Mechanical wall stress in abdominal aortic

aneurysm: Influence of diameter

and asymmetry
David A. Vorp, PhD, M. L. Raghavan, BS, and Marshall W. Webster, MD,
Pittsburgh, Pa.
Purpose: Risk for rupture of an abdominal aortic aneurysm is widely believed to be related to its maximum diameter. From a biomechanical standpoint, however, risk is probably more precisely related to mechanical wall stress. Many abdominal aortic aneurysms
are asymmetric (for example because of anterior bulging with posterior expansion limited by the vertebral column). The purpose of this work was to investigate the effect of
maximum diameter and asymmetric bulge on wall stress.
Methods: Three-dimensional computer models of abdominal aortic aneurysms were generated. In one protocol, maximum diameter was held constant while bulge shape factor
was varied. The shape factor took into account the asymmetric shape of the bulge. In a
second protocol, the shape of the aneurysmal wall was held constant while maximum
diameter was varied. Wall stress was computed in each instance with a commercial software package and assumption of physiologic intraluminal pressure.
Results: Both maximum diameter and the shape factor were found to have substantial
influence on the distribution of wall stress within the aneurysm. In some instances the
maximum stress occurred at the midsection, and in others it occurred elsewhere. The
magnitude of peak stress acting on the aneurysm increased nonlinearly with increasing
maximum diameter or increasing asymmetry.
Conclusions: Our computer models showed that the stress within the wall of an abdominal aortic aneurysm and possibly the potential for rupture are as dependent on
aneurysm shape as they are on maximum diameter. This information may be important
in determining severity of individual abdominal aortic aneurysms and in improving
understanding of the natural history of the disease. (J Vasc Surg 1998;27:632-9.)

Abdominal aortic aneurysm (AAA) disease

occurs among approximately 2% of the elderly population, and the incidence apparently is increasing.1
Rupture of an AAA is currently ranked as the 13th
most common cause of death in the United States.2
Surgical repair of AAA is performed to prevent death
from rupture and is associated with acceptable but
not insignificant morbidity and mortality rates.3-5 It
is important to determine when during the natural
history of an aneurysm, risk for rupture justifies the
operation and its potential attendant complications
From the Department of Surgery and Department of Mechanical
Engineering (Dr. Vorp) and Program in Bioengineering (Dr.
Vorp and Mr. Raghavan), University of Pittsburgh.
Supported by a grant from the Whitaker Foundation (D.A.V.).
Reprint requests: David A. Vorp, PhD, University of Pittsburgh,
Department of Surgery, Section of Vascular Surgery, A-1011
P.U.H., Pittsburgh, PA 15213.
Copyright 1998 by The Society for Vascular Surgery and
International Society for Cardiovascular Surgery, North
American Chapter.
0741-5214/98/$5.00 + 0 24/1/88728

632

and expense.6,7 At present the decision for elective

repair of an AAA is usually based on the maximum
diameter of the aneurysm. Several other anatomic
variables have been proposed for use as predictors of
aneurysm severity.1,6,8 These criteria are quite crude
and do not take into account certain important characteristics of individual aneurysms. For example,
AAAs with the same maximum diameter may have
differences in shape, wall thickness, or mechanical
properties that affect their propensity for growth
and rupture. Thus an operation based on a 5 cm
critical diameter or other similar criterion may be
unjustified (low risk for rupture) or too late (rupture
at less than 5 cm) for a particular patient.9-11
From a biomechanical perspective, the proper
definition of the critical state of an AAA is that at
which the mechanical stress within the aneurysmal
wall exceeds the tensile strength of the tissue. We
have reported on the tensile strength of aneurysmal
tissue,12,13 and there have been efforts to estimate
wall stress within AAAs.14-17 The purpose of this

JOURNAL OF VASCULAR SURGERY

Volume 27, Number 4

Vorp, Raghavan, and Webster

633

Fig. 2. Three-dimensional models of AAA represents the

two extreme cases studied for each protocol. Protocol 1,
variation of asymmetry parameter from 1.0 (axisymmetric) to 0.3 while maximum diameter (Dmax) was held constant at 6 cm. Protocol 2, variation of maximum diameter
from 4 cm to 8 cm while was held constant at 0.4. Three
intermediate cases were studied between the two extremes
for each protocol.

puter simulations showed that asymmetric shape is

an important factor affecting mechanical wall stress
within an AAA and that consideration of maximum
diameter alone may be insufficient as an estimation
of the severity of an individual AAA.
Fig. 1. Representative, three-dimensional, virtual AAA
model (top) and midsectional cross section (bottom). rp
and ra, maximum posterior and anterior wall dimensions,
respectively, measured from the dashed longitudinal line
defined by the centers of the two undilated ends. rp + ra =
maximum diameter. Inflection points (IP) are defined as
points on the AAA surface at which the local AAA wall
shape changes from concave outward to concave inward
and are indicated here only for the anterior and posterior
surfaces. The angular circumferential coordinate is
defined as shown.

study was to investigate the separate effects of AAA

bulge shape and diameter on wall stress distribution
in AAAs. Most AAAs are not axisymmetric.18,19 For
example, limitation of posterior expansion caused by
the vertebral column might result in preferential
anterior expansion of the aneurysmal wall and an
asymmetric configuration. An asymmetric shape
might greatly influence stresses on the wall of an
AAA and might be as important a clinical consideration as aneurysmal diameter.
We performed computational analyses using
three-dimensional computer models (or virtual
AAAs) wherein maximum diameter and degree of
asymmetry were individually controlled. Our com-

METHODS
Geometric modeling. Computer models of
hypothetical AAAs were generated with commercial
software (Pro-Engineer v. 16.0; Parametric Technology Waltham, Mass.) such that overall length was
12 cm and cross section at any axial position was circular, as has been shown to be typical.8,20 A representative model is shown in Fig. 1. Because wall
thickness was not studied in this investigation, it was
assumed to be uniform throughout at 1.5 mm. The
posterior and anterior wall profiles of the AAA models were generated from normal Gaussian distribution (bell curves). Important locations on the wall of
the virtual AAA, particularly the inflection points of
the surface, are shown in Fig. 1. These are locations
anywhere on the wall where the surface of the
aneurysm changes from concave outward to concave
inward.
For a separate investigation of the effect of maximum diameter and asymmetry on wall stress distribution in an AAA, the virtual AAAs were generated
according to two protocols (Fig. 2). In one protocol
maximum diameter was kept constant at 6 cm while
an asymmetry parameter was varied, where =
rp/ra. As shown in Fig. 1, rp is maximum posterior
wall dimension, and ra is maximum anterior wall

634 Vorp, Raghavan, and Webster

dimension. It can be seen that rp + ra = maximum

diameter. The anatomy of an AAA is usually such
that rp < ra ( < 1) because of the proximity of the
posterior wall to the vertebral column. Thus computer models were generated with that varied from
0.3 to 1.0. A value of 1.0 for represents an axisymmetric AAA with equal anterior and posterior
bulging; a value of 0.3 represents a highly asymmetric AAA with a highly preferential anterior bulge. In
the other protocol, models were generated for a
constant asymmetry ( = 0.4), and maximum diameter varied from 4 cm to 8 cm. The diameter of the
undilated aorta ranged from 2.1 cm to 2.4 cm in the
models.
Ten total virtual AAAs were constructed, five for
each protocol. The two extreme cases for each protocol are shown in Fig. 2. All AAA models have at
least two planes of symmetry, namely, the transverse
plane and the median plane. Because of the transverse plane of symmetry, there are two inflection
points in the AAA profile shape, one above the midsection and one below (Fig. 1).
Finite element model. The complex shape of
the aneurysmal wall precludes a solution for the wall
stress by means of straightforward analytic techniques. Instead we used a technique14-16,21 known as
finite element analysis for computation of the wall
stresses in each virtual AAA. In this technique, a body
of complex shape is divided into smaller, simpler
shaped elements. The stresses over the individual elements are computed, and the solution is patched
together to yield the stress distribution for the entire
complex body. Specific technical details regarding
finite element analysis may be found elsewhere.21
The three-dimensional AAA models were imported into the finite element software package ANSYS
(version 5.3; Ansys, Houston, Pa.) and discretized
into small elements with a linearly elastic, quadrilateral shell element. The number of elements ranged
from 3200 to 3800 depending on the shape and size
of the virtual AAA. The elastic modulus (500
Newtons/cm2) used in the analysis was obtained by
means of previous tensile tests of AAA tissue performed by us.12,13 It has been shown that vascular
tissue is nearly incompressible,22 and we used a
Poisson ratio of 0.49.
Static analysis was performed. A peak systolic arterial pressure load (120 mm Hg [1.6 Newtons/cm2])
was applied uniformly on the internal surface of the
AAA models. Shear stresses caused by flowing blood
were not considered because they are shown to be
small in magnitude compared with stresses caused by
distention of the wall.16,23 The outer surface of the

JOURNAL OF VASCULAR SURGERY

April 1998

AAA model was taken as load free, and the two ends
were held fixed in the axial direction. This longitudinal constraint took into account tethering of the aorta
by the surrounding connective tissue and the vertebral and other collateral arteries. To represent the
complex stress distribution in the wall of each virtual
AAA, von Mises stress distribution16 was computed
and inspected for each simulation. The von Mises
stress is derived from the distortion energy used in
studies of material failure. It is a function of the three
principle stresses in the body of an AAA.24
RESULTS
The three-dimensional distribution of mechanical wall stress for each of the ten virtual AAA models is shown in two views in Fig. 3. The results for
protocol 1 (varying asymmetry) are shown in Fig. 3,
A, and those for protocol 2 (varying maximum
diameter) are shown in Fig. 3, B. The effects of the
asymmetry parameter and maximum diameter on
the circumferential variation of wall stress around
the AAA midsection (along the dashed curve in Fig.
1) is shown in Fig. 4. The maximum stress at the
AAA midsection occurred on the posterior surface in
all models studied.
The effects of asymmetry and maximum diameter on the longitudinal variation of wall stress along
the anterior surface of the virtual AAA are shown in
Fig. 5. Both parameters were found to have similar
effects. Increasing the diameter or asymmetry of an
AAA causes an increase in wall stress at the inflection points of the profile shape (Fig. 1) while causing a decrease in stress at the midsection. The effects
of asymmetry and maximum diameter on longitudinal variation of wall stress along the posterior surface of a virtual AAA are shown in Fig. 6. The
effects are similar to those found for the anterior
surface but with the following differences. As the
aneurysm becomes more asymmetric, the greatest
stress on the posterior surface relocates from the
inflection points to the midsection (Fig. 6, A).
Conversely, as the aneurysm enlarges, the maximum
stress relocates from the midsection to the inflection
points (Fig. 6, B).
When an AAA is small, the maximum wall stress
occurs on the posterior wall at the midsection (compare Figs. 5, B and 6, B). As the AAA enlarges, the
stress on the posterior wall at the midsection remains
elevated, but the maximum stress occurs at the
inflection points on the anterior surface. The peak
stress within the virtual AAA was found to increase
nonlinearly with increasing diameter as well as with
increasing asymmetry (Fig. 7).

JOURNAL OF VASCULAR SURGERY

Volume 27, Number 4

Vorp, Raghavan, and Webster

Fig. 3. Distribution of mechanical wall stress in all ten virtual AAAs. A, Models for protocol
1 (varying asymmetry). B, Models for protocol 2 (varying maximum diameter). Each model is
shown in two views. The left columns of both A and B provide a view of the anterior, left lateral surface of the virtual AAA. The right column provides a view of the posterior, right lateral surface. The sketch at the top of each column shows the vertebral column and provides
anatomic reference. The magnitude of the mechanical wall stress for both protocols is given on
the individual color scales in Newtons/cm2 (note: 1 Newton/cm2 = 105 dynes/cm2).

635

636 Vorp, Raghavan, and Webster

Fig. 4. Effect of asymmetry parameter (A) and maximum diameter (B) on circumferential distribution of wall
stresses at the midsection of the AAA (along dashed curve
in Fig. 1). The coordinate is as defined in Fig. 1. A position of = 0 degrees corresponds to the posterior surface,
and 180 degrees corresponds to the anterior surface. For
this and subsequent Figures, a decrease in corresponds
to an increase in asymmetry.

DISCUSSION
Risk for rupture of AAA is widely believed to be
associated with maximum diameter. It has been
observed clinically, however, and autopsy studies
show that some large aneurysms do not rupture
while some small aneurysms do.9 There is a need for
a better definition of the severity of AAA. Aneurysm
rupture is caused by a gross mechanical failure of the
aortic wall and occurs when the acting mechanical
wall stress exceeds the strength of the tissue. We
believe that knowledge of the wall stresses in AAA
would provide clinicians with a more accurate estimate of the likelihood of rupture of an individual
aneurysm. Our study showed that the stress within
the wall of an AAA and possibly its propensity for
rupture depend on the shape and the diameter of the
aneurysm.
To our knowledge, this was the first study to
investigate the effects of asymmetry on three-dimensional stress distribution in the wall of AAAs. Previous

JOURNAL OF VASCULAR SURGERY

April 1998

Fig. 5. Effect of asymmetry parameter (A) and maximum diameter (B) on longitudinal distribution of wall
stresses along the anterior surface of an AAA. Axial positions of z = 0 cm and z = 12 cm correspond to the superior and inferior undilated ends of the AAA; z = 6 cm corresponds to the midsection.

consideration of AAA wall stresses have used the law

of Laplace,19 or investigators assumed axisymmetric
geometry.14-17 The law of Laplace is often incorrectly
used because in its most common forms, it is applicable only to cylinders or spheres. It is thus insufficient
to provide estimates of the complex stress distributions in AAA. Using the law of Laplace, one would
predict greatest wall stresses at the midsection of each
of the virtual AAAs studied because diameter was
maximum there. Our computer models showed that
this is usually not the case (Figs. 3, 5, and 6). We
showed here that stress distribution in asymmetric
AAA is markedly different from that in axisymmetric
AAA. Previous models with axisymmetric AAA shapes
(for which = 1) would fail to describe important
variations of stresses along the aneurysmal wall caused
by asymmetry (from preferential bulging, for example). Nonetheless, if their limitations are kept in mind,
previous models have provided important insight into
the biomechanical features of AAA.

JOURNAL OF VASCULAR SURGERY

Volume 27, Number 4

Vorp, Raghavan, and Webster

637

Fig. 7. Effect of asymmetry parameter (bottom axis)

and maximum diameter (top axis) on magnitude of peak
stress within a virtual AAA. Both increasing diameter and
increasing asymmetry (decreasing ) cause a nonlinear
increase in peak stress.

Fig. 6. Effect of asymmetry parameter (A) and maximum diameter (B) on longitudinal distribution of wall
stresses along the posterior surface of an AAA. The longitudinal axis is the same as in Fig. 5.

Stringfellow et al.14 used finite element analysis

to investigate the differences in stress distribution
for spherical versus cylindrical AAAs. Although they
performed their analysis on axisymmetric shapes,
they appear to have been the first to consider shape
effects on AAA stress distribution. Their results
showed that circumferential stress in the walls of
cylindrical aneurysms is greater than that in spherical
aneurysms and that longitudinal stress is unaffected.
One possible limitation of the investigation by
Stringfellow et al., however, was use of a sharp junction between the AAA and undilated aorta. Our
geometric modeling techniques allowed a smoother
transition, avoiding unrealistic stress concentrations.
Inzoli et al.16 also used an axisymmetric AAA model
to demonstrate the effects of maximum diameter
and presence of intraluminal thrombus and atherosclerotic plaque within the wall. In agreement with
our results (Fig. 7), they found that peak wall stress
increased with increasing maximum diameter. They
also showed that the presence of a calcified atherosclerotic plaque caused elevated stress concentrations and an increase in peak stress, whereas intralu-

minal thrombus caused a decrease in peak stress by

up to 30%. We suggested the mechanically protective consequence of intraluminal thrombus in previous experimental studies.25 Mower et al.15 demonstrated with an axisymmetric model that AAA wall
stress increases in proportion to maximum diameter
and in inverse proportion to wall thickness. As we
did in this study, Mower et al. determined that maximum stress in axisymmetric AAA models occurred
at the inflection points of the profile shape. Elger et
al.17 showed that the shape of an axisymmetric AAA
influences stress distribution. They suggested that
maximum wall stress is a function of the curvature of
the wall profile. We extended these previous studies
by considering aneurysms that were not axisymmetric but had more realistic asymmetric shapes produced by preferential or nonuniform bulging.
Our results indicated that for small AAAs (5
cm) maximum stress occurs on the posterior wall
but for larger AAAs peak stress occurs on the anterior surface. Mechanical failure of aortic tissue leads to
AAA rupture and occurs when local wall stress
exceeds local wall strength. We have reported on
wall strength of AAA tissue12,13 and have found that
the strength of the tissue near the neck or undilated
ends of an AAA are greater than that in the midsection, where diameter is maximum (unpublished
results). Although the stresses are maximum on the
anterior surface for larger AAAs, the actual propensity for rupture for that or any other surface depends

638 Vorp, Raghavan, and Webster

on the comparative local value of wall strength.

Because wall strength may be lower at the midsection and stresses are greatest on the posterior surface
(Fig. 4), it is possible that rupture would occur on
this surface before it would occur on the anterior
surface, where stresses are actually higher. Autopsy
studies have shown that there is no clear predilection
for site of rupture of AAA, but it seems to occur
most frequently along the posterior surface.9 To our
knowledge, no study has been performed to correlate diameter with rupture site. Comparison of our
results with those of such a study would allow more
rigorous interpretation of the importance of local
wall stress in AAA rupture.
The relative importance of maximum diameter
and can be demonstrated with inspection of Figs.
4 through 7. For example, Fig. 4, A shows that the
value of stress on the midsection of the posterior
wall is increased by more than 85% when the AAA
changes shape from axisymmetric ( = 1, stress 15
Newtons/cm2) to highly asymmetric ( = 0.3, stress
28 Newtons/cm2) with maximum diameter constant. Fig. 4, B shows that stress at the same site is
increased by only 33% because of an increase in maximum diameter from 4 cm (stress 18 Newtons/cm2)
to 8 cm (stress 24 Newtons/cm2) with constant.
Fig. 7 demonstrates that the value of peak wall stress
is increased by 137% with an increase in maximum
diameter from 4 cm (stress 19 Newtons/cm2) to 8
cm (stress 45 Newtons/cm2) but is increased by
45% with an increase in asymmetry from 1.0 (stress
22 Newtons/cm2) to 0.3 (stress 32 Newtons/cm2).
According to the critical diameter criterion often
used to guide decisions for elective AAA repair, all
AAAs of the same diameter have the same likelihood
of rupture. Our results show that this is probably not
the case. For example, Fig. 6, A depicts results for
five different virtual AAAs with a diameter of 6 cm.
The critical diameter criterion would suggest that
each of these AAAs has an equal predilection for
rupture. However, the stress on the midsection of
the posterior wall of a highly asymmetric AAA ( =
0.3) is twice that of an axisymmetric AAA ( = 1.0).
If the strength of the aortic wall were the same for
both, the asymmetric aneurysm would be more likely to rupture.
The finite element method provides a means to
inspect the stress distribution in individual AAA, and
this can lend itself as a powerful clinical and research
tool. With a stress map, it was possible to study the
individual effects of aneurysmal diameter and bulge
shape on stress distribution in virtual AAAs. Similar
techniques could be used as a reliable, noninvasive

JOURNAL OF VASCULAR SURGERY

April 1998

method to evaluate the severity of actual individual

AAAs. This estimate of severity would be based on
biomechanical wall stresses calculated with threedimensional reconstructions from computed tomographic scans or magnetic resonance images26 and
could be an important clinical tool to guide surgeons
in decisions about elective repair of AAA. Studying
stress distribution in real AAAs would be a useful
clinical research tool by providing means to study the
biologic consequences of mechanical stresses. For
example, as pointed out by Mower et al.,15 vasa vasorum and other small blood vessels in regions of
increased stress may tend to be compressed and not
provide nutrition to the vessel wall. Thus sites of
AAA with increased wall stress may undergo weakening with impaired repair or remodeling mechanisms.
Increased local stresses or stress gradients likewise
may regulate gene expression by the aortic wall,
which would have an influence on natural history.
Limitations. In this analysis it was assumed that
the AAA wall is homogeneous, isotropic, and linearly elastic and that it undergoes small strains, which is
not the case.12,27,28 However, the analysis provided
a first approximation of the effects of AAA asymmetry and diameter, which was the goal of this study.
We also assumed that the mechanical properties and
thickness of the AAA wall are uniform over the surface. This is likely not realistic because of localized
calcifications, for example, and variations of these
parameters should be accounted for when studying
stresses in real AAA. Though the same undilated
aortic diameter was not used for all models, the
slight variation was not enough to cause significant
differences in stress distribution.
We provide what we believe is the first demonstration that asymmetry of the AAA bulge is important in considering mechanical wall stresses. However,
all the virtual AAAs studied had at least two planes of
symmetry, the median plane and the transverse plane.
Although we used more realistic shapes than in previous studies, actual AAAs generally have even more
complex shapes18,19 and have no planes of symmetry.
Future analyses should evaluate the effect of other
asymmetries on mechanical wall stresses in AAA.
Conclusion. The asymmetry of an AAA, in
addition to aneurysm diameter, is an important
determinant of mechanical wall stress. Despite the
critical diameter criterion often used to assess severity of an AAA, aneurysms with the same diameter
may not necessarily have the same propensity for
rupture. This information may be important in
understanding the natural history of AAA and in the
clinical management of this disease.

JOURNAL OF VASCULAR SURGERY

Volume 27, Number 4

Vorp, Raghavan, and Webster

REFERENCES
1. Ouriel K, Green RM, Donayre C, Shortell CK, Elliott J,
DeWeese JA. An evaluation of new methods of expressing
aortic aneurysm size: relationship to rupture. J Vasc Surg
1992;15:12-20.
2. Patel MI, Hardman DTA, Fisher CM, Appleberg M. Current
views on the pathogenesis of abdominal aortic aneurysms. J
Am Coll Surg 1985;181:371-82.
3. McCombs RP, Roberts B. Acute renal failure after resection
of abdominal aortic aneurysm. Surg Gynecol Obstet
1979;148:175-9.
4. Johnston KW. Multicenter prospective study of nonruptured
abdominal aortic aneurysm; II: Variables predicting morbidity and mortality. J Vasc Surg, 1989;9:437-47.
5. Hollier LH, Taylor LM, Ochsner J. Recommended indications for operative treatment of abdominal aortic aneurysms.
J Vasc Surg 1992;6:1046-56.
6. Limet R, Sakalihassan N, Albert A. Determination of the
expansion rate and incidence of rupture of abdominal aortic
aneurysms. J Vasc Surg 1991;14:540-8.
7. Katz DA, Cronenwett JL. The cost-effectiveness of early
surgery versus watchful waiting in the management of small
abdominal aortic aneurysms. J Vasc Surg 1994;19:980-91.
8. Cronenwett JL, Sargent SK, Wall MH, Hawkes ML,
Freeman DH, Dain BJ, et al. Variables that affect the expansion rate and outcome of small abdominal aortic aneurysms.
J Vasc Surg 1990;11:260-9.
9. Darling RC, Messina CR, Brewster DC, Ottinger LW.
Autopsy study of unoperated abdominal aortic aneurysms.
Circulation 1977;56(II Suppl):161-4.
10. Szilagyi DE, Smith RF, DeRusso FJ, Elliott JP, Sherrin FW.
Contribution of abdominal aortic aneurysmectomy to prolongation of life. Ann Surg 1966;164:678-99.
11. Cronenwett JL, Murphy TF, Zelenock GB, Whitehouse
WM, Lindenauer SM, Graham LM, et al. Actuarial analysis of
variables associated with rupture of small abdominal aortic
aneurysms. Surgery 1985;98:472-83.
12. Raghavan ML, Webster MW, Vorp DA. Ex-vivo biomechanical
behavior of abdominal aortic aneurysm: assessment using a
new mathematical model. Ann Biomed Eng 1996;24:573-82.
13. Vorp DA, Raghavan ML, Muluk SC, Makaroun MS, Steed
DL, Shapiro R, et al. Wall strength and stiffness of aneurysmal and nonaneurysmal abdominal aorta. Ann NY Acad Sci
1996;800:274-6.
14. Stringfellow MM, Lawrence PF, Stringfellow RG. The influence of aorta-aneurysm geometry upon stress in the
aneurysm wall. J Surg Res 1987;42:425-33.
15. Mower WR, Baraff LJ, Sneyd J. Stress distributions in vascu-

16.

17.

18.

19.

20.

21.
22.

23.

24.
25.

26.

27.

28.

639

lar aneurysms: factors affecting risk of aneurysm rupture. J

Surg Res 1993;55:155-61.
Inzoli F, Boschetti F, Zappa M, Longo T, Fumero R.
Biomechanical factors in abdominal aortic aneurysm rupture.
Eur J Vasc Surg 1993;7:667-74.
Elger DF, Blackketter DM, Budwig RS, Johansen KH. The
influence of shape on the stresses in model abdominal aortic
aneurysms. J Biomech Eng 1996;118:326-32.
Sacks MS, Vorp DA, Raghavan ML, Federle MP, Webster
MW. A noninvasive surface geometric analysis of in-vivo
abdominal aortic aneurysms. American Society of Mechanical
Engineers Proc Bioeng Conf 1997;35:235-6.
Dobrin PB. Pathophysiology and pathogenesis of aortic
aneurysms: current concepts. Surg Clin North Am
1989;69:687-703.
Roach MR, Maclean NF. Changes in geometry of human
aortas with the development of abdominal aortic aneurysms.
In: Hosoda S, Yaginuma T, Sugawara M, Taylor MG, Caro
CG, editors. Recent progress in cardiovascular mechanics.
Chur, Switzerland: Harwood Academic Publishers; 1994. p.
269-84.
Hughes TJR. The finite element method. Englewood Cliffs,
N.J.: Prentice-Hall; 1987. p. 803
Chuong CJ, Fung YC. Compressibility and constitutive
equation of arterial wall in radial compression. J Biomech
1984;17:35-40.
Bluestein D, Niu L, Schoephoerster RT, Dewanjee MK.
Steady flow in an aneurysm model: correlation between fluid
dynamics and blood platelet deposition. J Biomech Eng
1996;118:280-6.
Shigley JE, Mitchell LD. Mechanical engineering design.
New York: McGraw-Hill; 1983. p. 869.
Vorp DA, Gorcsan J III, Webster MW. The potential influence of intraluminal thrombus on abdominal aortic aneurysm
as assessed by a new noninvasive method. Cardiovasc Surg
1996;4:732-9.
Simoni G, Perrone R, Cittadini G Jr, De Caro G, Baiardi A,
Civalleri D. Helical CT for the study of abdominal aortic
aneurysms in patients undergoing conventional surgical
repair. Eur J Vasc Surg 1996;12:354-8.
He CM, Roach MR. The composition and mechanical properties of abdominal aortic aneurysms. J Vasc Surg 1994;20:613.
Vorp DA, Raghavan ML, Rajagopal KR, Webster MW. A
nonlinear hyperelastic constitutive model for abdominal aortic aneurysm. Ann Biomed Eng 1996;24(1 Suppl):45.

Submitted Aug. 1, 1997; accepted Jan. 8, 1998.