate their subtle differences on US. This is especially important with fenestrated stent grafts, which have overlapping modular components that need to be fully assessed for apposition and integrity. Duplex ultrasound (DUS) is able to assess for endoleaks by detecting jets of blood flow into the aneurysm sac outside of the stent graft with high sensitivity (Figure 13). The origin of the blood flow into the sac can also be accurately assessed to determine the type of endoleak; however, thrombus may obscure this. In these
circumstances contrast-enhanced ultrasound (CEUS) using microbubbles enables accurate and easy detection of any blood flow outside the stent graft and within the aneurysm sac. Currently, 34% of UK centers primarily utilise DUS as their primary surveillance modality of choice following EVAR [27,28]. In 2009 the Society for Vascular Surgery suggested that patients with stable appearances of aortic stent grafts could be monitored annually with DUS in place of CT [27].
Figure 13. Type II lumbar artery endoleak lying between the limbs of the stent graft on (A) noncontrast and (B) arterial phase computed tomography images with corresponding (C) greyscale and (D) power Doppler ultrasound images.
10
A. S. Thakor et al. / Canadian Association of Radiologists Journal xx (2015) 1e14
Figure 14. Normal (A) anteroposterior and (B) lateral plain film radiograph of a Cook Zenith aortic stent graft.
In a large systemic review and meta-analysis, which analysed 25 studies (3975 paired scans), it was found that the pooled sensitivity of DUS was 74% and the specificity was 94% for the detection of all endoleaks when compared with CTA. Although the numbers are smaller, when looking at CEUS with 11 studies (961 paired cases), the pooled sensitivity was higher at 96%, but the specificity was lower at 85% compared with CTA. When identifying Type I and III endoleaks, the pooled sensitivity and specificity for both CEUS and DUS approached 100% (28). Interestingly, if CEUS were taken as the gold standard, analysis of the data demonstrated on CTA would have a pooled sensitivity of 70% and a specificity of 98%, indicating that CEUS is more accurate in the detection of endoleaks as compared to the current gold standard of CTA. In another prospective study (108 cases) in which the same patient underwent assessment with DUS, CEUS, CTA, and MR angiography (MRA), the results showed that using MRA and CTA as the gold standards, CEUS was superior to DUS in detecting endoleaks with a higher sensitivity (93% vs 58%) and specificity (100% vs 96%). In addition, CEUS picked up 4 endoleaks missed by CTA, with subsequent review attributing the missed CTA endoleaks to the presence of the metallic artifact from the stent graft [29]. Taken together, these studies show that DUS and CEUS are an acceptable alternative to monitor abdominal aortic stent grafts following EVAR compared with CTA with the advantage of having no ionizing radiation burden.
Figure 15. Magnified (A) anteroposterior and (B) lateral plain film radiograph of a Cook fenestrated aortic stent graft with multiple radio-opaque markers indicating fenestration and scallop orientation.
Radiological evaluation of abdominal EVAR / Canadian Association of Radiologists Journal xx (2015) 1e14
11
Figure 16. Magnified (A) anteroposterior plain radiograph and (B) corresponding thick coronal reformat showing inadequate overlap of the contralateral limb in this Cook Zenith aortic stent graft with the dashed line showing the ideal minimum overlap required as opposed to the current position indicated by the solid arrow.
Magnetic Resonance Angiography MRA is used as an alternative to CTA for patients with renal insufficiency or in patients who have an iodinated contrast medium allergy. The cost and time variables associated with MRA ultimately hinder its widespread use as the acquisition times are considerably longer compared to CTA. As well as the standard magnetic resonance exclusion
criteria, which include implanted metallic devices (eg, pacemakers) and patients who suffer from claustrophobia, MRA cannot be undertaken in patients who have received aortoiliac stent grafts made from stainless steel or elgiloy components as these metals cause susceptibility artifacts, which create large signal voids around the stent graft. None MR compatible endovascular embolization coils are also a limitation. However, newer metallic alloys such as nitinol (an
Figure 17. Magnified anteroposterior plain radiograph showing a significant kink (solid white arrow) in the iliac limb at the native aortic bifurcation (A) before and (B) after correction using a self-expanding stent insertion (solid white star).
12
A. S. Thakor et al. / Canadian Association of Radiologists Journal xx (2015) 1e14
Figure 18. Axial computed tomography in the (A) arterial phase and (B) venous phase showing an infected aortic stent graft with a thick enhancing aortic wall in the venous phase. Radiolabeled white cell scans in (C) the early phase and (D) at 24 hours showing a focal increase in uptake in the 24 hour image (white arrow).
alloy of nickel and titanium) have been shown to cause relatively little artifact and are fully MR compatible [30]. Furthermore, scanning protocols such as Volume Interpolated Breathhold Examination and Spectral Adiabatic Inversion Recovery (Siemens Healthcare, Erlangen, Germany) have the potential of producing diagnostic images with reduced artifact. Contrast-enhanced MRA (CE-MRA) can be used to assess for the presence of endoleaks using time-resolved imaging sequences following the administration of gadolinium. Time-resolved angiography is a CE-MRA technique that manipulates the filling of the K-space data set in a complex temporal manner to enable maximum edge
detection (for vessel wall interrogation) during certain time domains and maximum intravascular signal (for intraluminal interrogation) [31]. Through time-resolved pulse sequences, data acquisition can be up to 60 times faster compared with acquiring conventional K-space filling methods. Given the improved temporal resolution inherent to this technique, smaller quantities of contrast medium can be used for imaging in addition to less-motion artifact [31]. Superior temporal resolution in time-resolved angiography confers advantages to other CE-MRA techniques. For example, it may have a particular utility in endoleak characterization after endovascular aneurysm repair [32,33]. Flow into the stent and then the presence of an endoleak into the sac can be
Figure 19. A flow diagram outlining when and how to use each imaging modality to assess and follow up aortic stent grafts. AP/Lat ¼ anteroposterior/lateral; CEUS ¼ contrast-enhanced ultrasound; CT ¼ computed tomography; EVAR ¼ endovascular aortic aneurysm repair.
Radiological evaluation of abdominal EVAR / Canadian Association of Radiologists Journal xx (2015) 1e14
evaluated over minutes in a dynamic fashion as opposed to the limitations of CTA [34e36]. Consideration should be taken in those patients with renal insufficiency prior to performing CE-MRA; patients with a glomerular filtration rate of <30 mL/min/1.73m2 should be investigated with alternative modalities and care should be taken in patients with a glomerular filtration rate of between 30 and 59 mL/min/ 1.73m2 due to the risk of nephrogenic systemic fibrosis. The dose of contrast in the latter group of patients should be kept as low as possible [37]. Noncontrast MR sequences using steady-state free-precession sequences can monitor the aneurysm sac for regression or expansion. The age of any thrombus within the sac can be evaluated based on the inherent signal characteristics of the thrombus composition. A recent study has shown that non-contrast MRA has high sensitivity for endoleak detection with a high negative predictive value using steady state free precession sequences [38]. MRA has also been shown to be at least as sensitive as CTA in the overall detection of endoleaks and is thought to be more sensitive than CTA in detecting Type II endoleaks [39,40]. Newer techniques such Time Resolved Echo Shared Angiography Technique combined with high injection rates (1020 mL gadolinium at 2-5 mL/second), with a temporal resolution ranging between 2-4 seconds per data set, have been shown to be useful in differentiating between Type II and Types I/III endoleaks [41]. New blood pool MR contrast agents, such as ferumoxytol, have also been shown to improve the detection of slow flow endoleaks that are occult on CTA [42]. Plain Radiographs (X-Rays) Plain radiographs provide a simple overview of stent graft position and morphology. Radiographs should be taken in 2 orthogonal projections (anteroposterior [AP] and lateral) with the addition of oblique projections if clinically indicated (Figure 14). Whilst the AP and lateral views are useful for detecting stent graft migration and component separation, the oblique views are useful in detecting stent fractures. Consistent centering of the radiographs minimizes geometric distortion and allows easier comparison between serial films. Ideally, radiographs should not be acquired immediately after a CT examination as the contrast medium excreted by the kidneys may obscure parts of the stent graft [43]. Where possible, aortic stent radiographs should be reported by an interventional radiologist who is familiar with the manufacturers stent graft appearance. Assessment should include: 1) stent graft position; 2) migration of the graft or limbs; and 3) kink in the limbs, especially at the insertion points into the main body. The radio-opaque markers on each stent graft can be identified on the plain radiographs and are useful in assessing graft migration (Figure 15). If markers are not aligned, this may signify limb migration or imminent dislocation and hence further evaluation with CT is recommended to assess for a Type III endoleak (Figure 16). Significant kinking or stenosis of the main body of the graft or the iliac
13
limbs may require reinforcement by placing an additional stent within the stent graft (Figure 17). For more complicated fenestrated stent grafts, each fenestration needs to be carefully scrutinized for kinking and to ensure the correct amount of overlap with the main body. Nuclear Medicine Nuclear medicine scans using technetium (99mTC) sulfur colloid or 99mTC-tagged red blood cell have been used to detect endoleaks, however, they are less sensitive than CTA and not routinely used in clinical practice [44]. Nuclear medicine with radiolabeled white cells or positron emission tomography-CT however can be useful in determining graft related infections (Figure 18). In patients who have undergone transarterial embolization with high attenuation material (ie, coils, glue, lipiodol, or onyx), follow up assessment using CT for endoleaks may be hindered by beam hardening artifact. In these cases, 99mTC-labeled-human serum albumin diethylenetriamine pentaacetic acid single-photon emission computed tomography has been shown to provide a valuable alternative noninvasive tool for evaluating the therapeutic effect following embolization [45]. Conclusion EVAR provides a robust alternative to open surgical repair of aortic aneurysms. Figure 19 provides an overview of how the authors evaluate and follow up aortic stent grafts within their institution. Knowledge of potential complications allows the radiologist to provide a detailed and useful assessment of aortic stent grafts using different imaging modalities and when best to use each modality based on a specific situation or to answer a specific question. References [1] Parodi JC, Palmaz JC, Barone HD. Transfemoral intraluminal graft implantation for abdominal aortic aneurysms. Ann Vasc Surg 1991;5:491e9. [2] Forbes TL, Harris JR, Lawlor DK, et al. Midterm results of the Zenith endograft in relation to neck length. Ann Vasc Surg 2010;24:859e62. [3] De Bock S, Iannaccone F, De Beule M, et al. What if you stretch the IFU? A mechanical insight into stent graft instructions for use in angulated proximal aneurysm necks. Med Eng Phys 2014;36:1567e76. [4] Wilson A, Zhou S, Bachoo P, et al. Systematic review of chimney and periscope grafts for endovascular aneurysm repair. Br J Surg 2013;100: 1557e64. [5] White GH, Yu W, May J, et al. Endoleak as a complication of endoluminal grafting of abdominal aortic aneurysms: classification, incidence, diagnosis, and management. J Endovasc Surg 1997;4:152e68. [6] 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;35:1029e35. [7] Corriere MA, Feurer ID, Becker SY, et al. Endoleak following endovascular abdominal aortic aneurysm repair: implications for duration of screening. Ann Surg 2004;239:800e5. discussion 5e7. [8] Chaikof EL, Blankensteijn JD, Harris PL, et al. Reporting standards for endovascular aortic aneurysm repair. J Vasc Surg 2002;35:1048e60. [9] Vardulaki KA, Prevost TC, Walker NM, et al. Growth rates and risk of rupture of abdominal aortic aneurysms. Br J Surg 1998;85:1674e80.
14
A. S. Thakor et al. / Canadian Association of Radiologists Journal xx (2015) 1e14
[10] Jacobs TS, Won J, Gravereaux EC, et al. Mechanical failure of prosthetic human implants: a 10-year experience with aortic stent graft devices. J Vasc Surg 2003;37:16e26. [11] Mita T, Arita T, Matsunaga N, et al. Complications of endovascular repair for thoracic and abdominal aortic aneurysm: an imaging spectrum. Radiographics 2000;20:1263e78. [12] Patel A, Edwards R, Chandramohan S. Surveillance of patients postendovascular abdominal aortic aneurysm repair (EVAR). A webbased survey of practice in the UK. Clin Radiol 2013;68:580e7. [13] Armerding MD, Rubin GD, Beaulieu CF, et al. Aortic aneurysmal disease: assessment of stent-graft treatment-CT versus conventional angiography. Radiology 2000;215:138e46. [14] Rozenblit AM, Patlas M, Rosenbaum AT, et al. Detection of endoleaks after endovascular repair of abdominal aortic aneurysm: value of unenhanced and delayed helical CTacquisitions. Radiology 2003;227:426e33. [15] Hong C, Heiken JP, Sicard GA, et al. Clinical significance of endoleak detected on follow-up CT after endovascular repair of abdominal aortic aneurysm. AJR Am J Roentgenol 2008;191:808e13. [16] Stolzmann P, Frauenfelder T, Pfammatter T, et al. Endoleaks after endovascular abdominal aortic aneurysm repair: detection with dualenergy dual-source CT. Radiology 2008;249:682e91. [17] Iezzi R, Cotroneo AR, Filippone A, et al. Multidetector CT in abdominal aortic aneurysm treated with endovascular repair: are unenhanced and delayed phase enhanced images effective for endoleak detection? Radiology 2006;241:915e21. [18] Sun Z. Multislice CT angiography in abdominal aortic aneurysm treated with endovascular stent grafts: evaluation of 2D and 3D visualisations. Biomed Imaging Interv J 2007;3:e20. [19] Stavropoulos SW, Charagundla SR. Imaging techniques for detection and management of endoleaks after endovascular aortic aneurysm repair. Radiology 2007;243:641e55. [20] Maintz D, Fischbach R, Juergens KU, et al. Multislice CT angiography of the iliac arteries in the presence of various stents: in vitro evaluation of artifacts and lumen visibility. Invest Radiol 2001;36:699e704. [21] Sakai T, Dake MD, Semba CP, et al. Descending thoracic aortic aneurysm: thoracic CT findings after endovascular stent-graft placement. Radiology 1999;212:169e74. [22] Iezzi R, Santoro M, Dattesi R, et al. Multi-detector CT angiographic imaging in the follow-up of patients after endovascular abdominal aortic aneurysm repair (EVAR). Insights Imaging 2012;3:313e21. [23] Tanner-Steinmann B, Boggian K. Vascular endograft infection with listeria monocytogenes reated with surgical debridement but without graft removal. Case Reports Med 2011;2011:482815. [24] Becquemin JP, Majewski M, Fermani N, et al. Colon ischemia following abdominal aortic aneurysm repair in the era of endovascular abdominal aortic repair. J Vasc Surg 2008;47:258e63. discussion 63. [25] Tolenaar JL, van Keulen JW, Vonken EJ, et al. Fenestration of an iatrogenic aortic dissection after endovascular aneurysm repair. J Endovasc Ther 2011;18:256e60. [26] Alsac JM, Zarins CK, Heikkinen MA, et al. The impact of aortic endografts on renal function. J Vasc Surg 2005;41:926e30. [27] Chaikof EL, Brewster DC, Dalman RL, et al. The care of patients with an abdominal aortic aneurysm: the Society for Vascular Surgery practice guidelines. J Vasc Surg 2009;50(4 Suppl):S2e49. [28] Karthikesalingam A, Al-Jundi W, Jackson D, et al. Systematic review and meta-analysis of duplex ultrasonography, contrast-enhanced
[29]
[30]
[31]
[32]
[33]
[34] [35]
[36]
[37] [38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
ultrasonography or computed tomography for surveillance after endovascular aneurysm repair. Br J Surg 2012;99:1514e23. Cantisani V, Ricci P, Grazhdani H, et al. Prospective comparative analysis of colour-Doppler ultrasound, contrast-enhanced ultrasound, computed tomography and magnetic resonance in detecting endoleak after endovascular abdominal aortic aneurysm repair. Eur J Vasc Endovasc Surg 2011;41:186e92. Wang Y, Truong TN, Yen C, et al. Quantitative evaluation of susceptibility and shielding effects of nitinol, platinum, cobalt-alloy, and stainless steel stents. Magn Reson Med 2003;49:972e6. Vessie EL, Liu DM, Forster B, et al. A practical guide to magnetic resonance vascular imaging: techniques and applications. Ann Vasc Surg 2014;28:1052e61. Lookstein RA, Goldman J, Pukin L, et al. Time-resolved magnetic resonance angiography as a noninvasive method to characterize endoleaks: initial results compared with conventional angiography. J Vasc Surg 2004;39:27e33. Williams J, Vasanawala SS. Active gastrointestinal hemorrhage identification by blood pool contrast-enhanced magnetic resonance angiography. Pediatr Radiol 2011;41:1198e200. Korosec FR, Frayne R, Grist TM, et al. Time-resolved contrastenhanced 3D MR angiography. Magn Reson Med 1996;36:345e51. Swan JS, Carroll TJ, Kennell TW, et al. Time-resolved threedimensional contrast-enhanced MR angiography of the peripheral vessels. Radiology 2002;225:43e52. Lenhart M, Framme N, Volk M, et al. Time-resolved contrast-enhanced magnetic resonance angiography of the carotid arteries: diagnostic accuracy and inter-observer variability compared with selective catheter angiography. Invest Radiol 2002;37:535e41. Daftari Besheli L, Aran S, Shaqdan K, et al. Current status of nephrogenic systemic fibrosis. Clin Radiol 2014;69:661e8. Resta EC, Secchi F, Giardino A, et al. Non-contrast MR imaging for detecting endoleak after abdominal endovascular aortic repair. Int J Cardiovasc Imaging 2013;29:229e35. Engellau L, Larsson EM, Albrechtsson U, et al. Magnetic resonance imaging and MR angiography of endoluminally treated abdominal aortic aneurysms. Eur J Vasc Endovasc Surg 1998;15:212e9. Haulon S, Lions C, McFadden EP, et al. Prospective evaluation of magnetic resonance imaging after endovascular treatment of infrarenal aortic aneurysms. Eur J Vasc Endovasc Surg 2001;22:62e9. Cohen EI, Weinreb DB, Siegelbaum RH, et al. Time-resolved MR angiography for the classification of endoleaks after endovascular aneurysm repair. J Magn Reson Imaging 2008;27:500e3. Ersoy H, Jacobs P, Kent CK, et al. Blood pool MR angiography of aortic stent-graft endoleak. AJR Am J Roentgenol 2004;182: 1181e6. Murphy M, Hodgson R, Harris PL, et al. Plain radiographic surveillance of abdominal aortic stent-grafts: the Liverpool/Perth protocol. J Endovasc Ther 2003;10:911e2. Stavropoulos SW, Itkin M, Lakhani P, et al. Detection of endoleaks after endovascular aneurysm repair with use of technetium-99m sulfur colloid and (99m)Tc-labeled red blood cell scans. J Vasc Interv Radiol 2006;17:1739e43. Nakai M, Sato H, Ikoma A, et al. The use of technetium-99m-labeled human serum albumin diethylenetriamine pentaacetic acid singlephoton emission CT scan in the follow-up of type II endoleak treatment. J Vasc Interv Radiol 2014;25:405e9.
Vascular and Interventional Radiology / Radiologie vasculaire et radiologie d’intervention
Radiological Evaluation of Abdominal Endovascular Aortic Aneurysm Repair Avnesh S. Thakor, PhD, FRCRa, James Tanner, MB, BChirb, Shao J. Ong, MBBS, PhDb, Ynyr Hughes-Roberts, FRCRb, Shahzad Ilyas, FRCRb, Claire Cousins, FRCRb, Teik C. See, FRCRb, Darren Klass, MD, FRCRc, Andrew P. Winterbottom, FRCRb,* a
Department of Radiology, Stanford University, Stanford University Medical Center, Palo Alto, California, USA Department of Interventional Radiology, Addenbrooke’s Hospital, University of Cambridge, Cambridge, United Kingdom c Department of Interventional Radiology, Vancouver General Hospital, University of British Columbia, Vancouver, British Columbia, Canada b
Abstract Endovascular aortic aneurysm repair (EVAR) is an alternative to open surgical repair of aortic aneurysms offering lower perioperative mortality and morbidity. As experience increases, clinicians are undertaking complex repairs with hostile aortic anatomy using branched or fenestrated devices or extra components such as chimneys to ensure perfusion to visceral branch vessels whilst excluding the aneurysm. Defining the success of EVAR depends on both clinical and radiographic criteria, but ultimately depends on complete exclusion of the aneurysm from the circulation. Aortic stent grafts are monitored using a combination of imaging modalities including computed tomography angiography (CTA), ultrasonography, magnetic resonance imaging, plain films, and nuclear medicine studies. This article describes when and how to evaluate aortic stent grafts using each of these modalities along with the characteristic features of several of the main stent grafts currently used in clinical practice. The commonly encountered complications from EVAR are also discussed and how they can be detected using each imaging modality. As the radiation burden from serial follow up CTA imaging is now becoming a concern, different follow-up imaging strategies are proposed depending on the complexity of the repair and based on the relative merits and disadvantages of each imaging modality. Resume La reparation endovasculaire d’un anevrisme de l’aorte (REVA) est une solution de rechange qui presente des taux de mortalite et de mesure qu’ils gagnent en morbidite perioperatoires moins eleves que le traitement des anevrismes de l’aorte par chirurgie effractive. A experience, les cliniciens entreprennent des reparations complexes dans l’environnement anatomique difficile de l’aorte a l’aide d’endoprotheses munies de fen^etres ou de branches, auxquelles peut s’ajouter une cheminee, pour assurer la perfusion des branches viscerales et l’exclusion de l’anevrisme. La reussite d’une REVA est definie par des criteres a la fois cliniques et radiographiques, mais repose en definitive sur sa capacite a completement exclure l’anevrisme de la circulation. Les endoprotheses aortiques font l’objet d’une surveillance en combinant plusieurs modalites d’imagerie, notamment l’angiographie par tomodensitometrie (ATDM), l’echographie, l’imagerie par resonance magnetique (IRM), les images radiologiques sans produits de contraste et les examens de medecine nucleaire. Le present article decrit quand et comment chaque modalite se pr^ete a l’evaluation des endoprotheses, ainsi que les caracteristiques propres aux principales endoprotheses en usage dans la pratique clinique. Il aborde egalement les complications courantes de la REVA et la fac¸on de les detecter a l’aide de chacune des modalites d’imagerie. Enfin, compte tenu des inquietudes que suscite maintenant la realisation periodique d’ATDM de suivi sur le plan de la radioexposition, il propose diverses strategies de suivi qui varient selon la complexite de la reparation et selon les avantages et les inconvenients relatifs de chaque modalite. Ó 2015 Canadian Association of Radiologists. All rights reserved. Key Words: Endovascular aortic aneurysm repair; Fenestrated; Surveillance; Aortic aneurysm; Computed tomography; Ultrasound; Magnetic resonance imaging; X-ray; Nuclear medicine
* Address for Correspondence: Andrew P. Winterbottom, FRCR, Department of Interventional Radiology, Addenbrooke’s Hospital, University of Cambridge, Cambridge CB2 0QQ, United Kingdom.
E-mail address: [email protected] (A. P. Winterbottom).
0846-5371/$ - see front matter Ó 2015 Canadian Association of Radiologists. All rights reserved. http://dx.doi.org/10.1016/j.carj.2014.12.003
2
A. S. Thakor et al. / Canadian Association of Radiologists Journal xx (2015) 1e14
Endovascular aortic aneurysm repair (EVAR) was introduced in 1991, as an alternative to open surgical repair of aortic aneurysms, offering lower perioperative mortality and morbidity [1]. The repair involves placing a fabric covered stent graft into the aneurysmal segment of aorta to bypass the aneurysm thereby excluding it from systemic arterial blood pressure. The stent graft scaffold can be made of stainless steel, nitinol, or elgiloy (an alloy of cobalt, chromium, and nickel) and the graft fabric made of either polyester (eg, polyethylene terephthalate) or plastic (eg, polytetrafluoroethylene). Defining the success of EVAR depends on both clinical and radiographic criteria, but ultimately depends on complete exclusion of the aneurysm from the circulation. Abdominal Aortic Stent Graft Design Aortic stent grafts are modular in construction, made up of a bifurcated main body and 1 or 2 iliac limbs. During standard EVAR, a bifurcated stent graft is placed below the renal arteries, which has a main body that then extends into each iliac artery with a tube graft limb. When the bifurcation is narrow or when there is no suitable iliac landing zone (ie, due to an aneurysmal iliac artery), an aorto uni-iliac (AUI) stent graft can be used with a surgical femorofemoral cross over graft to maintain perfusion to the excluded leg. In cases where there is an insufficient seal at either the proximal or distal landing zone, extra components can be added (ie, proximal extensions/cuffs or covered stents) to create tighter apposition between the stent graft and the aortic wall. The construction of the aortic stent grafts consist of a combination of an exostent (ie, the stent structure is located outside the graft material) and endostent (ie, the stent structure is located inside the graft material). The AFX (Endologix, Irvine, CA) stent graft is a complete endostent with only the very top and bottom of the graft material adherent to the stent, thereby allowing the graft material to expand into irregular portions of the landing zones; this is an important feature of this graft as it can mimic an endoleak on follow up imaging (Figure 1). Most stent grafts have a bare metal stent to allow for suprarenal proximal fixation (Figure 2, B-D), often with barbs to prevent migration. Three exceptions are the C3 Excluder (Gore Medical, Flagstaff, AZ) (Figure 2A), Aorfix (Lombard Medical, Irvine, CA) (Figure 2, E and F), and Anaconda (Inchinnan, Renfrewshire, Scotland) stent grafts, which have infrarenal fixation. The shape of the top of the stent graft material also varies between devices; straight (Cook Zenith, Medtronic Endurant [Medtronic, Dublin, Ireland], Endologix AFX) (Figure 2, BD), castellated (Gore C3 Excluder) (Figure 2A), or fish mouth (Lombard Aorfix and Vascutek Anaconda) (Figure 2, E and F). The Ovation (TriVascular, Santa Rosa, CA) stent graft combines a conventional type suprarenal stent with active barb fixation along with a proximal sealing ring that is inflated using a polymer. The Nellix (Endologix) is a newer device, which uses a stent graft with surrounding inflatable endobags, which are filled with polymer that mould to the surrounding aneurysm sac (Figure 3).
Standard infrarenal stent grafts require a uniform infrarenal neck length between 10-15 mm to ensure an adequate seal between the stent graft and the aortic wall, however, neck lengths as short as 4 mm can be treated successfully [2]. An insufficient neck, which is too short, too angulated (ie, greater than 60 angle between the suprarenal aorta and proximal neck) [3] or too large in diameter (ie, juxta- or suprarenal aortic aneurysms), may require fenestrated or branched aortic stent grafts. Grafts for fenestrated endovascular aortic aneurysm repair have scallops (incomplete sections of the top of the stent graft material) and/or fenestrations (holes in the stent graft material) in their main body through, which additional covered stents can be deployed into visceral arteries (ie, celiac, superior mesenteric, and renal arteries) to maintain patency. Fenestrated stent grafts are made by Cook (Zenith) (Figure 2, G and H) and Vascutek (Anaconda). Other techniques include deploying chimneys or parallel grafts, which include snorkels and periscopes (for more information, please see the review by Wilson et al [4]) (Figure 4). Snorkels are branches that arise above the proximal end of the endograft to supply branches below the proximal endograft whereas periscopes are branches that arise below the distal end of the endograft to supply branches above the distal endograft. There are also sandwich grafts, which can run between 2 endograft components. When there is an unsuitable distal landing zone in the common iliac artery (ie, short or significant aneurysmal disease of the common iliac artery), the stent graft limb can be extended into the normal calibre external iliac artery, with prior embolization of the ipsilateral internal iliac artery (IIA) to prevent an endoleak (Figure 5, A and B). In these cases, flow in the contralateral IIA should be preserved to supply the pelvic organs and hence proper evaluation of the planning computed tomography angiography (CTA) should be undertaken to identify any IIA disease. Branched iliac devices allow the limb to extend and seal in a normal calibre external iliac artery whilst the branch portion of the device preserves flow into the IIA (Figure 5, C and D). Complications from Endovascular Aortic Aneurysm Repair The most common complications from EVAR are endoleaks. There are 5 types of endoleaks based on the source/ origin of blood flow into the aneurysm sac [5]: Type I endoleak e caused by blood flowing between the stent graft and the native arterial wall at either the proximal aortic (Type Ia) or distal (Type Ib) attachment site. A Type Ic endoleak occurs with incomplete occlusion of the contralateral iliac artery in AUI stent grafts. Type I endoleaks are most commonly seen at the time of stent graft deployment. Late development of Type I endoleaks (ie, delayed endoleaks) can be due to progressive aneurysmal disease involving the infrarenal neck (Type Ia; Figure 6, A and B) or iliac landing zone (Type Ib; Figure 6, C and D).
Radiological evaluation of abdominal EVAR / Canadian Association of Radiologists Journal xx (2015) 1e14
3
Figure 1. Axial arterial phase computed tomography images showing (A) the normal appearance of contrast medium lying outside the stent (black star) but inside the graft material of the Endologix AFX stent graft as opposed to (B) the normal appearance of contrast medium contained by both graft material and stent in a Medtronic Endurant stent graft.
Type II endoleak e retrograde flow of blood through vessels, which communicate with the aneurysm sac, most typically the lumbar and inferior mesenteric artery (IMA) (Figure 7). These are the most common type of endoleak encountered in clinical practice [6] and can be divided into simple, with only 1 patent retrograde branch (Type IIa), or complex, with 2 or more patent retrograde branches (Type IIb). Type III endoleak e structural failure of the stent graft as a result of junctional separation between stent graft components in modular devices (Type IIIa) (Figure 8) or as a result of a defect/tear within the graft fabric (Type IIIb).
Type IV endoleak e although now rare, this represent graft wall porosity and are identified immediately at the time of EVAR, most commonly due to the patient’s anticoagulated state (Figure 9). Type V endoleak (also known as endotension) e the continued expansion of the aneurysm sac without direct radiologic evidence of a leak [6]. Other complications include the stent graft becoming kinked or compressed, which may cause lower limb claudication, especially if 1 of the stent graft limbs becomes occluded. Kinking of either the main body or a limb of the stent graft can occur at the time of insertion due to tortuous
Figure 2. Volume rendered computed tomography images showing the proximal fixation of the aortic stent graft in relation to the renal arteries. (A) Gore C3 Excluder with no suprarenal fixation and castellated edge to the graft. (B) Medtronic Endurant, (C) Endologix AFX, and (D) Cook Zenith aortic stent grafts all with similar suprarenal fixation stent lying across the renal artery origins with a straight top edge to the stent graft. (E) Anteroposterior (AP) and (F) lateral views of the Lombard Aorfix stent graft with no suprarenal fixation and fish mouth configuration to the proximal edge of the stent graft. (G) AP and (H) lateral view of a Cook fenestrated stent graft with supraceliac bare stent fixation, scalloped top edge to the stent graft accommodating the celiac trunk, and 3 fenestrations for the superior mesenteric artery and renal arteries.
4
A. S. Thakor et al. / Canadian Association of Radiologists Journal xx (2015) 1e14
Figure 3. Axial computed tomography (CT) images showing an Endologix Nellix aortic aneurysm sealing system with 2 stents eccentrically surrounded by polymer filled endobags sealing the aneurysm lumen. (A) noncontrast CT showing intermediate density polymer bags (black stars) and (B) arterial phase showing luminal enhancement of the stent.
angulated anatomy or can be seen later due to changes in anatomy as the aneurysm sac reduces in size. In some cases, kinking can cause the modular components of the stent graft
to disconnect thereby resulting in a Type III endoleak. Remodelling of the aneurysm sac or improper sizing of the stent graft can also result in device migration over time, which may result in a late Type IA endoleak. Follow-Up of Endovascular Aortic Stent Grafts Lifelong surveillance is recommended for EVAR since endoleaks can develop at any time after stent graft placement, with cases being reported as late as 7 years after EVAR [7]. Remodelling of the aneurysmal sac along with progressive aneurysm disease involving the proximal and distal landing zones can result in 3-dimensional anatomical changes, which can jeopardize the integrity and function of endovascular stent grafts [8]. Decrease in the size of the aneurysmal sac is a surrogate marker of clinical success following EVAR [9]. Long-term durability of the metallic and fabric components is also a concern with reported cases of stent fracture and holes in the fabric [10]. Although CTA is the gold standard to monitor patients following EVAR, the cumulative radiation dose from repeated follow-up scans is an issue [11]. Hence, alternative ways of monitoring these patients are sought, including ultrasound (US), magnetic resonance (MR) imaging, and/or plain radiographs. Computed Tomography Angiography
Figure 4. Volume rendered computed tomography image showing a chimney stent graft (white arrow) within the SMA, which lies between the aortic wall and the second abdominal aortic stent graft (the celiac axis was intentionally covered as it was already occluded from long standing thrombus).
CTA is able to assess aneurysm size, graft stenosis, thrombosis, kinking, migration, detachment of modular components, visceral or peripheral arterial occlusion or embolization (resulting in organ ischemia), and endoleaks [11]. Despite the associated ionizing radiation dose and contrast nephrotoxicity, the reproducibility and spatial resolution of CTA have made it the preferred method of surveillance. A recent survey of UK practice (81 centers) demonstrated CTA to be the modality of choice for EVAR follow-up in the first year with a trend towards using US
Radiological evaluation of abdominal EVAR / Canadian Association of Radiologists Journal xx (2015) 1e14
5
Figure 5. Contrast-enhanced (A) volume rendered and (B) axial computed tomography (CT) images showing an internal iliac artery embolization with an Amplatzer plug (white arrowhead). Contrast-enhanced (C) volume rendered and (D) axial CT images showing an internal iliac artery branch graft (white arrow).
subsequently [12], reflecting the concern regarding cumulative radiation dose and resource allocation. Many of the responders in this survey stated that multiple contrast medium enhanced CT phases were used, each adding to the cumulative radiation dose. Traditionally, most centers use a triphasic CTA protocol (unenhanced, arterial and delayed phases). The unenhanced phase is used to differentiate small endoleaks in post-contrast medium images from aortic wall calcification, intra-thrombus calcification and metallic stents. The arterial phase is usually acquired following automated bolus tracking and requires an intraarterial threshold higher than 200 Hounsfield units; it is used to assess the integrity of the graft, the patency of visceral and distal vessels, the detection of endoleaks and to exclude any thrombus. CTA has excellent performance for the detection of endoleak, with sensitivity and specificity of approximately 92% and 90%, respectively [13]. The delayed phase is acquired 60-120 seconds post IV contrast medium administration and helps to identify slow flowing endoleaks that were inconclusive on the arterial phase as it allows time for contrast medium to pool within the aneurysm sac [14] (Figure 10). These type II endoleaks are less important as rarely their discovery leads to reintervention unless there is aneurysm sac enlargement, which can be detected with US [15].
To decrease radiation burden, many centers forego the delayed phase as most endoleaks can be detected on the arterial phase. Only when there is concern of continued aneurysm expansion (ie, endotension), then delayed phase images are performed to exclude an occult endoleak. Dual energy CT may aid in decreasing the radiation dose using complex reconstruction algorithms to provide a virtual unenhanced scan [16]. Studies have shown that the unenhanced images only need to be obtained at the first CT post-EVAR and these can be used as the unenhanced controls for subsequent studies without affecting endoleak detection rates [17]. However, with the more recent trend to use US as the primary EVAR follow-up modality, CT is often used as the problem solving tool and therefore a triphasic study (unenhanced, arterial, and delayed phase) is both useful and justified given the consequences of endoleak and the prior radiation dose saving provided by US. The stent graft position should be analysed for kinks and position, especially in relation to the planned proximal and distal landing zones (Figure 11), and this is best evaluated on the scout/localizer view, or when possible, a maximum intensity projection or curvilinear reformat [18]. However, CT topograms are inferior to plain abdominal radiographs for the assessment of endograft wire fractures and positional
6
A. S. Thakor et al. / Canadian Association of Radiologists Journal xx (2015) 1e14
Figure 6. Type Ia endoleak with contrast medium between the right lateral wall of the aorta and the stent graft (white arrow) due to inferior migration of the stent graft in the neck of the aneurysm on (A) axial computed tomography (CT) and (B) multiplanar reconstruction. Left aorto uni-iliac stent graft with Type Ib endoleak with contrast medium between the distal end of iliac limb and the left common iliac artery wall (white arrow) due to dilatation of the common iliac artery on (C) axial CT and (D) multiplanar reconstruction.
changes (please see section on plain radiographs). Comparison should be made to the operative fluoroscopic images taken at the time of stent graft insertion. The stent overlap at
component junctions should also be scrutinized to look for migration. Perfusion of the visceral vessels is paramount and the size and perfusion of each kidney should be carefully
Figure 7. Type II endoleak (black star) supplied by the inferior mesenteric artery (white arrow) seen on (A) axial computed tomography and (B) volume rendered images, which shows the direct route from the superior mesenteric artery (SMA) via the middle and left colic arteries to fill the aneurysm sac via retrograde flow in the inferior mesenteric artery (distal branches of the SMA and renal arteries have been removed).
Radiological evaluation of abdominal EVAR / Canadian Association of Radiologists Journal xx (2015) 1e14
7
Figure 8. Type III endoleak (black star) seen on (A) axial computed tomography and (B) sagittal curved reformat images that demonstrates limb dislocation.
evaluated on the enhanced images, especially if an accessory renal artery was covered at the time of the EVAR. The thickness and geometry of the suprarenal struts may generate artifacts, which may hinder interpretation of the aorta at this
Figure 9. An arteriogram showing a Type IV endoleak (white arrow) due to graft material porosity during stent graft insertion.
level [19]; however, newer stent grafts made from nitinol and elgiloy cause only minor streak artifacts [20]. The aneurysm sac should then be measured and compared to the pre-EVAR CTA to ensure it is not expanding; often no change in the size of the aneurysm sac occurs during the first year following EVAR. An increase in size should alert the clinician to look for an endoleak, which can be identified as an enhancing/ high attenuation area in the aneurysmal sac. This should then be cross-referenced with the unenhanced CT to confirm the absence of calcification or high-density thrombus in the same location. Type II endoleaks are common and often require no intervention if the sac is stable or decreasing in size. The iliac stent graft limbs should then be evaluated to ensure that there has been no stent graft migration and no in-stent thrombosis [11] (Figure 12). Partial stent graft thrombosis is reported to complicate EVAR in 3%-19% of cases [21]. Occlusion within the stent graft is best seen on axial and 3D reconstructions [22]. Mycotic pseudoaneurysms are a rare complication of EVAR (0.6%-3%), which carry a high mortality rate (25%88%) [23]. CTA findings are nonspecific and include perigraft thickening and fat stranding, which may however represent the normal host response to foreign material, and perigraft air, which can also be seen immediately after EVAR [11,23]; further assessment with a nuclear medicine white blood cell scintigram may help clarify the etiology of these findings. Colonic ischemia following EVAR of infrarenal abdominal aneurysms is a rare but recognized complication (1.5%-3%) and occurs more commonly in patients who have stenosis of the superior mesenteric artery (SMA), as the IMA will be occluded by the stent graft [24]. Assessment of the mesenteric vessels on the planning CTA is therefore paramount as surgical revascularization may need to be performed as part of a combined hybrid repair. Aortic dissection caused by injury from the introduction of the stent graft delivery system has also been reported [25].
8
A. S. Thakor et al. / Canadian Association of Radiologists Journal xx (2015) 1e14
Figure 10. Computed tomography axial (A) unenhanced, (B) arterial phase, and (C) portal venous phase images showing a Type II endoleak, which is only visible in the portal venous phase.
Ultrasound US is widely available and inexpensive, lacks ionizing radiation, and is dynamic. US assessment requires no nephrotoxic contrast medium, unlike CTA [26], which is important because a significant percentage of patients with aortic aneurysmal disease have some degree of renal impairment. Furthermore, renal creatinine clearance has also
been shown to decrease by an average of 10% within the first year after EVAR [26]. Disadvantages of US include its operator dependence, patient preparation (which generally requires fasting for 6 hours prior to examination to minimize bowel gas interference), and difficulties in assessing obese patients. Ultrasound operators are required to be familiar with the different types of stent grafts used within their institution to
Figure 11. Computed tomography (A) axial and (B) sagittal images of an aorto uni-iliac aortic stent graft with subsequent change in configuration following aneurysm shrinkage resulting in a kink between the body and limb seen on corresponding computed tomography (C) axial and (D) sagittal images.
Radiological evaluation of abdominal EVAR / Canadian Association of Radiologists Journal xx (2015) 1e14
9
Figure 12. Axial computed tomography images showing (A) concentric partial limb thrombosis and (B) complete limb occlusion due to thrombosis.
appreciate their subtle differences on US. This is especially important with fenestrated stent grafts, which have overlapping modular components that need to be fully assessed for apposition and integrity. Duplex ultrasound (DUS) is able to assess for endoleaks by detecting jets of blood flow into the aneurysm sac outside of the stent graft with high sensitivity (Figure 13). The origin of the blood flow into the sac can also be accurately assessed to determine the type of endoleak; however, thrombus may obscure this. In these
circumstances contrast-enhanced ultrasound (CEUS) using microbubbles enables accurate and easy detection of any blood flow outside the stent graft and within the aneurysm sac. Currently, 34% of UK centers primarily utilise DUS as their primary surveillance modality of choice following EVAR [27,28]. In 2009 the Society for Vascular Surgery suggested that patients with stable appearances of aortic stent grafts could be monitored annually with DUS in place of CT [27].
Figure 13. Type II lumbar artery endoleak lying between the limbs of the stent graft on (A) noncontrast and (B) arterial phase computed tomography images with corresponding (C) greyscale and (D) power Doppler ultrasound images.
10
A. S. Thakor et al. / Canadian Association of Radiologists Journal xx (2015) 1e14
Figure 14. Normal (A) anteroposterior and (B) lateral plain film radiograph of a Cook Zenith aortic stent graft.
In a large systemic review and meta-analysis, which analysed 25 studies (3975 paired scans), it was found that the pooled sensitivity of DUS was 74% and the specificity was 94% for the detection of all endoleaks when compared with CTA. Although the numbers are smaller, when looking at CEUS with 11 studies (961 paired cases), the pooled sensitivity was higher at 96%, but the specificity was lower at 85% compared with CTA. When identifying Type I and III endoleaks, the pooled sensitivity and specificity for both CEUS and DUS approached 100% (28). Interestingly, if CEUS were taken as the gold standard, analysis of the data demonstrated on CTA would have a pooled sensitivity of 70% and a specificity of 98%, indicating that CEUS is more accurate in the detection of endoleaks as compared to the current gold standard of CTA. In another prospective study (108 cases) in which the same patient underwent assessment with DUS, CEUS, CTA, and MR angiography (MRA), the results showed that using MRA and CTA as the gold standards, CEUS was superior to DUS in detecting endoleaks with a higher sensitivity (93% vs 58%) and specificity (100% vs 96%). In addition, CEUS picked up 4 endoleaks missed by CTA, with subsequent review attributing the missed CTA endoleaks to the presence of the metallic artifact from the stent graft [29]. Taken together, these studies show that DUS and CEUS are an acceptable alternative to monitor abdominal aortic stent grafts following EVAR compared with CTA with the advantage of having no ionizing radiation burden.
Figure 15. Magnified (A) anteroposterior and (B) lateral plain film radiograph of a Cook fenestrated aortic stent graft with multiple radio-opaque markers indicating fenestration and scallop orientation.
Radiological evaluation of abdominal EVAR / Canadian Association of Radiologists Journal xx (2015) 1e14
11
Figure 16. Magnified (A) anteroposterior plain radiograph and (B) corresponding thick coronal reformat showing inadequate overlap of the contralateral limb in this Cook Zenith aortic stent graft with the dashed line showing the ideal minimum overlap required as opposed to the current position indicated by the solid arrow.
Magnetic Resonance Angiography MRA is used as an alternative to CTA for patients with renal insufficiency or in patients who have an iodinated contrast medium allergy. The cost and time variables associated with MRA ultimately hinder its widespread use as the acquisition times are considerably longer compared to CTA. As well as the standard magnetic resonance exclusion
criteria, which include implanted metallic devices (eg, pacemakers) and patients who suffer from claustrophobia, MRA cannot be undertaken in patients who have received aortoiliac stent grafts made from stainless steel or elgiloy components as these metals cause susceptibility artifacts, which create large signal voids around the stent graft. None MR compatible endovascular embolization coils are also a limitation. However, newer metallic alloys such as nitinol (an
Figure 17. Magnified anteroposterior plain radiograph showing a significant kink (solid white arrow) in the iliac limb at the native aortic bifurcation (A) before and (B) after correction using a self-expanding stent insertion (solid white star).
12
A. S. Thakor et al. / Canadian Association of Radiologists Journal xx (2015) 1e14
Figure 18. Axial computed tomography in the (A) arterial phase and (B) venous phase showing an infected aortic stent graft with a thick enhancing aortic wall in the venous phase. Radiolabeled white cell scans in (C) the early phase and (D) at 24 hours showing a focal increase in uptake in the 24 hour image (white arrow).
alloy of nickel and titanium) have been shown to cause relatively little artifact and are fully MR compatible [30]. Furthermore, scanning protocols such as Volume Interpolated Breathhold Examination and Spectral Adiabatic Inversion Recovery (Siemens Healthcare, Erlangen, Germany) have the potential of producing diagnostic images with reduced artifact. Contrast-enhanced MRA (CE-MRA) can be used to assess for the presence of endoleaks using time-resolved imaging sequences following the administration of gadolinium. Time-resolved angiography is a CE-MRA technique that manipulates the filling of the K-space data set in a complex temporal manner to enable maximum edge
detection (for vessel wall interrogation) during certain time domains and maximum intravascular signal (for intraluminal interrogation) [31]. Through time-resolved pulse sequences, data acquisition can be up to 60 times faster compared with acquiring conventional K-space filling methods. Given the improved temporal resolution inherent to this technique, smaller quantities of contrast medium can be used for imaging in addition to less-motion artifact [31]. Superior temporal resolution in time-resolved angiography confers advantages to other CE-MRA techniques. For example, it may have a particular utility in endoleak characterization after endovascular aneurysm repair [32,33]. Flow into the stent and then the presence of an endoleak into the sac can be
Figure 19. A flow diagram outlining when and how to use each imaging modality to assess and follow up aortic stent grafts. AP/Lat ¼ anteroposterior/lateral; CEUS ¼ contrast-enhanced ultrasound; CT ¼ computed tomography; EVAR ¼ endovascular aortic aneurysm repair.
Radiological evaluation of abdominal EVAR / Canadian Association of Radiologists Journal xx (2015) 1e14
evaluated over minutes in a dynamic fashion as opposed to the limitations of CTA [34e36]. Consideration should be taken in those patients with renal insufficiency prior to performing CE-MRA; patients with a glomerular filtration rate of <30 mL/min/1.73m2 should be investigated with alternative modalities and care should be taken in patients with a glomerular filtration rate of between 30 and 59 mL/min/ 1.73m2 due to the risk of nephrogenic systemic fibrosis. The dose of contrast in the latter group of patients should be kept as low as possible [37]. Noncontrast MR sequences using steady-state free-precession sequences can monitor the aneurysm sac for regression or expansion. The age of any thrombus within the sac can be evaluated based on the inherent signal characteristics of the thrombus composition. A recent study has shown that non-contrast MRA has high sensitivity for endoleak detection with a high negative predictive value using steady state free precession sequences [38]. MRA has also been shown to be at least as sensitive as CTA in the overall detection of endoleaks and is thought to be more sensitive than CTA in detecting Type II endoleaks [39,40]. Newer techniques such Time Resolved Echo Shared Angiography Technique combined with high injection rates (1020 mL gadolinium at 2-5 mL/second), with a temporal resolution ranging between 2-4 seconds per data set, have been shown to be useful in differentiating between Type II and Types I/III endoleaks [41]. New blood pool MR contrast agents, such as ferumoxytol, have also been shown to improve the detection of slow flow endoleaks that are occult on CTA [42]. Plain Radiographs (X-Rays) Plain radiographs provide a simple overview of stent graft position and morphology. Radiographs should be taken in 2 orthogonal projections (anteroposterior [AP] and lateral) with the addition of oblique projections if clinically indicated (Figure 14). Whilst the AP and lateral views are useful for detecting stent graft migration and component separation, the oblique views are useful in detecting stent fractures. Consistent centering of the radiographs minimizes geometric distortion and allows easier comparison between serial films. Ideally, radiographs should not be acquired immediately after a CT examination as the contrast medium excreted by the kidneys may obscure parts of the stent graft [43]. Where possible, aortic stent radiographs should be reported by an interventional radiologist who is familiar with the manufacturers stent graft appearance. Assessment should include: 1) stent graft position; 2) migration of the graft or limbs; and 3) kink in the limbs, especially at the insertion points into the main body. The radio-opaque markers on each stent graft can be identified on the plain radiographs and are useful in assessing graft migration (Figure 15). If markers are not aligned, this may signify limb migration or imminent dislocation and hence further evaluation with CT is recommended to assess for a Type III endoleak (Figure 16). Significant kinking or stenosis of the main body of the graft or the iliac
13
limbs may require reinforcement by placing an additional stent within the stent graft (Figure 17). For more complicated fenestrated stent grafts, each fenestration needs to be carefully scrutinized for kinking and to ensure the correct amount of overlap with the main body. Nuclear Medicine Nuclear medicine scans using technetium (99mTC) sulfur colloid or 99mTC-tagged red blood cell have been used to detect endoleaks, however, they are less sensitive than CTA and not routinely used in clinical practice [44]. Nuclear medicine with radiolabeled white cells or positron emission tomography-CT however can be useful in determining graft related infections (Figure 18). In patients who have undergone transarterial embolization with high attenuation material (ie, coils, glue, lipiodol, or onyx), follow up assessment using CT for endoleaks may be hindered by beam hardening artifact. In these cases, 99mTC-labeled-human serum albumin diethylenetriamine pentaacetic acid single-photon emission computed tomography has been shown to provide a valuable alternative noninvasive tool for evaluating the therapeutic effect following embolization [45]. Conclusion EVAR provides a robust alternative to open surgical repair of aortic aneurysms. Figure 19 provides an overview of how the authors evaluate and follow up aortic stent grafts within their institution. Knowledge of potential complications allows the radiologist to provide a detailed and useful assessment of aortic stent grafts using different imaging modalities and when best to use each modality based on a specific situation or to answer a specific question. References [1] Parodi JC, Palmaz JC, Barone HD. Transfemoral intraluminal graft implantation for abdominal aortic aneurysms. Ann Vasc Surg 1991;5:491e9. [2] Forbes TL, Harris JR, Lawlor DK, et al. Midterm results of the Zenith endograft in relation to neck length. Ann Vasc Surg 2010;24:859e62. [3] De Bock S, Iannaccone F, De Beule M, et al. What if you stretch the IFU? A mechanical insight into stent graft instructions for use in angulated proximal aneurysm necks. Med Eng Phys 2014;36:1567e76. [4] Wilson A, Zhou S, Bachoo P, et al. Systematic review of chimney and periscope grafts for endovascular aneurysm repair. Br J Surg 2013;100: 1557e64. [5] White GH, Yu W, May J, et al. Endoleak as a complication of endoluminal grafting of abdominal aortic aneurysms: classification, incidence, diagnosis, and management. J Endovasc Surg 1997;4:152e68. [6] 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;35:1029e35. [7] Corriere MA, Feurer ID, Becker SY, et al. Endoleak following endovascular abdominal aortic aneurysm repair: implications for duration of screening. Ann Surg 2004;239:800e5. discussion 5e7. [8] Chaikof EL, Blankensteijn JD, Harris PL, et al. Reporting standards for endovascular aortic aneurysm repair. J Vasc Surg 2002;35:1048e60. [9] Vardulaki KA, Prevost TC, Walker NM, et al. Growth rates and risk of rupture of abdominal aortic aneurysms. Br J Surg 1998;85:1674e80.
14
A. S. Thakor et al. / Canadian Association of Radiologists Journal xx (2015) 1e14
[10] Jacobs TS, Won J, Gravereaux EC, et al. Mechanical failure of prosthetic human implants: a 10-year experience with aortic stent graft devices. J Vasc Surg 2003;37:16e26. [11] Mita T, Arita T, Matsunaga N, et al. Complications of endovascular repair for thoracic and abdominal aortic aneurysm: an imaging spectrum. Radiographics 2000;20:1263e78. [12] Patel A, Edwards R, Chandramohan S. Surveillance of patients postendovascular abdominal aortic aneurysm repair (EVAR). A webbased survey of practice in the UK. Clin Radiol 2013;68:580e7. [13] Armerding MD, Rubin GD, Beaulieu CF, et al. Aortic aneurysmal disease: assessment of stent-graft treatment-CT versus conventional angiography. Radiology 2000;215:138e46. [14] Rozenblit AM, Patlas M, Rosenbaum AT, et al. Detection of endoleaks after endovascular repair of abdominal aortic aneurysm: value of unenhanced and delayed helical CTacquisitions. Radiology 2003;227:426e33. [15] Hong C, Heiken JP, Sicard GA, et al. Clinical significance of endoleak detected on follow-up CT after endovascular repair of abdominal aortic aneurysm. AJR Am J Roentgenol 2008;191:808e13. [16] Stolzmann P, Frauenfelder T, Pfammatter T, et al. Endoleaks after endovascular abdominal aortic aneurysm repair: detection with dualenergy dual-source CT. Radiology 2008;249:682e91. [17] Iezzi R, Cotroneo AR, Filippone A, et al. Multidetector CT in abdominal aortic aneurysm treated with endovascular repair: are unenhanced and delayed phase enhanced images effective for endoleak detection? Radiology 2006;241:915e21. [18] Sun Z. Multislice CT angiography in abdominal aortic aneurysm treated with endovascular stent grafts: evaluation of 2D and 3D visualisations. Biomed Imaging Interv J 2007;3:e20. [19] Stavropoulos SW, Charagundla SR. Imaging techniques for detection and management of endoleaks after endovascular aortic aneurysm repair. Radiology 2007;243:641e55. [20] Maintz D, Fischbach R, Juergens KU, et al. Multislice CT angiography of the iliac arteries in the presence of various stents: in vitro evaluation of artifacts and lumen visibility. Invest Radiol 2001;36:699e704. [21] Sakai T, Dake MD, Semba CP, et al. Descending thoracic aortic aneurysm: thoracic CT findings after endovascular stent-graft placement. Radiology 1999;212:169e74. [22] Iezzi R, Santoro M, Dattesi R, et al. Multi-detector CT angiographic imaging in the follow-up of patients after endovascular abdominal aortic aneurysm repair (EVAR). Insights Imaging 2012;3:313e21. [23] Tanner-Steinmann B, Boggian K. Vascular endograft infection with listeria monocytogenes reated with surgical debridement but without graft removal. Case Reports Med 2011;2011:482815. [24] Becquemin JP, Majewski M, Fermani N, et al. Colon ischemia following abdominal aortic aneurysm repair in the era of endovascular abdominal aortic repair. J Vasc Surg 2008;47:258e63. discussion 63. [25] Tolenaar JL, van Keulen JW, Vonken EJ, et al. Fenestration of an iatrogenic aortic dissection after endovascular aneurysm repair. J Endovasc Ther 2011;18:256e60. [26] Alsac JM, Zarins CK, Heikkinen MA, et al. The impact of aortic endografts on renal function. J Vasc Surg 2005;41:926e30. [27] Chaikof EL, Brewster DC, Dalman RL, et al. The care of patients with an abdominal aortic aneurysm: the Society for Vascular Surgery practice guidelines. J Vasc Surg 2009;50(4 Suppl):S2e49. [28] Karthikesalingam A, Al-Jundi W, Jackson D, et al. Systematic review and meta-analysis of duplex ultrasonography, contrast-enhanced
[29]
[30]
[31]
[32]
[33]
[34] [35]
[36]
[37] [38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
ultrasonography or computed tomography for surveillance after endovascular aneurysm repair. Br J Surg 2012;99:1514e23. Cantisani V, Ricci P, Grazhdani H, et al. Prospective comparative analysis of colour-Doppler ultrasound, contrast-enhanced ultrasound, computed tomography and magnetic resonance in detecting endoleak after endovascular abdominal aortic aneurysm repair. Eur J Vasc Endovasc Surg 2011;41:186e92. Wang Y, Truong TN, Yen C, et al. Quantitative evaluation of susceptibility and shielding effects of nitinol, platinum, cobalt-alloy, and stainless steel stents. Magn Reson Med 2003;49:972e6. Vessie EL, Liu DM, Forster B, et al. A practical guide to magnetic resonance vascular imaging: techniques and applications. Ann Vasc Surg 2014;28:1052e61. Lookstein RA, Goldman J, Pukin L, et al. Time-resolved magnetic resonance angiography as a noninvasive method to characterize endoleaks: initial results compared with conventional angiography. J Vasc Surg 2004;39:27e33. Williams J, Vasanawala SS. Active gastrointestinal hemorrhage identification by blood pool contrast-enhanced magnetic resonance angiography. Pediatr Radiol 2011;41:1198e200. Korosec FR, Frayne R, Grist TM, et al. Time-resolved contrastenhanced 3D MR angiography. Magn Reson Med 1996;36:345e51. Swan JS, Carroll TJ, Kennell TW, et al. Time-resolved threedimensional contrast-enhanced MR angiography of the peripheral vessels. Radiology 2002;225:43e52. Lenhart M, Framme N, Volk M, et al. Time-resolved contrast-enhanced magnetic resonance angiography of the carotid arteries: diagnostic accuracy and inter-observer variability compared with selective catheter angiography. Invest Radiol 2002;37:535e41. Daftari Besheli L, Aran S, Shaqdan K, et al. Current status of nephrogenic systemic fibrosis. Clin Radiol 2014;69:661e8. Resta EC, Secchi F, Giardino A, et al. Non-contrast MR imaging for detecting endoleak after abdominal endovascular aortic repair. Int J Cardiovasc Imaging 2013;29:229e35. Engellau L, Larsson EM, Albrechtsson U, et al. Magnetic resonance imaging and MR angiography of endoluminally treated abdominal aortic aneurysms. Eur J Vasc Endovasc Surg 1998;15:212e9. Haulon S, Lions C, McFadden EP, et al. Prospective evaluation of magnetic resonance imaging after endovascular treatment of infrarenal aortic aneurysms. Eur J Vasc Endovasc Surg 2001;22:62e9. Cohen EI, Weinreb DB, Siegelbaum RH, et al. Time-resolved MR angiography for the classification of endoleaks after endovascular aneurysm repair. J Magn Reson Imaging 2008;27:500e3. Ersoy H, Jacobs P, Kent CK, et al. Blood pool MR angiography of aortic stent-graft endoleak. AJR Am J Roentgenol 2004;182: 1181e6. Murphy M, Hodgson R, Harris PL, et al. Plain radiographic surveillance of abdominal aortic stent-grafts: the Liverpool/Perth protocol. J Endovasc Ther 2003;10:911e2. Stavropoulos SW, Itkin M, Lakhani P, et al. Detection of endoleaks after endovascular aneurysm repair with use of technetium-99m sulfur colloid and (99m)Tc-labeled red blood cell scans. J Vasc Interv Radiol 2006;17:1739e43. Nakai M, Sato H, Ikoma A, et al. The use of technetium-99m-labeled human serum albumin diethylenetriamine pentaacetic acid singlephoton emission CT scan in the follow-up of type II endoleak treatment. J Vasc Interv Radiol 2014;25:405e9.