Introduction
An abdominal aortic aneurysm (AAA) is defined as a localized dilatation of the abdominal
aorta of 3 cm or greater in sagittal or transverse orientation or, alternatively,
as a maximum infrarenal aortic diameter of at least 1.5 times larger than the expected
diameter of the normal adjacent infrarenal aortic segment [1 ]. Screening studies have reported a prevalence for AAA of approximately 4–8 % [2 ]
[3 ]
[4 ]
[5 ].
Men are four to six times more frequently affected than women [5 ]
[6 ]
[7 ]. Further risk factors for developing AAA include age > 60 years, tobacco use, hyperlipidemia,
hypertension, chronic obstructive pulmonary disease (COPD), coronary, cerebrovascular,
or peripheral arterial disease, and family history of AAA [8 ]
[9 ]. Moreover, people of Caucasian race are at higher risk of developing an AAA [10 ].
The pathophysiology of the atherosclerotic AAA is multifactorial and not yet fully
understood. In short, preexisting atherosclerotic wall changes lead to the migration
of inflammatory cells. These initial changes promote the degradation of elastic collagen
fibers and apoptosis of vascular smooth muscle cells by expression of matrix metalloproteases
and secretion of proinflammatory cytokines [11 ]
[12 ]. Weakening of the arterial wall results in localized dilatation of the aorta. The
dilatation leads to blood flow stasis with formation of turbulent blood flow profiles,
favoring formation of an intraluminal thrombus (ILT) [13 ]. Thick ILT may cause local hypoxia of the vessel wall, promoting local inflammatory
processes and resulting in further weakening of the arterial vessel wall [14 ]. However, from a biomechanical perspective, the formation of a mural thrombus is
thought to reduce the risk of rupture by reducing aortic wall shear stress of blood
flow [15 ].
AAA is associated with the risk of dissection and consecutive aortic rupture as a
life-threatening complication. According to Laplace’s law, aortic wall stress is proportional
to the vessel radius. Rupture occurs when the aortic wall stress exceeds the tensile
strength of the aortic wall. Clinical data showed an exponential association between
aortic diameter and rupture risk [16 ]. For example, the risk of rupture was estimated at 10 % per year for AAA from 5.5
to 6.9 cm in diameter but > 33 % per year for AAA ≥ 7 cm in diameter [17 ].
According to their morphology, AAAs can be further classified as saccular or fusiform
aneurysms. Other image-based classifications consider the presence of an intraluminal
thrombus (ILT) or possible involvement of the renal arteries (infrarenal AAA vs. juxtarenal
or pararenal). The latter is particularly relevant for planning of an interventional
procedure, as treatment of juxta- and pararenal AAA is performed using branched or
fenestrated endoprostheses [18 ].
Urgent repair is recommended in symptomatic patients. Indications for elective aortic
aneurysm repair in asymptomatic patients with AAA are aneurysm growth of more than
5 mm in 6 months, or a maximum aortic diameter of at least 5.5 cm in men or 5.0 cm
in women, respectively [19 ].
Therefore, monitoring of the maximum aortic diameter by imaging techniques (ultrasound,
contrast-enhanced computed tomography (CT), or magnetic resonance angiography (MRA))
are an integral part of the clinical management of patients with AAA.
Technical and practical considerations of different aortic imaging techniques
Ultrasound
Abdominal ultrasound is the mainstay imaging modality for AAA screening and surveillance
due to its noninvasiveness, low cost, wide availability, high sensitivity, and high
specificity [20 ]
[21 ]. Ultrasound allows accurate assessment of abdominal aortic morphology such as aortic
diameter, kinking, and involvement of the renal as well as iliac artery branches.
In addition, it enables assessment of the presence of an ILT, which has been associated
with rapid aneurysm growth and early AAA rupture [22 ]
[23 ]. An ILT is present in more than 75 % of all large aneurysms requiring surgical intervention
and in about half of smaller AAAs [22 ].
Ultrasound screening programs
One-time ultrasound screening for AAAs in men or women 65 to 75 years of age with
a history of tobacco use is recommended in current practice guidelines [19 ]. In Germany, men over the age of 65 with statutory health insurance have been entitled
to one-time ultrasound screening of the abdominal aorta since 2018. Similar programs
were established earlier in other countries. In the United States, Medicare has covered
one-time AAA ultrasound screening at age 65 for men who have smoked at least 100 cigarettes
and for women with a family history of AAA disease since 2007 [24 ]. Since then, AAA screening programs have been introduced in the United Kingdom,
Scotland, and Sweden [25 ]
[26 ].
According to the current practice guidelines of the Society for Vascular Surgery,
surveillance ultrasound imaging is recommended at 3-year intervals for patients with
an AAA between 3.0 and 3.9 cm in diameter, at 12-month intervals for patients with
an AAA with a diameter of 4.0 to 4.9 cm, and at 6-month intervals for patients with
an AAA with a diameter of 5.0 to 5.4 cm [19 ].
Ultrasound – technical considerations
B-mode ultrasound is the method of choice for anatomical visualization. Typically,
low-frequency curved transducers (2–5 MHz) are used. An expansion of the frequency
range to 1 MHz can be helpful in obese patients. Slight inspiration assists in the
visualization of the proximal portion of the abdominal aorta. Color duplex ultrasound
provides additional information on aortic flow characteristics and may occasionally
be helpful to differentiate between perfused lumen and mural thrombus.
The maximum AAA diameter is measured orthogonally in the anteroposterior (AP) and
transverse plane ([Fig. 1 ]). When compared with diameter measurements derived from computed tomography angiography
(CTA), systematic underestimation has been reported for ultrasound [27 ]
[28 ]. In addition to modality-associated influences, examiner-dependent influences, such
as the angulation of the transducer and patient-dependent (physiological) pulsation-related
deviations of up to 4 mm between systole and diastole add to variations in the assessment
of the aortic diameter [29 ].
Fig. 1 Abdominal ultrasound in an abdominal aortic aneurysm (AAA) patient with indication
for operation. Transverse (A ) and sagittal (B ) cross-section of a 69-year-old male patient with a circumferentially thrombosed
infrarenal AAA. The intraluminal thrombus presents as a layered circular structure
with mixed echogenicity. Ultrasound demonstrated a maximal aneurysm diameter of 5.6 cm.
In the sagittal cross-section, the cranial end of the aorta should always be displayed
on the left side of the monitor, the caudal end on the right.
Although there are no generally accepted guidelines whether to use an outer-to-outer
or inner-to-inner diameter measurement, previous studies have demonstrated a consistent
and significant underestimation of inner-to-inner-measurements by 4–5 mm on average
[30 ]
[31 ]. [Fig. 2 ] shows ultrasound-guided determination of the maximum infrarenal diameter in an exemplary
patient with size-progressive AAA by measuring the outer-to-outer diameter.
Fig. 2 Ultrasound-based follow-up of an AAA. A Transverse ultrasound in a 78-year-old male AAA patient with a semicircular posterior
intraluminal thrombus. Note that the posterior wall caliper must be positioned on
the outside of the dorsal intraluminal thrombus, rather than the inside, to correctly
determine aortic size. B Long-term follow-up in the same patient after 3.5 years reveals progression of the
aortic aneurysm diameter from 3.3 cm to 5.1 cm.
AAA imaging using three-dimensional (3D) ultrasound is largely independent of the
angulation of the transducer and may thus improve comparability of diameter assessment
[32 ]
[33 ]. In addition, 3D ultrasound makes it possible to quantify the aortic volume, which
has been proposed as a more sensitive indicator of AAA progression compared to using
only the diameter [34 ]
[35 ].
Contrast-enhanced ultrasound is a complementary tool for AAA characterization. Recent
studies have demonstrated its potential as a sensitive tool for endoleak surveillance
following endovascular aneurysm repair (EVAR), which was comparable to that of CTA
[36 ].
Another drawback of ultrasound is its decreased reproducibility of AAA diameter measurements,
which is often compromised by examiner dependence. Furthermore, the limited field
of view can result in missed contained ruptures. If clinically suspected, clarification
by CT or MRI should always be performed. An example of a symptom-free female patient
with known AAA under ultrasound monitoring who presented to our institution is shown
in [Fig. 3 ]. The chronic contained rupture shown on MR and CT imaging had not been detected
until then despite repeated previous ultrasound examinations.
Fig. 3 MRI and CT of a chronic contained AAA rupture in a symptom-free 63-year-old female
that was missed during repeated previous ultrasound monitoring examinations. A chronic
contained rupture with hematoma in the left major psoas muscle (arrowheads) is shown
on A axial T2-weighted MRI as well as on B axial portal-venous CT. The chronic covered AAA had not been known until then despite
repeated previous ultrasound monitoring examinations. Upon request, the patient reported
a sudden abdominal pain event about 10 years prior.
Computed tomography angiography
CTA provides thin-sliced, three-dimensional, high-resolution anatomical images, which
are of particular importance for surgical planning. CTA makes it possible to determine
AAA diameter, thrombus burden, and aneurysm morphology, i. e., assessment of the aortic
neck angle, landing zone, diameter and length, and involvement of aortic side branches
as well as pelvic arteries. Moreover, CT can be useful to determine calcium burden
of the aortic wall [27 ]. High calcium burden has been proposed as a risk marker for a worse overall cardiovascular
outcome, i. e., risk factor for AAA rupture [37 ] and both higher overall and cardiovascular mortality [37 ] in AAA patients. Moreover, CTA enables an exact determination of the position of
existing implants containing metal.
Scan volume should cover supraaortic vessels, the entire aorta as well as the pelvic
axis to visualize concomitant aneurysms and relevant stenoses of the commonly applied
transfemoral access pathway for endovascular aortic repair (EVAR). CTA should be acquired
with a slice thickness of ≤ 1 mm to allow for exact multiplanar reconstruction and
thus planning of endovascular procedures.
Owing to its high spatial resolution and large scan volume, CTA detects incidental
findings of varying clinical relevance in the thorax or abdomen more frequently. It
was reported that 2/3 of AAA patients had findings of immediate or potential clinical
relevance. Of these, the most common were colorectal and lung tumors, bladder wall
thickening, and pneumonia. The incidence of malignant findings at the time of the
initial CTA scan was 6.5 % [38 ].
However, disadvantages of CTA include its increased cost compared to ultrasound, ionizing
radiation exposure, and requirement of intravenous contrast for adequate evaluation
of aortic morphology. The use of iodinated contrast agents can cause adverse reactions
including hyperthyroidism, acute renal failure, and allergic reactions.
CTA – technical considerations
The current diagnostic reference level set by the German Federal Office for Radiation
Protection (BfS) for CTA of the entire aorta is 10 CTDIvol [mGy] [39 ]. Patient doses for CTA can vary significantly depending on the CT system, scanning
parameters, and protocol used. Advanced iterative reconstruction algorithms make it
possible to obtain high-quality images with < 5 mGy of radiation exposure by reducing
the tube voltage [40 ]
[41 ].
Most CTA protocols of the aorta include multiphase imaging [42 ]. However, for preoperative imaging purposes (without clinical suspicion of a contained
rupture), performing thin-sliced single arterial phase imaging (slice thickness ≤ 1 mm)
is sufficient for EVAR planning [25 ].
CTA-based assessment of the maximum aortic diameter
Precise assessment of the maximum aneurysm diameter is pivotal since the AAA represents
one of the surgical conditions in which size is such a critical factor in the timing
of surgery.
Diameter measurements performed on axial images become unreliable when aortic angulations
are > 25° because tortuous vessels result in correspondingly elliptical cross-sections
on axial images, resulting in overestimation of AAA size [43 ]. Therefore, determination of the maximum aneurysm diameter derived from CT imaging
based on an outer-to-outer wall measurement perpendicular to the centerline of the
aorta is considered the gold standard by current practice guidelines of the Society
for Vascular Surgery [19 ]. 3D imaging using multiplanar reformation may allow correction for vessel angulation.
Many vascular centers use dedicated vessel imaging software with automatic centerline
positioning for semiautomatic diameter measurement. However, it should be noted that
most algorithms for diameter assessment operate on a threshold basis. Especially in
larger aneurysms, where an ILT is usually present, the hypodense/hypointense ILT is
excluded from the initial evaluation. In such a case, manual correction of the outer
vessel wall may be required to determine a correct outer-to-outer diameter. However,
in the absence of intraluminal thrombus, a semi-automatic centerline is a powerful
tool for determining aortic diameters in order to plan oversizing of the stent grafts
at the level of the landing zones and for determining the distance between the involved
aortic branches, even in tortuous vessels. [Fig. 4 ] presents an AAA after 3D reconstruction, using IntelliSpace Portal software (Philips
Healthcare) to semiautomatically determine the maximum infrarenal diameter along the
centerline.
Fig. 4 CTA-based semiautomatic assessment of the maximum infrarenal diameter of an AAA.
A Coronal CT angiography of a 69-year-old male patient with AAA and circumferential
intraluminal thrombus. B 3D rendering of the abdominal aorta shows atherosclerotic calcifications (arrowheads)
of the infrarenal aorta. Note that the 3D rendering image visualizes only the contrast-enhanced
perfused lumen of the circumferentially thrombosed AAA. C Curved planar reconstruction of the infrarenal aorta with automatically generated
centerline (green line). D Straight reconstruction with semiautomatic determination of aortic diameters fails
to assess the maximal aortic diameter due to the threshold-based algorithm, which
aligns along the contrast-enhanced perfused lumen of the circumferentially thrombosed
AAA. E Straight reconstruction with manual segmentation of the AAA and thus correct estimation
of the maximal aortic diameter of 5.4 cm.
Magnetic resonance angiography
Magnetic resonance angiography (MRA) is an ionizing radiation-free imaging technique
with excellent soft-tissue contrast. MRA remains the procedure of choice for patients
in whom the use of an iodinated contrast agent or ionizing radiation should be avoided
[44 ]. Move content to the end of the MRI section. According to the other imaging modalities,
the “disadvantage section” was moved to the end. To avoid redudancy, this paragraph
and the last paragraph on page 7 were merged.
MRA – technical considerations
Contrast-enhanced MRA is still considered the first-line technique due to its high
signal-to-noise ratio, which is largely independent from flow-related artifacts [45 ]
[46 ]
[47 ]. ECG gating helps to minimize motion artifacts when performing MRA of the thoracic
aorta but is not required for the abdominal aorta because of its location in the retroperitoneal
space. Non-contrast MRA has become increasingly important in clinical practice in
recent years due to its significant artifact reduction and is therefore established
in routine clinical practice for certain patient groups.
Non-contrast MRA using both two-dimensional (2D) and three-dimensional (3D) techniques
has been available for several decades. 2D techniques often provide a more robust
image quality but preclude multiplanar reconstructions. Flow-independent techniques
have the advantage of being less dependent on rapid inflow of unsaturated spins into
the imaging volume, which may be of relevance in slow flow conditions, e. g., in the
aneurysm sac. Current non-contrast MRA of the aorta is typically based on flow-independent
2D or 3D steady-state free precession pulse (bSSFP) imaging [48 ]
[49 ]. However, a major drawback of bSSFP imaging is off-resonance artifacts caused by
B0 inhomogeneities, which occur particularly at higher field strengths ≥ 3 T [50 ].
Currently used 2D MRA techniques include, for example, cardiac-gated inflow-dependent
quiescent-interval slice-selective (QISS) MRA (Siemens Healthineers) [51 ]
[52 ]. This technique is a notably useful option for patients with additional peripheral
arterial disease (pAVD), which is of particular interest in light of the more than
3-fold increased incidence of pAVD in AAA patients compared with non-AAA patients
[53 ]. QISS MRA has been shown to be less susceptible to severe artifacts compared with
the quite similar time-of-flight (TOF) MRA.
Regarding 3D imaging, a flow-independent non-gated relaxation-enhanced 3 D angiography
without contrast (REACT) technique with magnetization-prepared, non-balanced, dual-echo
acquisition and generalized Dixon (Philips Healthcare) has been introduced [54 ]
[55 ]. Experience in routine clinical use of this sequence at our site indicates robust
applicability for the thoracic [56 ] as well as the abdominal aorta (unpublished data). In [Fig. 5 ], non-contrast 3D REACT MRA is illustrated in an in-house AAA patient.
Fig. 5 MRA-based semiautomatic assessment of the maximum diameter of an AAA. A Sagittal reconstruction of a non-contrast 3D REACT MRA sequence of a 71-year-old
male with AAA. B Maximum intensity projection (MIP) of the 3D REACT MRA with automatic centerline
positioning of the aorta (green line). C MIP visualizing automatic determination of the vessel contours (turquoise circle)
perpendicular to the centerline. D Vessel cross-section for determination of the maximum diameter based on automatic
segmentation of the vessel wall (orange line). The largest aneurysm diameter (MaxD,
41.9 mm) and smallest diameter (MinD, 37.7 mm) at this level are determined automatically.
MRA = magnetic resonance angiography, REACT = relaxation-enhanced 3D angiography without
contrast.
Intraluminal thrombus characterization
ILT is recognized as a biologically active component of AAA acting as a factor in
AAA progression and rupture [22 ]
[23 ]. However, its exact role in AAA pathogenesis remains unclear and continues to be
controversial. Suppression of the blood signal using black blood imaging can be helpful
in assessing thrombus burden and age by providing a sharp delineation of the surrounding
vessel wall [57 ]
[58 ].
ILT is a heterogeneous structure consisting of various components, such as fibrous
matrix, lipid-rich necrotic core, calcifications, or hemorrhage. MRI using T1- and
T2-weighted imaging allows detailed characterization of ILT components [59 ]. Based on ILT compositional variations, it is possible to differentiate between
i) old solid/fibrotic organized thrombi, ii) fresh disintegrated/hemorrhagic thrombi,
and iii) partially organized multilayered thrombi with mixed old as well as fresh
components.
An old fibrotic ILT is characterized by a predominantly iso- to hypointense signal
intensity on T1- and T2-weighted imaging relative to the major psoas muscle, whereas
a fresh disintegrated thrombus is predominantly hyperintense. A partially organized
ILT has a multilayered appearance with both hyper- and hypointense components. Fresh
thrombi were shown to be associated with more rapid aortic growth than old thrombi
[60 ]
[61 ]. MR images of exemplary thrombi (fresh, old, partially organized) from in-house
patients are shown in [Fig. 6 ].
Fig. 6 Thrombus characterization using T2- and T1-weighted MR imaging. Axial T2-weighted
and T1-weighted fat-saturated (FS) black blood MR imaging of three different intraluminal
thrombi (ILT). A Old fibrotic ILTs (arrowheads) are characterized by a predominantly iso- to hypointense
signal intensity on both T1- and T2-weighted imaging relative to the major psoas muscle
(asterisk). B Fresh disintegrated ILTs are predominantly hyperintense on both T1- and T2-weighted
imaging (arrowheads). C Partially organized ILTs have a multilayered appearance with both hyper- and hypointense
components (arrowheads). T2w = T2-weighted MRI, T1w FS black blood = T1-weighted fat-saturated
black blood MRI, SI = signal intensity.
In addition to the examiner-dependent visual ILT classification, there is also a semi-automatic
approach to determine the age of the thrombus. The so-called ILT signal ratio (ILTr ) is calculated by dividing the T1-weighted signal intensity of the ILT by the signal
of the major psoas muscle (signalILT / signalmajor psoas muscle ). AAAs with a high relative signal ratio on T1-weighted MRI (> 1.2) have been shown
to be associated with a higher AAA growth rate [62 ].
However, in addition to being more expensive and less available, disadvantages of
MRI include incompatibility with certain medical devices (e. g., pacemakers) and claustrophobic
patients. Especially in patients who have undergone previous implantation of metallic
implants, image quality may be reduced due to corresponding magnetic susceptibility
artifacts, preventing diagnostic evaluation of the aorta at this level and adjacent
structures. Further drawbacks are a significantly longer acquisition time, precluding
examination of individuals with suspected acute AAA rupture, and the potential requirement
of intravenous contrast. The latter is often omitted in patients with renal failure
due to controversies regarding potential retention in tissues and nephrogenic systemic
fibrosis [63 ]
[64 ].
18 F-FDG PET-CT
18 F-fluoro-deoxy-glucose positron emission tomography (PET/CT) is currently not implemented
in the routine pre-procedural diagnosis and treatment of atherosclerotic AAA. However,
it may provide valuable information regarding the presence of chronic aortic wall
inflammation [65 ]
[66 ]. The potential clinical application of positron emission tomography (PET) in predicting
aneurysm growth may be further optimized by the development and use of more specific
radiotracer agents targeting, for instance, angiogenesis [67 ].
Emerging imaging techniques and technical applications
Aneurysmal dilatation of the aorta favors not only a reduction of flow velocity but
also the occurrence of turbulent blood flow. In recent years, it has been proposed
that local hemodynamics contribute significantly to ILT formation and growth [22 ]
[23 ].
Four-dimensional (4D) flow MRI is a noninvasive functional imaging technique that
makes it possible to obtain time-resolved information on human blood flow dynamics
[68 ]
[69 ]
[70 ]
[71 ]
[72 ]. This technique is based on phase contrast imaging, an MRI technique that allows
simultaneous co-registration of morphological images and velocity data. It has been
shown that flow stasis in AAA can be visualized and quantified using 4D flow MRI [73 ].
Wall shear stress and relative residence time derived from 4D MRI have been suggested
as potential risk factors for AAA development [74 ]. However, prospective longitudinal 4D flow MRI studies are lacking, which would
allow the identification of flow-based biomarkers for better risk stratification of
AAA growth. Exemplary 4D flow MRI-based blood flow profiles of i) a patient without
an AAA, ii) a patient with a small AAA, and iii) with a large AAA are displayed in
[Fig. 7 ].
Fig. 7 4D flow MRI-based characterization of abdominal aortic blood flow profiles. A 4D flow MRI demonstrates a laminar flow profile in the abdominal aorta (AA) of a
27-year-old male control subject without AAA. B 52-year-old male with a small AAA with a diameter of 4.6 cm. Reduced flow velocity
in the aneurysm sac is illustrated by blue streamlines. C 87-year-old woman with a large AAA with a diameter of 6.7 cm. In addition to the
slowed flow in the aneurysm sac, there is significant kinking in the cranial portion
of the abdominal aorta causing turbulent flow (arrow) and flow acceleration as illustrated
by red streamlines. IVC = inferior vena cava, AA = abdominal aorta.
Mathematical models utilizing finite element analysis (FEA) and computational fluid
dynamics (CFD) can estimate aortic stress distribution based on various geometries
(e. g., CTA scans) and material properties. These techniques aim at predicting local
hemodynamics contributing to ILT deposition and AAA rupture [75 ]
[76 ]. For example, an association between the hemodynamics of near-wall particles and
the onset of ILT formation in AAA has been demonstrated, suggesting a critical role
of regional shear stress in the pathogenesis of AAA [77 ].
Speckle tracking ultrasound is another innovative technique that has primarily been
applied to quantify cardiac function, specifically to assess strain. This technique
tracks tissue movement by analyzing speckle patterns on the B-mode ultrasound image.
Interestingly, speckle tracking ultrasound has also been evaluated to assess displacement
of aortic wall motion throughout the cardiac cycle in AAA patients [78 ].
Nevertheless, it should be noted that all techniques discussed in this chapter are
currently considered experimental and are not yet established in the clinical routine.
Further studies are needed to demonstrate the utility of these techniques in the diagnosis
and treatment of AAA.
