Acta chir belg, 2003, 103, 81-86

Three-Dimensional CT Evaluation for Endovascular Abdominal Aortic

Aneurysm Repair. Quantitative Assessment of the Infrarenal Aortic Neck
K. A. Filis, Fr. R. Arko, G. D. Rubin, Chr. K. Zarins
Division of Vascular Surgery, Stanford University Hospital, Stanford, California, USA.

Key words. Three-dimensional CT ; aortic aneurysm ; aortic neck ; endovascular repair.

Abstract. Endovascular grafting of abdominal aortic aneurysms should be offered only to those patients with suitable
anatomy. This is especially true at the level of the proximal aortic neck in order to secure long-term proximal fixation.
Aortoiliac anatomy is easy to understand conceptually, however, it is difficult to define and measure quantitatively. In
this article, we discuss the use of three dimensional computed tomographic angiography to determine aneurysm mor-
phology and select patients for endovascular repair. Specifically, we apply our methods to define and measure angula-
tion of the aorta and iliac arteries. The anatomic definition of the angulation of the proximal aortic neck is emphasized.

Imaging modalities for operative treatment of
abdominal aortic aneurysms
Ultrasonography is the procedure of choice for the ini-
tial diagnosis of abdominal aortic aneurysms (AAA) and
surveillance of small aneurysms. It is non invasive and
inexpensive, uses no radiation, and provides structural
detail of vessel wall and atherosclerotic plaque as well
as accurate AAA size. However, it is highly technician
dependent and image resolution is impaired by body size
and presence of intestinal gas.
The gold standard preoperative test for abdominal
aortic aneurysm assessment was arteriography up until
the early-1980s. Despite issues of cost, discomfort, com-
plications and inaccurate determination of the AAA size,
arteriography had a central diagnostic role on operative
decision making because it could image : a) the supra-
renal extension of the aneurysm, b) concomitant iliac or
peripheral aneurysms, c) renal or mesenteric artery dis-
ease d) accessory renal arteries and e) lower extremity
occlusive disease.
In 1972 computed tomography (CT) was developed
as an interactive technique for combining traditional X-
ray imaging with computed acquisition and processing
of multiple sequential tomographic images. Some years
later (1980-90), CT replaced arteriography for the pre-
operative study of AAAs because it was less invasive
and more accurate in evaluating AAA size and diagnos-
ing inflammatory, leaking, and ruptured AAAs.
Since 1990, helical or spiral CT combined with three
dimensional rendering techniques allows for continuous
circumferential and longitudinal imaging along the
length of the aorta and iliac arteries. Spiral CT with
intravenous contrast enhancement (spiral CTA) provides
the ideal preoperative imaging for selection of patients
for endovascular therapy. Spiral CTA allows for precise
diameter measurements of the proximal aortic neck,
aneurysm size, and access vessels. Furthermore, path
lengths of the proximal aortic neck, renal to aortic bifur-
cation, and distance to the iliac bifurcation are easily
obtained. However, its use can be limited in patients
with renal insufficiency.
Three-dimensional reconstruction of spiral CTA
images
The image data from a spiral CTA acquired in 1-3 mm
thick sections, can be computer processed through soft-
ware packages to develop three dimensional reconstruc-
tion images. These images can be displayed in several
different and complementary formats. They are compre-
hensive and detailed images that allow evaluation of
intraluminal blood flow channels, location of branch
vessels, and vessel wall characteristics. Reconstruction
to any visual angle is possible with satisfactory resolu-
tion of arterial branches 1 to 2 mm in size. Each slice of
the aortoiliac system can be displayed as well as the
axial skeleton of the supra- and infrarenal aorta, the
aneurysm, the iliac arteries and the renal and mesenteric
arteries (2).
To facilitate the measurements of the aortoiliac sys-
tem for endovascular grafting, the Median Center Line
Path (MCLP) is created (3). This is performed through a
computerized process. After extraction of the contrast
enhanced flow channel from spiral CT angiographic
images the software creates orthogonal cross-sections
at 1 mm intervals between two limits of calculation
which are manually selected by the operator. The center
82
Fig. 1
The Median Center Line Path is the sum of the middle points
of all subsequent cross sections of the aortoiliac system.
point for each cross-section is then automatically
marked and all these subsequent center points (at 1 mm
intervals) are visualized at the display and they compose
the Median Center Line Path (Fig. 1). The MCLP pro-
vides in a line the aortoiliac tortuosity in a three dimen-
sional image and can be used as a realistic model of the
aortoiliac angulation. With the addition of the orthogo-
nal cross sections we can measure lengths, diameters,
angles, cross section areas and volumes of any aortoili-
ac segment. This technology permits precise measure-
ments based on a three dimensional reconstructed image
in contrast to the measurements based on a two dimen-
sional image from a conventional CT or angiography
(Fig. 2).
Aortoiliac morphology and patient selection
The endografts that are currently approved for clinical
use in the United States and those in clinical trials differ
in several design features. Thus, each graft has its own
indications for use making some patients eligible for one
graft and not for another. However, certain anatomic fac-
tors are considered essential for all types of endovascu-
lar grafts and should be carefully evaluated and mea-
sured before endovascular grafting, independently of the
graft specific protocol. It is recognized that the methods
used for measuring and describing aortoiliac morpholo-
gy influence treatment results of endovascular therapy
of abdominal aortic aneurysms.
Complete hemodynamic sealing and secure fixation
at the graft attachment sites are the absolute require-
ments for permanent aneurysm exclusion. Therefore, it
is necessary to clearly define the anatomic characteris-
tics that are crucial for aneurysm exclusion. The two
of the most important factors include : 1) anatomic
K. A. Filis et al.
Fig. 2
Three dimensional CT images provide orthogonal cross sec-
tions (sections vertical to flow) which facilitate the accurate
measurement of the diameter.
parameters of the aortic neck and 2) anatomic para-
meters of the iliac arteries. Evaluation of proximal
neck anatomy includes determination of five factors :
a) length, b) diameter, c) configuration or shape,
d) angulation , e) and presence of atheroma or thrombus
within (4). Judgement about the distal fixation zones
(iliac arteries) includes determination of three factors :
a) measurement of length of a suitable segment (not
aneurysmal, without excessive thrombus or atheroma) of
the iliac artery to be used as a distal attachment site,
b) minimal diameter of the iliac arteries to allow access
with the endovascular equipment (sheaths etc), c) and
iliac tortuosity. Presence of internal iliac aneurysms is
also a significant challenge depending on the acceptance
of the risk of occluding one or both internal iliac arter-
ies, or proceeding with an additional surgical treatment
of these aneurysms. These anatomic criteria are used to
determine whether a patient is a candidate for endovas-
cular grafting and to select the type of the graft to use.
Anatomic parameters of the aortic neck
The proximal aortic neck is defined as the distance from
the lower renal artery to the start of the aneurysm
(Fig. 3). In our series of patients with the AneuRx stent
graft (Medtronic, Santa Rosa, CA) a neck length of at
least 10 mm was required. The aortic neck was defined
on the three-dimensional image as the segment between
two points of the MCLP of the aorta. These two points
(orifice of the lower renal artery and start of the
aneurysm) are defined by scrolling on the computed
CTA images and clicking on the anatomic points. The
point of the orifice of the lower renal is visually detect-
ed and manually selected by the vascular surgeon who
does the evaluation. Although the start of the aneurysm
Aortoiliac Anatomy for Endovascular Grafting
Fig. 3
The proximal to the aneurysm aortic segment is an aneurysm.
This patient has no aortic neck (he has a juxtarenal aortic
aneurysm).
is generally considered to be defined easily, the precise
anatomic definition of this point is dependent consider-
ably on the operator. The point at which the aorta starts
to be an aneurysm is also manually selected by the oper-
ator (Fig. 4).
Neck Length : The measurement of the length of the
proximal fixation zone is calculated between the point of
the MCLP at the orifice of the lower renal artery and the
point of the MCLP at the start of the aneurysm.
Neck diameter : The diameter of the neck is measured
on the orthogonal cross sections of the neck, from the
outer wall to the outer wall. Orthogonal cross sections of
the neck can be obtained at 1 mm intervals along the
length of the neck. The shape of each cross section is not
an ideal circle, and each cross section is not the same
along the aortic neck length. Therefore, we can measure,
on the same cross section, different aortic neck diame-
ters (diameters at different positions), as well as the
diameter at the same position of different cross sections
of the neck. As a result, we can measure the minimum,
the maximum, and the mean value of all diameters of the
neck. The diameter that is used to decide if the neck is
suitable for endovascular repair, is the maximum diame-
ter of the orthogonal cross section just below the lower
renal artery (5, 6). Although today there are no practical
clinical applications for the measurement of all diame-
ters, the distribution of the diameters of the neck along
its length allows the creation of a numerical estimation
83
Fig. 4
Different (minimum and maximum) diameters along the aortic
neck length.
of the neck’s shape. The mean value of the difference of
the diameters of the subsequent cross sections can define
the shape of the neck. For an aortic neck which is an
ideal cylinder, the mean of the differences between the
diameters would be nil.
Neck shape : The aortic neck is ideally a cylinder.
However, this is a simplification of the true morphology
of the neck. In fact, any aortic neck which does not have
a morphology close to a true cylinder is not a suitable
proximal fixation zone for endovascular repair (Fig. 5).
Therefore, aortic necks suitable for proximal fixation
zones for endovascular repair can be assumed to have a
straight (or near straight) MCLP and the diameter
change, along its length, should be small (less than
5 mm ) (7) (Fig. 4).
The shape or configuration of the neck is usually
determined by visual examination of coronal views of
the aorta at different angle projections. Necks that have
a reverse funnel shape are poor candidates (Fig. 5). The
distribution of the diameters of the neck at 1 mm inter-
vals gives numerical data for the shape determination of
the aortic neck. The mean of the differences of all the
diameters can be a parameter that defines the irregulari-
ty of the neck shape. Numerical limits for this number
can then -through future clinical research- be created
after comparing different numerical values of various
necks to the clinical outcome after endovascular repair.
Neck atheroma and thrombus : Aortic neck atheroma
and thrombus formation are factors which are visually
detected and judgement on suitability is dependent on
experience and practice. A heavily calcified aorta must
be approached with caution as it may preclude graft
deployment. A considerable amount of thrombus in the
neck poses a question about permanent fixation and
84
Fig. 5
Different types of proximal aortic necks which are not ideal for use as a proximal fixation zone for EVG
Fig. 6
Definition of the proximal and the distal aortic neck angles
sealing of the endograft. A semi-quantitative method has
been proposed by BALM et al. (7) for the presence of aor-
tic neck calcification. According to this method, the
number of quadrants (anterior/left/posterior/right) per
slice that contain calcifications larger than 5 mm is
counted and divided by the total number of quadrants.
However, there is lack of a generally accepted method to
calculate calcification and thrombus, our practice is to
determine the suitability of the neck in relation to calci-
fication and thrombus by visual inspection of coronal
and sagittal views.
K. A. Filis et al.
Neck angulation : We have considered that the aortic
neck suitable for endovascular repair has a morphology
close to a cylinder. This allows us to define the axis of
the aortic neck as the straight line between the MCLP
of the aorta at the level of the orifice of the lower renal
artery and the MCLP of the aorta at the start of the
aneurysm. The aortic neck is the anatomic segment of
the aorta which lies between two segments. The first
segment is the aorta above the renal arteries, the second
is the cylindrical neck, and the third segment is the
aneurysm. Between these three segments two angles are
defined. The proximal aortic neck angle is the angle
between the axis of the neck and the axis of the
suprarenal aorta. The distal aortic neck angle is the
angle between the axis of the neck and the axis of the
lumen of the aneurysm (Fig. 6). The axis of the
suprarenal aorta is defined as the straight line between
the MCLP of the aorta at the level of the orifice of the
superior mesenteric artery and the MCLP of the aorta at
the orifice of the lower renal artery. The axis of the
lumen of the aneurysm is the straight line between the
MCLP at the start of the aneurysm and the MCLP at the
end of the aneurysm. When the lumen of the aneurysm
is not straight, the first 2.0 cm at its proximal end is used
to determine the axis of the aneurysm (Fig. 7).
It is well known that in the majority of the published
reports, aortic neck angulation is defined as the distal
aortic neck angle. The definition of neck angulation by
describing and measuring two angles, the proximal and
the distal, is for us more realistic than defining neck
Aortoiliac Anatomy for Endovascular Grafting
Fig. 7
An example of a nil proximal aortic neck angle. Because the
MCLP of the aneurysm is not straight, the axis of the aneurysm
is defined by the MCLP at its start and the MCLP at 2 cm dis-
tally.
angulation only by measuring the distal aortic neck
angle. The fact that the proximal aortic neck angle is
usually nil explains in part why it is not usually
considered as a parameter of neck angulation (Fig. 7).
However, this is not always true. Using the three dimen-
sional CTA images and by the definitions mentioned
above we measured the proximal neck angle in a series
of our patients (n = 334) and we found this angle to be
> 5 degrees in half of the cases evaluated.
In order to confirm the clinical value of the proximal
aortic neck angle, we conducted a retrospective analysis
at our institution. We compared two groups of patients
who had undergone endovascular AAA repair with the
AneuRx stent graft and had long term follow up for
more than three years. Group I consisted of 15 patients
who required secondary procedures to secure stent graft
fixation or treat complications (endoleaks, migrations)
and group II consisted of 18 patients who did not under-
go any secondary procedure. The proximal aortic neck
angle was greater in group I (22.6 ± 16.2 degrees) com-
pared to group II (11.9 ± 6.9 degrees) (p < 0.01).
However, the distal aortic neck angle was not different
between groups (group I : 17.6 ± 12.4, group II : 18.8 ±
9.4). The measurement of the distal aortic neck angle
was a parameter which was included in the criteria for
suitability for endovascular AAA repair, though the
measurement of the proximal aortic neck angle was not.
We concluded that the proximal aortic neck angle may
be related to the need for secondary procedures after
endovascular repair, though the distal aortic neck angle
was not. However, the latter may have been due to the
fact that an inclusion criterion in relation to the distal
aortic neck angle (< 60 degrees) had been applied
(Fig. 8).
85
Fig. 8
The proximal and distal neck angles in post-implantation three
dimensional CT images.
The literature contains a number of recommendations
regarding aortic neck angulation. The Eurostar trial (8),
after an analysis of 2146 patients, showed that neck
angulation was related to migration or type I endoleaks.
Similar results were reported by ALBERTINI et al. (9) and
STRENBERG et al. (10). However, these studies refer pri-
marily to the distal aortic neck angle. PETRIK et al. (11)
reported no difference in neck angulation (distal aortic
neck angle), between patients with and without type I
endoleaks. In this study, no patient with a distal aortic
neck greater than 60 degrees was offered an endovascu-
lar repair as was the case in this series we report.
Anatomic parameters of the iliac arteries
The value of the definition of the anatomy of iliac ar-
teries in endovascular AAA repair is related to :1) the
presence of a segment to serve as the distal fixation zone
and 2) the diameter and tortuosity of these vessels, to
allow access with wires, sheaths and stent grafts.
Iliac artrery diameter : The minimal diameter of the
iliac arteries (common and external iliac) is measured to
make sure that access can be obtained with the sheath
and/or device. The measurements are based on the cross
sectional areas of the flow lumen of the arteries and
therefore the diameter measured is the diameter of the
lumen of the artery. The presence or absence of iliac
artery occlusive disease, should be noted (12).
Iliac artery calcification and thrombus : These para-
meters cannot be estimated by a validated quantitative
method. Experience influences visual judgement.
Iliac artery tortuosity : Angulation and tortuosity of
the iliac arteries makes deployment of the endograft
86
more difficult. Increased angulation and toruosity can
cause embolization and dissection. With methods cur-
rently available, it is difficult to describe tortuosity of the
iliac arteries since they can be variably tortuous through-
out their length. Angulation may change along the
length of the iliac arteries and may be particularly dif-
ferent in the common and the external iliac arteries. We
believe as others (11) that the proximal iliac angle is an
important parameter of aortoiliac tortuosity that may
play a role in postoperative clinical outcome. The prox-
imal iliac angles are defined as the angles between the
axis of each iliac artery (c : right, d : left) and the axis of
the lumen of the aneurysm. The axis of each iliac artery
is defined as the straight line between the MLCP at the
level of aortic bifurcation and the MLCP of each iliac
artery at 2.0 cm distally. The axis of the aneurysm is
described earlier, but when its lumen is not straight, the
last 2 cm at its distal end is used to determine its axis.
Conclusion
Endovascular surgery has undoubtedly had a dramatic
effect in the treatment of abdominal aortic aneurysm
repair. Although short and medium term results are
encouraging, there are still reports of isolated adverse
events of rupture, migrations and late onset endoleaks,
especially in patients with increased angulation of the
proximal aortic neck. Although, the value of the pre-
operative anatomic evaluation has been strongly empha-
sized, the definition and quantification of the anatomic
parameters still needs further definition. Three dimen-
sional reconstructed CTA images provide a suitable
imaging method for clinical research in this erea. We
have defined aortic neck angulation as two separate
angles. Our data support that the proximal aortic neck
angle is a parameter of aortic neck angulation that has to
be measured. A prospective study is required to answer
the question of what is the limit of the size of the proxi-
mal aortic neck angle that prevents the need for secon-
dary procedures after endovascular AAA repair. Future
research should focus on the precise and standard defini-
tion of aortic neck shape, aortic calcification and throm-
bus and a quantitative determination of the overall aorto-
iliac tortuosity.
K. A. Filis et al.
References
1. PARODI J. C., PALMAZ J. C., BARONE H. D. Transfemoral intralumi-
nal graft implantation for abdominal aortic aneurysms. Ann Vasc
Surg, 1991, 5 : 491-9.
2. RUBIN G. D., DAKE M. D., NAPEL S. A., MCDONNELL C. H.,
JEFFREY R. B. Jr. Three-dimensional spiral CT angiography of the
abdomen : initial clinical experience. Radiology, 1993, 186 : 147-
52.
3 RUBIN G. D., PAZK D. S., JOHNSON P. C., NAPEL S. A. Quantification
of the aorta and its branches using helical CT. Radiology, 1998,
206 : 823-829.
4. BEEBE H. G. Selection criteria for endovascular treatment of
abdominal aortic aneurysm. In : ERNST C., STANLEY J. (eds.)
“Current Therapy for Vascular Surgery”, Mosby Inc., 2001.
5. BAYLE R., BRANCHEREAU A., ROSSET E., BEAURAIN P., FERDANI M.,
JAUSSERAN J. morphologic assessment of abdominal aortic
aneurysms by spiral computed tomographic scanning. J Vasc
Suirg, 1997, 26 : 238-46.
6. RUBIN G. D., SHIAU M. C., LEUNG A. N. et al. Aorta and iliac arte-
ries : Single versus multiple detector – row helical CT angio-
graphy. Radiology, 2000, 215 : 670-6.
7. BALM R., STOKKING R., KAATE R., BLANKESTEIJN J. D., EIKEL-
BOOM B. C., VAN LEEUWEN M. S. Computed tomographic angio-
graphic imaging of abdominal aortic aneurysms : Implications for
transfemoral endovascular aneurysm management. J Vasc Surg,
1997, 26 : 231-7.
8. MOHAM I. V., LAHEIJ R. J., HARRIS P. L., EUROSTAR COLLABORATORS.
Risk factors for endoleak and the evidence for stent-graft over-
sizing in patients undergoing endovascular aneurysm repair. Eur J
Vacs Endovasc Surg, 2001, 21 : 344-9.
9. ALBERTINI J. N., SALLIAFAS S., TRAVIS S., YUSUF S. W., MACIERE-
WICS J. A., WHITAKER S. C. et al. Anatomical risk factors for proxi-
mal perigraft endoleak and graft migration following endovascu-
lar repair of abdominal aortic aneurysms. Eur J Vasc Endovasc
Surg, 2000, 19 : 308-12.
10. STERNBERG W. C., CARTER G., YORK J. W., YOSELEVITZ M.,
MONEY S. R. Aortic neck angulation predicts adverse outcome
with endovascular abdominal aortic aneurysm repair. J Vasc Surg,
2002, 35 : 482-6.
11. PETRIK P., MOORE W. S. Endoleaks following endovascular repair
of abdominal aortic aneurysms : the predictive value of preopera-
tive anatomic factors – a review of 100 cases. J Vasc Surg, 2001,
33 : 739-44.
12. ANN S. S., RUTHERFORD R. B., JOHNSTON K. W., MAY J., VEITH F. J.,
BAKER J. D. et al. Reporting standards for infrarenal endovascular
abdominal aortic aneurysm repair. J Vasc Surg, 1997, 25 : 405-10.
Dr. Chr. K. Zarins, M.D.
Division of Vascular Surgery
Stanford University Medical Center
300 Pasteur Drive
H-3642, Stanford, California, 94305-5642
Tel. : 650-725-7830
Fax : 650-498-6044
E-mail : zarins@stanford.edu