British Journal of Radiology 75 (2002),700-707 © 2002 The British Institute of Radiology
Complications following endovascular abdominal aortic aneurysm repair
R Magennis, DMRD, FRCR
1
E Joekes, MD
1
J Martin, DCR(R), PG Dip, MSc
2
D White, DCR(R)
1 and
R G McWilliams, FRCS, FRCR
1
Departments of 1 Radiology and 2 Vascular Surgery, Royal Liverpool University Hospital, Prescot Street, Liverpool L7 8XP, UK
 |
Abstract
|
|---|
There is a growing population of patients who have undergone endovascular abdominal aortic aneurysm repair (EVAR) and thus there is an increasing likelihood that radiologists who are unfamiliar with this technique and its complications will have to report radiological investigations on one of these patients. The purpose of this review is to describe and illustrate the normal and abnormal radiological appearances after EVAR on plain radiography, ultrasound and CT.
|
Introduction
|
|---|
EVAR was first reported in 1989, gained popularity in the 1990s [1] and is now the subject of a large, randomized trial in the UK.
Aneurysms are treated by EVAR using an endoluminal device that combines metallic stentsand graft material. The concept is easily grasped by referring to Figure 1
. Most devices aremodular, two-piece bifurcated devices. Other graft configurations may be used: some bifurcated devices have a unitary design with ingenious mechanisms for delivery; others come in three pieces with a body and two separate limbs to connect to this; patients may also be treated with a tube device from the aorta to one iliac artery, with endovascular occlusion of the contralateral iliac limb and a surgical femorofemoral cross-over graft (Figure 2). Aorto-aortic tube devices have been used where the distal point of attachment isin the aorta just above the aortic bifurcation. These devices have performed poorly owing to loss of seal in the distal aorta, and they are rarely used.
View larger version (194K):
[in this window]
[in a new window]
|
Figure 1. Computer-generated image illustrating a modular, two-piece bifurcated graft. The main component comprises a body and one limb, introduced from the left femoral artery in this case (curved arrows). There is a short limb in the main component on the sideopposite to that of introduction (straight arrow). An extension limb (open arrows) is docked with this short limb to complete the bifurcated device.
|
|
View larger version (128K):
[in this window]
[in a new window]
|
Figure 2. Plain anteroposterior abdominal radiograph showing a stent-graft passing from the aorta into the right common iliac artery; this is an aorto-uni-iliac device. A separate component covered with fabric is used to occlude the left common iliac artery (arrow).
|
|
Some devices have stents only at the attachment sites, and others are stented throughout. The graft material may be on the inside or the outside of the stent system, and it is generally attached to the stent system throughout the length of the device. In one of the devices currently used, the Powerlink, Endologix, the fabric is on the outside of the stent skeleton and is attached only at the ends of the device. The fabric may thus billow away from the stent creating a false impression of perigraft flow (Figure 3). An understanding of the device used is important when attempting to interpret follow-up images.
View larger version (127K):
[in this window]
[in a new window]
|
Figure 3. Axial contrast enhanced CT showing a well defined, contrast-filled space surrounding the metal frame of this stent-graft (arrows). In this device the fabric is on the outside of the stent and is sutured only at the proximal and distal ends. The fabric thus billows away from the stent giving the false impression of perigraft flow.
|
|
|
Technique
|
|---|
The deployment technique for a two-piece modular device involves passage through one femoral artery of a sheath containing a body andthe ipsilateral iliac limb. The superior end ofthe body is positioned just below the lowermost renal artery. The contralateral iliac limb isinserted through the opposite femoral artery and is overlapped with a short stump in the bodyof the device. The end result is a percutaneous Y-graft, with some important differences: the attachment sites in the infrarenal aortic neck, iliac arteries and within the graft are not sutured and the aortic side branches are not ligated.
Since the attachment sites are not hand-sutured, they rely on the radial force of the stents to provide a haemostatic seal. Fixation at the attachment sites is also important to prevent migration of the stent-graft. Some manufacturers achieve this by adding hooks or barbs on the infrarenal stent. Some devices incorporate a bare suprarenal stent with or without hooks. This suprarenal stent may extend as high as the superior mesenteric artery (SMA).
The aim of treatment is to prevent aneurysm rupture, and this is dependent on exclusion of theaneurysm from blood flow and pressurization. Failure of treatment occurs if the aneurysm ruptures or remains at risk of rupture.
|
Complications following EVAR
|
|---|
The main problems following EVAR can be anticipated from the nature of the procedure andmaterials.
Migration
As the attachment sites are not sutured, theremay be movement at the aortic or iliac attachments or at points of connection within modular grafts.
Endoleak
Flow in the aortic aneurysm sac outside the stent-graft is called endoleak. The term endoleak was coined to distinguish this perigraft flow fromaortic rupture with retroperitoneal or intraperitoneal leak [2]. If the sac remains pressurized because of blood flow then it may enlarge and rupture [3].
Classification of endoleak is based on the source of the blood flow in the perigraft space.The most important types of endoleak are Types 1, 2 and 3. Type 1 endoleak refers toperigraft flow at the proximal aortic or iliac attachment sites; Type 2 endoleak refers to flowinto the perigraft space from aortic side branches; Type 3 endoleak refers to perigraft flowarising from a fabric tear at any point of thedevice, or owing to graft disconnection of modular (2-piece) or multi-modular (3-piece) devices.
Aortic side branches, which are not ligated during the procedure, may allow blood to flow into and out of the sac after EVAR. The relevant side branches include lumbar arteries, the inferior mesenteric artery (IMA) and accessory renal arteries. The most common branches causing Type 2 endoleak are the fourth lumbar arteries and the IMA.
Structural failure and graft distortion
Metals subjected to aortic pulsation pressures may fracture. The iliac limbs of the device may become distorted and angulated within the residual space of the aneurysm.
|
Imaging appearances
|
|---|
Plain radiography
Most devices are bifurcated and are composed of two modular pieces (Figure 4). Embolisation coils may be seen following adjunctive embolisation of aortic side branches or an internal iliac artery.
View larger version (114K):
[in this window]
[in a new window]
|
Figure 4. Anteroposterior plain radiograph showing a bifurcated stent-graft. There are radio-opaque markers at the top and bottom of the short limb of this bifurcated device (solid arrows). The third pair of radio-opaque markers is at the top of the limb (open arrow), which has been introduced from the right groin to overlap this short leg, completing the bifurcated device. Embolisation coils can be seen at the origin of the inferior mesenteric artery (curved arrow).
|
|
Important questions at plain radiography are:- Is there evidence of structural failure (Figure 5)?
- Is there evidence of migration of the device ateither the aortic or iliac attachment zones (Figure 6)? Comparable previous images are necessary to make this diagnosis unless there is evidence of gross movement. The most useful film in follow-up is the lateral abdominal radiograph. As the aorta and lumbar spine usually lie in the same plane, useful comparisons of position can still be made despite different centering points. If the proximal stent can be seen to have migrated distally on the lateral view then migration has occurred.
- Is there severe distortion of the graft (Figure 7)?
- Is there evidence of modular disconnection ofthe two pieces of a bimodular device (Figure 8)?
View larger version (101K):
[in this window]
[in a new window]
|
Figure 5. (a) The arrows point to fractures within the stent rings in the upper body of this device. (b) Thearrowpoints to separation of the stents within the body of this device owing to breakage of sutures designedtohold these components together. The apices of the stents may ultimately erode the fabric in this situation.
|
|
View larger version (96K):
[in this window]
[in a new window]
|
Figure 6. (a) Lateral abdominal radiograph showing the superior extent of this stent-graft lying at the level of the L1/2 disc (arrow). This stent-graft is composed of individual nitinol wires attached to the graft material. The individual serpentine graft wires are not sutured together unlike the device shown in Figure 5b . The separation of thestents in this device is a normal feature and is not owing to suture breakage. (b) 2 years later the stent-graft has migrated distally and the top stent now lies at the level of the lower L2 vertebral body (arrow). There is angulation of the device anteriorly owing to some buckling as the proximal end migrated distally.
|
|
View larger version (120K):
[in this window]
[in a new window]
|
Figure 7. This device has become markedly distorted. The marker at the distal end of the right limb can now be seen to lie within the aneurysm sac (arrow). This is owing to dislocation of this component from its attachment site in the right common iliac artery.
|
|
View larger version (107K):
[in this window]
[in a new window]
|
Figure 8. There has been separation at the connection between the components of this modular bifurcated device. There is no longer continuity of the stent system between the radio-opaque markers, which can be seen to have separated (arrows).
|
|
Ultrasound
At ultrasound the upper end of the stent-graft should lie close to the renal arteries. If these cannot be visualized then the relationship of the device to the SMA should be sought. Increasing distance between these fixed anatomical landmarks and the upper end of the device indicates distal migration of the device.
Flow should be detected in both limbs and can be detected through the metal stent skeleton of the device (Figure 9). Areas of turbulence on colour imaging with velocity increase on spectral Doppler may be owing to a haemodynamically significant kink in the stent-graft. This is associated with graft limb occlusion.
View larger version (88K):
[in this window]
[in a new window]
|
Figure 9. Colour is clearly seen within the limbs of this bifurcated device. There is no flow in the surrounding thrombus.
|
|
Endoleak may be seen at colour Doppler ultrasound (Figure 10a) [4] and this is an important observation as there is a risk of fatal sac rupture from untreated endoleak. Spectral Doppler confirmation of endoleak is recommended (Figure 10c), as artefactual colour spillover into adjacent anechoic thrombus occurs andmay lead to a false positive diagnosis of endoleak [5].
View larger version (90K):
[in this window]
[in a new window]
|
Figure 10. (a) Axial colour Doppler image at the level of the proximal aorta. Colour can be seen outside the limits of the stent-graft (arrows), in keeping with endoleak. (b) CT confirms the presence of endoleak with a contrast blush within the aneurysm outside the confines of the stent-graft (curved arrows). (c) Spectral Doppler confirms the presence of arterial flow outside the stent-graft. The arrow points to the Doppler cursor, which is in the aneurysm sac behind the stent-graft.
|
|
CT
Endoleak is recognized by the accumulation ofcontrast medium in the aneurysm sac outside the stent-graft. Contrast enhanced spiral CT timed for the arterial phase is the method of choice for detecting endoleak, although some slow-filling side branch endoleaks may not be seenuntil a later phase after contrast medium injection. The distinction between graft-related and side branch-related endoleak is based on the relationship of the accumulated contrast medium to the aortic wall and its side branches and to the stent-graft and its attachment sites [6]. Both types of endoleak may co-exist where the aortic side branches are acting as the outflow for an endoleak due to a graft-related problem. Examples ofproximal Type 1 (Figure 11), distal Type 1 (Figure 12) and Type 2 endoleak (Figure 13) are given.
View larger version (105K):
[in this window]
[in a new window]
|
Figure 11. Axial contrast enhanced CT at the level of the proximal aneurysm. There is a large endoleak related to the proximal graft (arrows). This is a Type 1, or graft-related, endoleak.
|
|
View larger version (142K):
[in this window]
[in a new window]
|
Figure 12. There is endoleak at the level of the distal aortic sac owing to incomplete seal of the right iliac limb (arrow). This is a distal Type 1, graft-related endoleak.
|
|
View larger version (54K):
[in this window]
[in a new window]
|
Figure 13. (a) There is a contrast blush arising from the anterolateral aspect of the aorta on the left (curved arrow) and a small area of calcification at this level. This is the origin of the inferior mesenteric artery. This is a type 2, side branch endoleak. There is also some peri-aneurysmal fibrosis (solid arrow), which has obstructed the right ureter (open arrow). (b) There is a faint blush between the limbs of the stent-graft (curved arrow) at the level of the left L4 lumbar artery (straight arrows). This was confirmed angiographically as a Type 2 lumbar endoleak.
|
|
Endoleak should be distinguished from areas oflongstanding increased density in the mural thrombus. This distinction may be difficult and may require an unenhanced study at a later date. It is generally possible to overcome this uncertainty if images before EVAR are available (Figure 14).
View larger version (60K):
[in this window]
[in a new window]
|
Figure 14. (a) There is calcification within the mural thrombus posteriorly on the left before endovascular repair (arrow). (b) There is calcification and high density within the thrombus outside the stent-graft (arrow). This should not be confused with endoleak, as this conforms to the appearances on the pre-operative image.
|
|
The aneurysm may continue to grow even in the absence of demonstrable endoleak and this is considered to be a treatment failure with risk of rupture. It is well documented that the aneurysm sac may remain under pressure after endovascular repair, a condition described as "endotension", even in the absence of demonstrable endoleak [7]. Sac diameter measurements are an important component of follow-up. Sac diameter reduction after EVAR is considered a definite indication oftreatment success. If the diameter remains static then we remain unclear as to whether the treatment has succeeded by arresting growth or failed in that there has been no shrinkage. A sac diameter increase of greater than 5 mm is a worrying observation that should lead to specialist review.
When measuring the aortic diameter significant interobserver variations occur owing to the slice chosen for measurement and the orientation—anteroposterior, transverse or oblique—of the "diameter". Many aneurysms have non-geometric shapes with asymmetric bulges, and a range of "diameters" may be chosen for measurement (Figure 15). Ideally, previous images will be reviewed and comparable diameters at comparable axial levels will be taken.
View larger version (164K):
[in this window]
[in a new window]
|
Figure 15. This aneurysm has an irregular shape and there is no true diameter of this non-geometric shape. A range of diameters may be selected: anteroposterior, transverse or oblique. The resulting diameter varies from 7.1 cm to 8.3 cm. Consistency in the choice of diameter is necessary to avoid the false impression of aneurysm growth on follow-up studies.
|
|
There are other issues in surveillance following EVAR, but we have attempted to highlight the more important complications that may be encountered outside the specialist unit. If investigation outside a specialist unit reveals a problem relating to EVAR, then consultation with the unit where the original procedure was performed is recommended.
Received for publication August 7, 2001.
Revision received November 12, 2001.
Accepted for publication November 21, 2001.
|
References
|
|---|
- Parodi JC, Palmaz JC, Barone HD. Transfemoral intraluminal graft implantation for abdominal aortic aneurysms. Ann Vasc Surg 1991;5:491–9.[Medline]
- White GH, Yu W, May J, Chaufour X, Stephen MS. Endoleak as a complication of endoluminal grafting of abdominal aortic aneurysms: classification, incidence, diagnosis and management. J Endovasc Surg 1997;4:152–68.[Medline]
- Lumsden AB, Allen RC, Chaikof EL, et al. Delayed rupture of aortic aneurysms following endovascular stent grafting. Am J Surg 1995;170:174–8.[Medline]
- McWilliams RG, Martin J, White D, et al. Use of contrast-enhanced ultrasound in follow up after endovascular aortic aneurysm repair. J Vasc Interv Radiol 1999;10:1107–14.[Medline]
- Sato DT, Goff CD, Gregory RT, et al. Endoleak after aortic stent graft repair: diagnosis by color duplex ultrasound scan versus computed tomography scan. J Vasc Surg 1998;28:657–63.[Medline]
- Görich J, Rilinger N, Sokiranski R, et al. Leakages after endovascular repair of aortic aneurysms: classification based on findings at CT, angiography and radiography. Radiology 1999;213:767–72.[Abstract/Free Full Text]
- Gilling-Smith GL, Brennan JA, Harris PL, Bakran A, Gould DA, McWilliams RG. Endotension after endovascular aneurysm repair: definition, classification and strategies for surveillance and intervention. J Endovasc Surg 1999;6:305–7.[Medline]
Top
Abstract
Introduction
