See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/6422586

Modeling Plaque Fissuring and Dissection during Balloon Angioplasty

Intervention

Article in Annals of Biomedical Engineering · June 2007

DOI: 10.1007/s10439-007-9258-1 · Source: PubMed

CITATIONS READS

97 352

2 authors:

Thomas Christian Gasser Gerhard Holzapfel

KTH Royal Institute of Technology Graz University of Technology
208 PUBLICATIONS 14,679 CITATIONS 520 PUBLICATIONS 35,821 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Gerhard Holzapfel on 15 May 2016.

The user has requested enhancement of the downloaded file.

Annals of Biomedical Engineering, Vol. 35, No. 5, May 2007 (
2007) pp. 711–723
DOI: 10.1007/s10439-007-9258-1

Modeling Plaque Fissuring and Dissection during

Balloon Angioplasty Intervention
T. CHRISTIAN GASSER1 and GERHARD A. HOLZAPFEL 1,2
1
Department of Solid Mechanics, School of Engineering Sciences, Royal Institute of Technology (KTH), Osquars Backe 1,
SE-100 44, Stockholm, Sweden; and 2Institute for Biomechanics, Center for Biomedical Engineering, Graz University of
Technology, 8010 Graz, Austria
(Received 1 September 2006; accepted 9 January 2007; published online 24 March 2007)

Abstract—Balloon angioplasty intervention is traumatic to practice. In particular, the fracture mechanism occur-
arterial tissue. Fracture mechanisms such as plaque fissuring ring during balloon angioplasty,41 which to model is
and/or dissection occur and constitute major contributions to one focus of the present work, is fundamentally trau-
the lumen enlargement. However, these types of mechani-
cally-based traumatization of arterial tissue are also contrib- matic to arterial tissue.6,61 Balloon angioplasty involves
uting factors to both acute procedural complications and denudation of the endothelium, disruption of the intima
chronic restenosis of the treatment site. We propose physical and the atherosclerotic plaque with frequent separation
and finite element models, which are generally useable to from or dissection of the media, and overstretching of
trace fissuring and/or dissection in atherosclerotic plaques remnant non-diseased tissue in lesions (see Humphrey37
during balloon angioplasty interventions. The arterial wall is
described as an anisotropic, heterogeneous, highly deform- and references therein). Plaque fissuring and/or local
able, nearly incompressible body, whereas tissue failure is dissection (probably secondary to plaque fissuring)
captured by a strong discontinuity kinematics and a novel are particular forms of balloon angioplasty-induced
cohesive zone model. The numerical implementation is based arterial trauma involving laceration and/or cleavage of
on the partition of unity finite element method and the the arterial wall.9,41,47 Angioplasty-induced dissections
interface element method. The later is used to link together
meshes of the different tissue components. The balloon have been implicated as a contributing factor to both
angioplasty-based failure mechanisms are numerically stud- acute procedural complications (abrupt re-closure,
ied in 3D by means of an atherosclerotic-prone human ischemia, myocardial infarction, emergency surgery
external iliac artery, with a type V lesion. Image-based 3D and coronary micro-embolization1,10,11,36,55) and
geometry is generated and tissue-specific material properties chronic restenosis of the treatment site.38
are considered. Numerical results show that in a primary
phase the plaque fissures at both shoulders of the fibrous cap Although arterial dissection is a frequently occur-
and stops at the lamina elastica interna. In a secondary ring phenomenon and a challenging clinical entity, the
phase, local dissections between the intima and the media underlying biomechanics remains largely unclear. Only
develop at the fibrous cap location with the smallest a few studies investigate the underlying biomechanics
thickness. The predicted results indicate that plaque fissuring of failure initiation and propagation in atherosclerotic
and dissection cause localized mechanical trauma, but
prevent the main portion of the stenosis from high stress, plaques. For experimental studies see the reviews on
and hence from continuous tissue damage. the biomechanics of atherosclerotic tissues and plaque
rupture by Richardson50,51 and Salunke and Topoles-
Keywords—Artery, Atherosclerotic plaque, Balloon angio- ki,56 and the literature survey on failure properties of
plasty, Cohesive zone model, Dissection, FEM, Plaque the aorta by Sommer et al.58 In the later experimental
fissuring. study58 also dissection properties of the human aortic
media are documented allowing the development of
suitable constitutive models. It seems that only one
INTRODUCTION computational model is yet available in the literature
Arterial dissections may occur spontaneously or that allows the study of failure propagation in arterial
traumatic and are frequently observed in clinical tissue caused by balloon angioplasty.16 Computational
models are best suited to uncover the mechanisms of
arterial failure, and hence helpful to optimize balloon
Address correspondence to Gerhard A. Holzapfel, Department angioplasty interventions.
of Solid Mechanics, School of Engineering Sciences, Royal Institute
of Technology (KTH), Osquars Backe 1, SE-100 44, Stockholm, Besides three-dimensional morphological data,
Sweden. Electronic mail: [email protected] the constitutive response of the involved tissue
711
0090-6964/07/0500-0711/0
2007 Biomedical Engineering Society
712 T. C. GASSER AND G. A. HOLZAPFEL

components is of fundamental importance in order to components occurring during balloon dilatation of this
reliably predict the evolutions of tissue stresses and specific atherosclerotic lesion are discussed in detail,
strains during balloon angioplasty. The (rubber-like) and possible consequences are deduced.
non-collagenous matrix, the (leather-like) collagen
fibers, and the smooth muscle cells control the
mechanical properties of arterial walls. In particular, in MODELING ASSUMPTIONS
the media of elastic arteries these three components are
present in the medial lamellar units,7,65 each of which is In this section the modeling assumptions are briefly
about 10 lm thick.49 The highly organized structure of described. In particular the underlying continuum
the arterial wall49 causes its local cylindrical ortho- mechanical basis and the proposed finite element
tropy,46,64,67 and, in particular, its laminated structures implementation are discussed to some extent. An
are prone to split by creating a cleavage plane between artery, when expanded during balloon angioplasty, is
adjacent lamellae. subject to (quasi-static) mechanical loads which are
The present work aims to model fissuring of tissue much higher than the loads under physiological
components, and dissection from each other typically conditions. Hence, it is justified to neglect all dynamic
occurring during balloon angioplasty intervention. We effects inherent to the cardiovascular system so that we
pursue the strong discontinuity approach, and model a pursue a quasi-static approach herein. A comprehen-
material surface separation by a jump in the displace- sive introduction and derivation of the underlying
ment field. We postulate the existence of a fracture framework is provided by Gasser and Holzapfel.19
process zone and regard plaque fissuring and dissection
as a gradual process in which separation between
material surfaces is resisted by a cohesive traction. In Continuum Mechanical Basis
particular, recent histological investigations58 highlight Strong Discontinuity Kinematics
that (collagen and elastin) fiber bridging might play a
dominant role in arterial dissection, which motivates to We pursue the strong discontinuity approach in
use this approach rather than the theory of sharp crack which the discontinuity in the displacement field
tips. We employ a (discrete) constitutive description of describes tissue failure, i.e. fissuring and/or dissection.
the cohesive zone, which is based on an isotropic We assume that a discontinuity ¶W0d separates a body W0
traction separation law with isotropic damage. into two sub-bodies, which themselves occupy the sub-
The proposed computational 3D failure model domains W0+ and W0-, so that @X0d \ X0þ > \X0
¼ ;
combines the cohesive zone representation of failure and X0þ [ @X0d [ X0
¼ X0 . Here we denote elements
with the partition of unity finite element method in the reference configuration by the subindex 0. For
(PUFEM) and the interface element method (IEM). In simplicity, a single discontinuity is assumed to describe
particular, IEM is applied to interfaces with non- the underlying kinematics. However, note that accord-
matching nodes allowing the discretization of the ing to the numerical schemas described below, we can
involved tissue components independently from each handle multiple discontinuities, and also interactions
other and to link them together. The proposed between tissue fissuring and dissection.
numerical concept is based on a consistent lineariza- A deformation vðXÞ, applied to the reference con-
tion of the underlying variational formulation which figuration, maps W0+ and W0- into their current con-
serves as a basis for an effective and robust numerical figurations W+ and W), where X denotes the referential
description of failure propagation in arterial tissue. position of a material point. Consequently, the
The balloon angioplasty-based failure mechanism is displacement at X is assumed to be uðXÞ ¼
numerically studied by means of an atherosclerotic- uc ðXÞ þ HðXÞue ðXÞ, where uc and ue are regular and
prone human external iliac artery. The considered enhanced displacement fields, respectively.2,44 Here
region of the atherosclerotic lesion is of type V HðXÞ denotes the Heaviside function with the values 0
according to the histological classification of Stary.59 and 1 for X 2 X0
and X 2 X0þ , respectively. In
The model’s geometry was generated from high-reso- addition, we introduce the unit normal vector N(X),
lution magnetic resonance imaging (hrMRI) data, and which yields the orientation of @X0d in X, see Fig. 1.
the tissue-specific material, structural and cohesive Standard derivation24,43 and the use of the property
parameters were considered. The balloon angioplasty GradHðXÞ ¼ dd NðXd Þ defines the corresponding
intervention is numerically performed by solving the deformation gradient
balloon/artery interaction as a 3D contact problem.
FðXÞ ¼ I þ Grad uðXÞ ¼ I þ Grad uc ðXÞ
Novel results on plaque fissuring, tissue dissection and
on the stress distribution in the individual tissue þ HGrad ue ðXÞ þ dd ðXÞue ðXÞ  NðXd Þ; ð1Þ
 Modeling Plaque Fissuring and Dissection during Balloon Angioplasty Intervention 713

FIGURE 1. Strong discontinuity kinematics.

which serves as the basic expression for the right and left Z 9
Cauchy-Green tensors, C = FTF and b = FFT, symðgradc duc Þ : rc dv >
>
>
>
respectively. Here I denotes the identity tensor and the >
>
X
Z >
>
material gradient operator is defined by GradðÞ ¼ >
>
>
@ðÞ=@X. In addition, dd denotes the Dirac-delta func- þ symðgrade duc Þ : re dv
dPc ðduc Þ ¼ 0;>
ext >
>
>
>
tional with dd = 0 and dd ! 1 for X 62 @X0d and X
=
Z þ
ð2Þ
X 2 @X0d , respectively. >
symðgrade due Þ : re dv >
>
In order to provide a discrete constitutive descrip- >
>
>
>
tion of the cohesive traction acting at the displacement Xþ Z >
>
>
>
discontinuity, we introduce the gap displacement b uðXÞ þ t  due ds
dPext >
>
e ðdue Þ ¼ 0; >
>
and the average displacement uðXÞ related to a point >
;
@Xd
x located at the fictitious spatial discontinuity @Xd , see
Fig. 1. The spatial orientation of @Xd in x is defined where dPcext and dPeext are external contributions
by the unit normal vector n which can be interpreted which refer to the domains W0- and W0+, respectively.
as a weighted push-forward operation of the covariant In addition, dv and ds are the spatial volume and
vector N. The fictitious spatial discontinuity @Xd is surface elements, respectively.
assumed to be in the middle between the two physical Here, rc ¼ J
1 T
1 T
c PðFc ÞFc and re ¼ Je PðFe ÞFe denote
surfaces @Xdþ and @Xd
, which originate from @X0d . the Cauchy stress tensors and t = TdS/ds is the Cauchy
According to the introduced kinematics, a referen- traction vector associated with a fictitious discontinuity
tial point X maps into the two spatial points x+ and ¶Wd, while T is the first Piola-Kirchhoff traction vector
x), which uniquely defines the gap and average dis- associated with the related fictitious discontinuity @X0d .
placements of the discontinuity, i.e. b u ¼ xþ
x
and The spatial gradients in (2) are defined according to
u ¼ ðxþ þ x
Þ=2, see Fig. 1. gradc ðÞ ¼ GradðÞF
1
1
c ; grade ðÞ ¼ GradðÞFe , and
T
sym() = (() + () )/2 furnishes the symmetric part of
Variational Formulation
(). For a consistent linearization of the statements (2), we
The variational formulation for a quasi-static finite refer to Gasser and Holzapfel,15 and references therein.
element model Ris based on a single-field variational
principle,24 i.e. X0 Grad du : PðFÞdV
dPext ðduÞ ¼ 0,
Interface Representation of the Discontinuity
where P(F) and du denote the first Piola-Kirchhoff stress
tensor and the admissible variation of the displacement It is convenient to focus here on the referential vari-
field u, respectively. The integration is taken over the ational formulation, which, for a quasi-static interface
reference configuration W0, where dV denotes the ref- element model, is based
R on the single-field variational
erential volume element. In addition, the contributions principle45 dP ¼ @X0d T  db udS ¼ 0. We assume now
due to external loading are summarized in the virtual that19 T ¼ T½buðxÞ; nðb
uðxÞÞ; d, and, therefore, T depends
external potential energy dPext. We assume that loads on the gap displacement b u, the unit normal vector n onto
do not depend on the deformation of the body. the fictitious discontinuity @X0d , and d 2 ½0; 1 which
According to the introduced displacement field its denotes an internal damage variable recording the
admissible variation reads du ¼ duc þ Hdue , and after history of the failure process. By neglecting external
some algebraic manipulations, we get the two spatial loading on the interface the consistent linearization of
variational statements15 the traction term gives
714 T. C. GASSER AND G. A. HOLZAPFEL
Z Z
According to Holzapfel and Weizsäcker35 we assume
DdP ¼ u  Cbu Dbu dS þ
db u  Cn Du ndS
db an additive decomposition of the strain-energy function.
@X0d
Z
@X0d In addition, we propose that the isochoric strain energy
W stored in the non-collagenous and collagenous
P com-
þ u  Cd Dbu ddS;
db ð3Þ
ponents is according to WðC; Hi Þ ¼ Wg ðCÞ þ i¼1;2 Wfi
@X0d ðC; Hi ða0i ; jÞÞ, where C denotes the modified right
where Cbu ¼ @T=@b u; Cn ¼ @T=@n; Cd ¼ @T=@d describe Cauchy-Green tensor.24 While the non-collagenous
the stiffness of the cohesive zone with respect to the contribution is modeled by the neo-Hookean strain-
gap displacement, the rotation and the growing energy function Wg ðCÞ ¼ lðtrC
3Þ=2, the strain en-
damage. The notation Da() indicates the Gâteaux ergy stored in the collagen fibers is proposed to be
derivative of () in the direction of a, which is captured by the function20
explained in more detail by Gasser and Holzapfel15 k1
and Holzapfel.24 Wfi ðC; Hi Þ ¼ fexp½k2 Ei 2 
1g;
2k2 ð5Þ
Ei ¼ Hi : C
Hi : I; i ¼ 1; 2;
Constitutive Formulation where Hi : C denotes an invariant of the symmetric
In order to work with the above introduced varia- generalized structure tensors Hi, and the symmetric
tional statements, we have to provide continuous and tensor C. Consequently, the set (l,k1,k2) of the
discontinuous constitutive formulations to describe the material parameters needs to be quantified from
bulk and the cohesive material responses. We assume experimental data, while the set (a0i,j) of the structure
both constitutive formulations to be independent from parameters is characterized by the histology of the
each other. related tissue component.
We start with the bulk constitutive model, and We proceed now by describing the cohesive con-
review that the arterial wall is anisotropic, heteroge- stitutive model. The complex irreversible changes in
neous, highly deformable, nearly incompressible and the microstructure of the individual tissue components
shows pseudo-elastic behavior.20,27,29,37 In the present due to, e.g., fissuring and dissection are lumped into a
work, the complex architecture of an arterial layer (discrete) cohesive zone, which is mechanically defined
is represented mechanically as a fiber-reinforced by the cohesive potential w. A justification for the
composite,25,27 in which two families of collagen fibers application of the theory of cohesive zones rather than
are embedded in an isotropic groundmatrix. In addi- assuming sharp crack tips is mainly motivated by the
tion, the dispersion of the collagen fiber orientation, (collagen and elastin) fiber cross-bridging, as can be
as pointed out in, e.g., Finlay et al.,12 is also repre- seen through histological images.58
sented.20 In particular, we introduce (symmetric)
R gen-
Within the present work we employ an isotropic
1 cohesive model similar to the Rose, Smith and Ferr-
eralized structure tensors, defined by Hi ¼ 4p x qðMÞ
M  Mdx, which serve as a continuum representation ante universal binding law,53,54 and used in the recent
of dispersed fiber orientations within the ith family work by Gasser and Holzapfel,19 namely
of collagen fibers. Here q(M) denotes the normal- c
ized density function of the referential orientation wði1 ; dÞ ¼ expð
adÞi1 ; ð6Þ
2
M, |M| = 1, and dx ¼ sin HdHdU is the unit
sphere with the two Eulerian angles H 2 ½0; p and where i1 ¼ bub
u is the first invariant of the symmetric
U 2 ½0; 2p. tensor bubu and the non-negative parameters a,c and
Based on experimental quantifications of collagen d characterize the traction separation properties of
formation in arterial layers,12 we assume a transversely the cohesive zone. In particular, d 2 ½0; 1 denotes
isotropic distribution of the collagen within a partic- an internal damage variable. The term exp()ad) is
ular fiber family such that equivalent to (1 ) d) known as the reduction factor
from the isotropic continuum damage theory, where
Zp the damage variable d 2 ½0; 1 is the ratio between
1
Hi ¼ jI þ ð1
3jÞa0i  a0i ; j¼ qðHÞ sin3 HdH the damaged and the initial cross sectional areas of
4
0 a material point.24,39 With standard arguments,
ð4Þ the Coleman-Noll procedure8 implies the physical
expression
hold. Here, a0i denotes the (mean) preferred direction
of the collagen, whereas j characterizes the related @w
T¼ ¼ c expð
adÞb
u ð7Þ
dispersion around a0i.20 @b
u
 Modeling Plaque Fissuring and Dissection during Balloon Angioplasty Intervention 715

for the first Piola-Kirchhoff traction vector T. From (7) Partition of Unity Finite Element Method (PUFEM)
it is clear that c has the meaning of the initial stiffness
According to the PUFEM, the displacement field u
of the cohesive zone.
is interpolated as
Moreover, with the introduction of the R 1 mode I
fracture
R1 energy, denoted by GI , as G I ¼ 0 T  db
u¼ X
nelem X
nelem

c expð
adÞb u db
u ¼ c=a 2
, and the cohesive tensile u¼ NI uIc þ H NI uIe ; ð9Þ
0
I¼1 I¼1
strength T0, which is the maximum traction obtained
from model (7) to be T0 = cexp()1)/a, we may deduce where NI are the standard (polynomial) shape func-
the useful relations tions, and I is an index running between 1 and the total
number of finite element nodes, denoted by nelem. Note
T20 expð2Þ T0 expð1Þ
c¼ ; a¼ ð8Þ that characters indicated by underlines denote the
GI GI matrix notation of the associated tensor or vector (for
for the introduced material parameters c and a, example, u is the matrix representation of vector u). In
respectively. Finally, in order to complete the cohesive (9), regular and enhanced nodal displacements are
description, a damage surface /ðb u; dÞ ¼ jb
uj
d ¼ 0 in denoted by uI c and uI e, respectively, and the Heaviside
the 3D gap displacement space is introduced, and it is function H enriches (9) in order to achieve good local
assumed that d_ ¼ jb u_ j captures the evolution of the approximation properties.
internal (damage) variable d. Interpolation (9) in conjunction with the variational
The proposed model has the main advantage of a statements (2) render the following linearized algebraic
smooth transition from the stiffening into the soft- set of equations for a particular finite element node I
ening branch (continuous envelope) of the cohesive to15
zone, which is particularly helpful for dynamical
I
I
ext I
int I
Kuc uc Kuc ue Duc f uc f
computations. However, the capability of the cohesive ¼ ext
uc ; ð10Þ
model to fit experimental data is limited, e.g., relation Kue uc Kue ue i
1 Due i f ue i
1 f intue i
1
(8)1 indicates that a large fracture energy causes a
where i, i)1 denote the iteration steps associated with a
low initial stiffness, which might not always meet
global Newton iteration, and Duc and Due denote the
experimental observations. The introduction of an
increments of the regular and the enhanced nodal
additional material parameter, as shown by Gasser
displacements, respectively.
and Holzapfel,14 can, to some extend, circumvent this
In (10) the vectors fext ext int int
uc , fue and fuc , fue denote the
drawback.
external and internal nodal force vectors, respectively.
Finally, the penalty constraint wpen ¼ cpen hb uni2 =2
For a detailed derivation of these vectors, and the
against penetration is added to the cohesive potential
contribution of the sub-matrices KðÞ (with the abbre-
(6), where hðÞi ¼ ½ðÞ þ jðÞj=2 , defines the Macau-
viation () standing for ucuc, ucue, ueuc or ueue) to the
ley bracket function, while cpen is a non-negative pen-
nodal stiffness matrix, as introduced in (10).19 Finally,
alty parameter. The penalty formulation is exact when
we emphasize that the cohesive traction T causes an
cpen ! 1. This approach is suitable as long as small
off-diagonal contribution Kue uc and a diagonal contri-
sliding of the interface is present, however, for large
bution Kue ue to the element stiffness matrix.14,15
sliding the term needs to be replaced by a standard
The proposed finite element model has been imple-
contact formulation.66
mented into the multi-purpose finite element analysis
program FEAP.60 In addition, a separate package was
developed to handle the geometrical representation
Finite Element Implementation and propagation of the developing cracks, which are
We restricted our finite element model in the sense utilized by the macro command language in FEAP
that dissection type of failure can only propagate along after each load step.18
interfaces between the different arterial layers, e.g.,
along the lamina elastica interna and externa. In con-
Interface Element Method (IEM)
trast, tissue fissuring is under no such restriction; it is
solely driven by the stress field. According to these In order to allow the generation of finite element
assumptions, dissection failure is numerically repre- meshes for the individual tissue geometries independent
sented by the IEM, while fissuring of layers is captured from each other, we use the IEM for non-matching
by the PUFEM. The two different numerical imple- nodes to link together the different three-dimensional
mentations, i.e. IEM and PUFEM of arterial failure, geometries. Note that the considered IEM must handle
are discussed in the following. the enhanced degrees of freedom used in PUFEM in
716 T. C. GASSER AND G. A. HOLZAPFEL

order to allow an interaction between the fissuring and balloon angioplasty, see, e.g., the study by Coulden
dissection of the individual tissue layers. et al.9 In particular, we have chosen an atherosclerotic-
Consequently, we introduce prone human external iliac artery (65 years, female),
if if
see Fig. 2(a), from which the different tissue types were
nelem nelem
X X prepared and mechanically tested in axial and cir-
b
u¼ bI uIc þ H
N bI uIe ;
N cumferential directions. For the related anamnesis and
I¼1 I¼1
if if
ð11Þ the mechanical data (uniaxial tensile stress-stretch
nelem nelem
X X responses and ultimate tensile stresses and stretches) of
u¼ NI uIc þ H NI uIe ; the different tissues, see Holzapfel et al.,32 specimen I
I¼1 I¼1
therein. In addition, the three-dimensional recon-
where b u and u denote the gap displacement and struction of the most important tissue boundaries of
average displacement located at the fictitious spatial this specific atherosclerotic artery is performed by
discontinuity, and N bI and NI are the interpolation means of hrMRI and a specifically developed software
functions associated with the Ith node, related to b u tool which is based on a deformable model. Details on
and u, respectively. The index I in the sums of (11) the method, the reconstruction technique and the val-
runs between 1 and the total number of nodes of the idation approach is provided in the recent work by
interface element, denoted by nifelem . Auer et al.3 (for a longitudinal section see also Fig. 1 in
According to relation (3), a straightforward alge- Holzapfel et al.,33 where the same atherosclerotic
braic manipulation gives the following linearized artery was used to analyze the changes in its mechan-
algebraic set of equations for a particular node I on the ical environment during interaction with different
interface element stents). As can be seen from the segmented macro-

if I
I
if I scopic view in Fig. 2(a), the artery consists of seven
K Kif Duc f different tissue components represented by the bor-
¼
; ð12Þ
Kif Kif i
1 Due i f if i
1 derlines32 (A = adventitia, M-nos = non-diseased
media, I-nos = non-diseased intima, I-fc = fibrous
where i, i)1 denote the iteration steps associated with a
cap, I-fm = fibrous intima at the medial border,
global Newton iteration, and Duc , Due denote again
I-lp = lipid pool, M-f = fibrous (diseased) media).
the increments of the nodal displacements. The nodal
Based on the histological composition and structure
force vector fIif due to internal loading, and the stiff-
this atherosclerotic lesion is of type V59 (for a parti-
ness sub-matrix KifI of the Ith node are defined as
tioning of the pathogenesis of atherosclerosis into
Z Z
different stages of plaque formation see Fig. I in
f ifI ¼ bI dS; Kif ¼ bI C u N
b
TN I N b J dS Stary59). A type V lesion (fibroatheroma) contains
@Xe0d @Xe0d mainly reparative smooth muscle cells and fibrous
Z
tissue and, additionally, two or more lipid pools of

bI Cn ngradNJ dS;
N ð13Þ unequal size separated from each other by cells and
@Xe0d fibrous tissue.
For our numerical investigation we have now dis-
respectively, where @Xe0d denotes the discontinuity do-
cretized the different tissue components of this ath-
main of the considered interface element in the reference
erosclerotic lesion by means of tetrahedral finite
configuration. It is important to note that in the case
elements. For simplicity, we combined M-nos and M-f
that the enhanced degrees of freedoms are active, the
to one tissue component, labeled as ’media’, and I-nos,
nodal stiffness in (12) becomes rank deficient, and hence
I-fc and I-fm to another tissue component, labeled as
the interface formulation can only be used in combina-
’intima’. Subsequently, we associate the media with the
tion with the partition of unity (PU) finite elements.
material properties of the non-diseased media (M-nos),
The particular IEM with the related model has been
and the intima with the fibrous cap (I-fc). Finally, we
realized by linking together two triangular facets.
consider just a slice of the atherosclerotic lesion as-
Hence, the interface element has 6 nodes, and it has
sumed to be at rest and stress free in the unloaded
been implemented into FEAP by using the user-defined
(initial) configuration, and composed of four different
element routine.
tissue components, as can be seen in Fig. 2(b).

BALLOON ANGIOPLASTY MODEL Definition of the Material Axis

In this section plaque fissuring and local dissection In order to apply the anisotropic constitutive model,
of the atherosclerotic plaque are studied in more detail, as proposed above, the (mean) preferred direction of
since this type of failure is frequently identified after the collagen a0i and the related dispersion j have to be
 Modeling Plaque Fissuring and Dissection during Balloon Angioplasty Intervention 717

FIGURE 2. Human external iliac artery: (a) segmented macroscopic view of a type V lesion according to the histological clas-
sification of Stary59 (adopted from Holzapfel et al.32). The borderlines represent the different tissue components. A = adventitia, M-
nos = non-diseased media, I-nos = non-diseased intima, I-fc = fibrous cap, I-fm = fibrous intima at the medial border, I-lp = lipid
pool, and M-f = fibrous (diseased) media); (b) 3D discretization of a slice of the atherosclerotic lesion composed of four different
tissue components (labeled as adventitia, media, intima, lipid pool; for an explanation see the text) by means of tetrahedral finite
elements. The individual tissues are meshed separately and linked together using the interface element method.

defined. For our analysis structural homogeneity is can then be defined in a straightforward way. Subse-
assumed, i.e. j is constant and a0i has the same com- quently, a pull-back operation with the rotation tensor
ponents with respect to a local coordinate system QðÞ ¼ Q
1 ðÞ, with () denoting ðR; H; ZÞ , defines
ðR; H; ZÞ within each tissue component (see Fig. 3). the radial R, circumferential H and axial Z directions
Apart from the histological quantification of a0 i and j, in W0, see Fig. 3. Here, the orthogonal tensor Q is
the local coordinate system, i.e. the local material axes defined according to the multiplicative decomposition
ðR; H; ZÞ, has to be defined in order to work with an F ¼ QU, where U denotes the right stretch tensor.24
anisotropic constitutive model. Note that the proposed method requires the definition
Although diffusion MRI, which is based on the of the configuration X0 , and this has some influence on
underlying water diffusion properties, is basically able the prediction of the material axes. In the current
to determine the microstructure of different athero- study, on a trial and error basis, we attempted to
sclerotic components,62 this technique does not pro- generate the eccentric thick-walled tube X0 such that
vide the resolution required for the present study. the deformation v is minimized.
To overcome this shortcoming, we propose a novel
theoretically-oriented method allowing the definition
Modeling Details
of a local coordinate system ðR; H; ZÞ within all tissue
components of the whole stenotic artery. The basic While we had access to reliable geometrical data of
idea is to map the reference configuration W0 of the the involved tissue types, their structural quantification
artery to an eccentric thick-walled tube X0 , according is partly based on assumptions. In particular, there are
to the deformation v with F ¼ @v=@X, as can be seen no data available in the literature regarding the (local)
in Fig. 3. Hence, in view of the simple geometry of X0 , collagen structure of stenotic iliac arteries. However,
the system ðR; H; ZÞ located in X0 , i.e. the local radial structural data can, to some extend, be estimated from
R, the circumferential H and the axial Z directions, mechanical tests. To this end the structural parameters
can be treated like material parameters and both

TABLE 1. Material and structural parameters describing the

bulk material of the involved arterial tissue components
(taken from Holzapfel et al.30).

Tissue l (kPa) k1 (kPa) k2 (-) a () j (-)

Intima 78.9 23.7 26.3 0.0 0.0

Media 15.0 4.0 2.3 7.0 0.0
Adventitia 1.75 65.6 61.8 49.0 0.0
FIGURE 3. Transformation of the reference configuration W0 Lipid Pool 0.1 0.0 – – –
of the stenotic artery to an eccentric thick-walled tube X0 .
718 T. C. GASSER AND G. A. HOLZAPFEL

(structural and material) parameters can then be Fig. 2 to be stress free. The different tissue components
estimated from mechanical experiments. In the present were discretized separately using tetrahedral second-
analysis we adopted parameters defined in that way. order finite elements and the mesh was generated with
They are provided by Holzapfel et al.30 Note that the software-package NETGEN.57 Subsequently, the
the (mean) preferred direction a0i of the collagen is different meshes, with non-matching nodes at the
assumed to have no radial component in the local interfaces, were linked together by using the IEM, as
coordinate system ðR; H; ZÞ such that a0i can uniquely discussed in the previous section. A program that has
be described by an angle a defined between the fiber been specifically developed is able to detect the differ-
reinforcement (orthotropy) and the circumferential ent interfaces in an automatic way and provides the
direction H in the individual layers. The employed input data for FEAP.
structural and material parameters of the involved The mechanics of the balloon angioplasty interven-
arterial tissue components are summarized in Table 1. tion is modeled as a 3D contact problem, where the
Next, we quantify the failure properties of the mechanical load caused by the balloon/artery contact
individual tissue components. There are not much data finally leads to tissue failure. For simplicity we assumed
available in the literature, and hence some estimations the balloon to be a circular cylinder with continuously
are required. The estimation of the tissue resistance to increasing diameter. This approach seems to capture
fissuring is based on the literature,32,40,48 and on the real situation at higher inflation pressure, which
preliminary experiments performed in our own labo- coincides with the loading state at which tissue failure is
ratory. Therefore, these data need to be seen as a rough expected. Note that at lower inflation pressures a bal-
estimation from available (published and unpublished) loon of, e.g., the Grüntzig type, is not yet unfolded such
sources. In particular, no data are available to quantify that the pressure in the balloon acts directly on the
the softening region, i.e. the fracture energy and/or the artery, and hence a pressure-boundary condition seems
shape of the cohesive law for tissue fissuring. The to be more realistic. In our study we do not model the
dissection properties are based on recent experimental unfolding process of the balloon, which takes place at
results obtained from peeling (healthy) aortic medias,58 low inflation pressure at which complex (dynamic)
and these data aim to represent the cohesive properties unfolding effects might be present.
of the lamina elastica interna (intima/media interface), For the present computation a specifically devel-
and the lamina elastica externa (media/adventitia oped automatic crack initialization algorithm captures
interface). The used cohesive parameters describing the onset of tissue fissuring. To this end crack root
tissue fissuring and dissection are summarized in elements are adopted for the case that the stress-based
Table 2. failure initialization criterion is met.18 For algorithmic
According to our experience, highly stenotic arteries purposes we have to avoid that crack root elements are
show very little or no axial in situ pre-stress, i.e. the too close to existing tissue fissures. (For the case that
pre-stretch in the axial direction is approximately 1, crack root elements are initialized in a too narrow
which is not the case for healthy arteries. Hence, in distance to existing cracks, the developing new crack
order to account for this, we fixed the bottom and top might interact with the existing one, which cannot be
faces of the analyzed slice in the axial direction. In handled by the current implementation.) In the present
addition, we neglected residual strains in the load-free analysis
pffiffiffiffiffiffi we chose a minimum tolerable distance of
configuration, and assumed the geometry shown in 4:0 3 Ve , with Ve denoting the referential element
volume.

Predicted Results
TABLE 2. Cohesive parameters describing fissuring of and
dissection from the involved arterial tissue components. At the beginning of the computation, where the
Cohesive zone T0 (kPa) Gf (mJ) balloon/artery contact occurred, the 3D problem
could not be solved in a quasi-static way. Therefore, to
Fissuring of the Intima 400.0 40 stabilize the scheme we added some numerical viscosity
Fissuring of the Media 500.0 100
Fissuring of the Adventitia 2000.0 300
and solved the resulting first-order transient problem
Lamina elastica interna 140.1 6.35 using the Newmark method provided by FEAP. The
(Intima/media dissection) added viscosity was (very) small so that it did not affect
Lamina elastica externa 140.1 6.35 the development of plaque fissuring and dissection in
(Media/adventitia dissection) the atherosclerotic lesion, which occurs at a much
Intima/lipid pool interface rigid rigid
higher load. After 132 load steps and a balloon
The values for T0 and Gf are estimated from experimental diameter of db = 4.42 mm, the large mesh distortion
tests.32,40,48,58 in the lipid pool at the dissection site required a
 Modeling Plaque Fissuring and Dissection during Balloon Angioplasty Intervention 719

termination of the computation. Re-meshing would where the fibrous cap is thinner. As long as intact, the
have been required to overcome this drawback of the innermost layer of the lesion carries a significant part
numerical model, which was not considered in the of the load. This load carrying mechanism is similar to
present study. that of an inflated thick-walled tube, where stress
The deformed state of the stenosis at the maximum concentrations also occur at the inner wall. By tearing
lumen diameter of db = 4.42 mm is shown in Fig. 4. the intima and (partly) dissecting it from the media, a
For clarity the PU finite elements, which are those redistribution of the stresses occurs from the intima to
elements at which the failure criterion were met and the media, as clearly shown in Fig. 5. As expected, the
discontinuities have been embedded, are removed. As stress in the adventitia increases during balloon infla-
can be seen from the results in Fig. 4, the plaque tion, however, the stress level remains relatively low,
fissured at both shoulders of the intima, i.e. where the even behind the dissection. A possible explanation for
fibrous cap converts into the (healthy) intimal tissue. the low stress level can be found in the intact media,
However, only at the side where the fibrous cap is which itself carries the main load. However, once the
thinner a pronounced cleavage of the arterial wall media gets more damaged a redistribution of the stress
between the intima and the media developed, i.e. a from the media to the adventitia will occur. Finally, the
dissection propagates along the lamina elastica inter- computational analysis shows that the lipid pool is
na. Interestingly, the tissue failure does not propagate under a small compressive hydrostatic stress (not
into the lipid pool, which would lead to an extrusion shown in Fig. 5), which allows the extrusion of the fat
of the lipid pool into the lumen, the typically into the lumen and a partial redistribution into the
observed type of revascularization during balloon wall, as proposed by Toussaint et al.63 for this type of
angioplasty of fatty (yellow) plaques.63 It needs to be lesion (here it is assumed that a channel between the
emphasized, that, once developed, the tissue dissec- lipid pool and the lumen exists). This mechanism is
tion propagates very much under mode I condition, not considered in the present computation, but it
and peeling seems to be the underlying dissection contributes to the enlargement of the lumen of fatty
mechanism. Hence, in that sense the used cohesive (yellow) plaques.63
material parameters, which were obtained from Surprisingly, after intimal fissuring occurs, the stress
peeling experiments of human aortic medias,58 are level in the overall tissue remains relatively moderate;
appropriate. it does not much exceed 300.0 kPa (apart from some
The predicted distributions of the maximum prin- small regions at the non-dissected shoulder of the
cipal (Cauchy) stresses within the atherosclerotic lesion intima). Hence, it is suggested that the propagation of
at balloon diameters db = 3.52 mm and 4.42 mm are the dissection effectively avoids stress-based damage in
illustrated in Fig. 5. The stresses are given in kPa and other regions of the lesion, as it has been confirmed by
are shown for the intima, media and the adventitia. At our previous work.16
the inflation state, corresponding to db = 3.52 mm,
intimal fissuring starts at that side of the shoulder
CONCLUSION

Plaque fissuring and/or dissections of arterial tissue

are frequently observed in clinical practice (see the
reviews by Richardson50,51), and these types of failure
mechanisms are known to be also caused during (bal-
loon) angioplasty interventions.9,41,47 Plaque fissuring
and/or dissections represent major contributions to the
lumen enlargement caused by balloon angioplasty.41
Plaque fissuring and/or dissections have been shown to
be an important predictor of the clinical outcome after
balloon angioplasty intervention.4,38 Interestingly,
only a few studies investigate the underlying biome-
chanics of tissue failure in atherosclerotic plaques.
In the present work we proposed physical and fi-
nite element models in 3D, which are generally use-
able to trace fissuring and/or dissection in
atherosclerotic plaques during balloon angioplasty
FIGURE 4. Deformed slice of the atherosclerotic lesion at a
lumen diameter of db = 4.42 mm. Partition of unity finite ele- interventions. The underlying continuum mechanical
ments are removed, for the sake of clarity. framework considered the atherosclerotic plaque as a
720 T. C. GASSER AND G. A. HOLZAPFEL

FIGURE 5. Distribution of maximum principal (Cauchy) stresses in the atherosclerotic lesion of a human external iliac artery
during balloon angioplasty (type V lesion according to Stary59). Stresses are given for the intima, media and the adventitia at
balloon diameters db = 3.52 mm and 4.42 mm.

highly deformable, locally anisotropic and heteroge- based on a single-field variational formulation. The
neous body, which was assumed to be stress free in implementation of the related linearized problem
the unloaded (initial) configuration. Since the wall represents a robust and efficient numerical basis for
stresses due to artery (balloon) expansion are much the study of the propagation of tissue failure in 3D.
higher than those due to physiological loading, we We modeled and investigated the evolution of plaque
neglected dynamic effects inherent to the cardiovas- fissuring and dissection in an atherosclerotic-prone
cular system, and claim that plaque fissuring and human external iliac artery (type V lesion according
dissection caused during angioplasty intervention is to Stary59) during balloon angioplasty intervention,
most likely to be unaffected by blood-flow induced for which the balloon/artery interaction was solved as
forces. Tissue failure was captured by means of the a 3D contact problem. Finally, the lesion’s geometry
strong discontinuity kinematics, which characterize a was reconstructed by means of hrMRI and a specif-
material surface separation by a jump in the dis- ically developed software tool. Material, structural
placement field. Recent histological investigations58 and cohesive parameters of the involved arterial tis-
have highlighted the dominant role of (collagen and sue components were considered.
elastin) fiber bridging in arterial dissection, which was The performed 3D computations showed that in a
the motivation to use the theory of cohesive zones primary phase the plaque fissured at both shoulders of
rather than the theory of sharp crack tips. The failure the intima, and then stopped at the lamina elastica
hypothesis of atherosclerotic plaques was numerically interna. In a secondary phase, local dissections
realized by means of PUFEM and IEM such that between the intima and the media developed at the
plaque fissuring and dissection, or a combination of shoulder, where the fibrous cap is thinner. Interest-
them can be studied. Both numerical methods were ingly, the dissection effectively reduced the stresses in
 Modeling Plaque Fissuring and Dissection during Balloon Angioplasty Intervention 721

the stenosis and avoided continuous damage in the mechanism rather than rupture of collagen fibers. To
lesion. Hence, the mechanical trauma due to balloon some extend this observation has been justified by our
angioplasty seems to remain localized at the dissection experimental dissection studies.58
site rather than spread out all over the lesion. It is known that beyond a certain loading point, the
The used numerical concept to describe tissue failure arterial widening becomes permanent due to balloon-
in a phenomenological way was based on cohesive induced overstretching, see the related experimental
zones, which is known to be the proper concept to seminal study on dilated cadaveric arteries by Cas-
handle strain softening problems. In particular, the taneda-Zuniga et al.6 In particular, during dilation
numerical concept does not suffer from mesh-sensitivity remnant non-diseased tissues in lesions are loaded
inherent to several alternative numerical approaches.15 beyond the elastic (physiological) limit, and the over-
However, a crucial requirement for obtaining mesh- stretch contributes to the lumen enlargement, a
independent results is that the finite element size needs mechanism suggested by Hjemdahl-Monsen et al.23
to be smaller than the characteristic length of the Some of our earlier constitutive models13 assumed
cohesive zone, which can be estimated from the cohe- tissue damage due to accumulation of plastic defor-
sive law.22 Consequently, the tissue failure under con- mation in the non-collagenous matrix. Hence, perma-
sideration requires such a fine computational grid that nent deformations of the tissue components and the
a large-scale analysis would have been computationally associated changes of the mechanical environment
very expensive. Certainly, h refinement at locations during balloon angioplasty can be addressed.17,26,30,34
where dissection regions appear could significantly It seems, however, that mainly fibrotic stenosis are
reduce the computational costs. effected by this mechanism.63 Since we studied a lipid-
Within this work we assumed that the load-free filled plaque we have, consequently, neglected the
configuration of the lesion was free of stress, which is inelastic effect due to permanent tissue deformation.
most probably not the case. Residual stresses in the Apart from all these limitations the proposed physical
load-free configuration have a big influence on the local and numerical model should be regarded as a further
stress field, and a moderate on the global response of step towards computational aided angioplasty.
healthy arteries.20,27,28 It needs to be emphasized
that the complex geometry and heterogeneity of ath- ACKNOWLEDGMENTS
erosclerotic arteries cause also complex three-dimen-
sional residual stress fields,31 which cannot easily be We would like to thank Martin Auer and Dimitrios
incorporated, e.g., by ’closing an opened-up configu- E. Kiousis for their helpful support to generate the
ration’, as it is usually performed for healthy arteries. finite element grids. Financial support for this research
Certain assumptions were required to provide was partly provided by the Austrian Science Founda-
material and structural parameters for the present tion under START-Award Y74-TEC. This support is
analysis since there is (still) a considerable lack of gratefully acknowledged.
comprehensive experimental data in the literature.
Although a small number of studies document exper-
imental data regarding ultimate tensile stresses and REFERENCES
strains,32,40 no data are known regarding the softening
region (fracture energy and/or the shape of the cohe- 1
Agarwal, R., U. Kaul, V. Dev, S. Sharma, and P. Venug-
sive law). Note that balloon angioplasty locally exceeds opal. The morphology of coronary arterial dissection
the deformation at which the maximum cohesive ten- occurring subsequent to angioplasty and its influence on
sile strength occurs, and predictions then depend on acute complications. Int. J. Cardiol. 31:59–64, 1991.
2
Armero, F. and K. Garikipati. An analysis of strong dis-
the used data describing the softening regions. continuities in multiplicative finite strain plasticity and their
A fundamental and in-depth understanding of fail- relation with the numerical simulation of strain localization
ure mechanisms in soft biological tissues is still miss- in solids. Int. J. Solids Struct. 33:2863–2885, 1996.
3
ing. While dissections of arterial tissues have been Auer, M., R. Stollberger, P. Regitnig, F. Ebner, and G. A.
investigated in detail in previous works,5,42,52,58 the Holzapfel. 3-D reconstruction of tissue components for
atherosclerotic human arteries based on high-resolution
failure mechanisms of plaque fissuring are unclear. MRI. IEEE Trans. Med. Imaging 25:345–357, 2006.
Within this work we assumed a Rankine failure crite- 4
Black, A. J., D. L. Namay, A. L. Niederman, N. J. Lembo,
rion, in which the occurring maximum principal stress G. S. Roubin, J. S. Douglas Jr., and S. B. King III. Tear or
was compared with the tissue-specific stress limit. dissection after coronary angioplasty Morphologic corre-
However, recent failure studies indicated that shear lates of an ischemic complication. Circulation 79:1035–
1042, 1989.
stress in the non-collagenous matrix might trigger 5
Carson, M. W. and M. R. Roach. The strength of the
tissue failure.21 To be more specific, pull-out and aortic media and its role in the propagation of aortic
peeling of collagen fibers might be the dominate failure dissection. J. Biomech. 23:579–588, 1990.
722 T. C. GASSER AND G. A. HOLZAPFEL
6 22
Castaneda-Zuniga, W. R., A. Formanek, M. Tadavarthy, Hillerborg, A., M. Modeer, and P. E. Petersson. Analysis
Z. Vlodaver, J. E. Edwards, C. Zollikofer, and K. Amplatz. of crack formation and crack growth in concrete by means
The mechanism of balloon angioplasty. Radiology of fracture mechanics and finite elements. Cement Concr.
135:565–571, 1980. Res. 6:773–782, 1976.
7 23
Clark, J. M. and S. Glagov. Transmural organization of Hjemdahl-Monsen, C. E., J. A. Ambrose, S. Borrico, M.
the arterial media: The lamellar unit revisited. Arterio- Cohen, W. Sherman, D. Alexopoulos, R. Gorlin, and V.
sclerosis 5:19–34, 1985. Fuster. Angiographic patterns if balloon inflation during
8
Coleman, B. D. and W. Noll. The thermodynamics of percutaneous transluminal coronary angioplasty: Role of
elastic materials with heat conduction and viscosity. Arch. pressure-diameter curves in studying distensibility and
Rat. Mech. Anal. 13:167–178, 1963. elasticity of the stenotic lesion and the mechanism of dila-
9
Coulden, R. A., H. Moss, M. J. Graves, D. J. Lomas, tion. J. Am. Coll. Cardiol. 16:569–575, 1990.
24
D. S. Appleton, and P. L. Weissberg. High resolution Holzapfel, G. A. Nonlinear Solid Mechanics. A Contin-
magnetic resonance imaging of atherosclerosis and the re- uum Approach for Engineering. John Wiley & Sons,
sponse to balloon angioplasty. Heart 83:188–191, 2000. Chichester, 2000.
10 25
Cripps, T. R., J. M. Morgan, and A. F. Rickards. Outcome Holzapfel, G. A., and T. C. Gasser. A viscoelastic model
of extensive coronary artery dissection during coronary for fiber-reinforced composites at finite strains: Continuum
angioplasty. Br. Heart J. 66:3–6, 1991. basis, computational aspects and applications. Comput.
11
Detre, K. M., D. R. Holmes Jr., R. Holubkov, M. J. Methods Appl. Mech. Eng. 190:4379–4403, 2001.
26
Cowley, M. G. Bourassa, D. P. Faxon, G. R. Dorros, L. G. Holzapfel, G. A. and T. C. Gasser. Computational stress–
Bentivoglio, K. M. Kent, and R. K. Myler. Incidence and deformation analysis of arterial walls including high-
consequences of periprocedural occlusion: The 1985-1986 pressure response. Int. J. Cardiol. 116:78–85, 2007.
27
National Heart, Lung, and Blood Institute Percutaneous Holzapfel, G. A., T. C. Gasser, and R. W. Ogden. A new
Transluminal Coronary Angioplasty Registry. Circulation constitutive framework for arterial wall mechanics and a
82:739–750, 1990. comparative study of material models. J. Elasticity 61:1–48,
12
Finlay, H. M., P. Whittaker, and P. B. Canham. Collagen 2000.
28
organization in the branching region of human brain Holzapfel, G. A., T. C. Gasser, and R. W. Ogden. Com-
arteries. Stroke 29:1595–1601, 1998. parison of a multi-layer structural model for arterial walls
13
Gasser, T. C. and G. A. Holzapfel. A rate-independent with a Fung-type model, and issues of material stability.
elastoplastic constitutive model for (biological) fiber-rein- J. Biomech. Eng. 126:264–275, 2004.
29
forced composites at finite strains: Continuum basis, algo- Holzapfel, & Ogden (Eds) Biomechanics of Soft Tissue in
rithmic formulation and finite element implementation. Cardiovascular Systems. Springer-Verlag, Wien – New York,
Comput. Mech. 29:340–360, 2002. 2003.
14 30
Gasser, T. C. and G. A. Holzapfel. Geometrically non- Holzapfel, G. A., C. A. J. Schulze-Bauer, and M. Stadler.
linear and consistently linearized embedded strong dis- Mechanics of angioplasty: Wall, balloon and stent. In
continuity models for 3D problems with an application to Casey, J. and G. Bao (eds.) Mechanics in Biology. New
the dissection analysis of soft biological tissues. Comput. York: The American Society of Mechanical Engineers
Methods Appl. Mech. Eng. 192:5059–5098, 2003. (ASME). AMD-Vol. 242/BED-Vol. 46, 2000, pp. 141–
15
Gasser, T. C. and G. A. Holzapfel. Modeling 3D crack 156.
31
propagation in unreinforced concrete using PUFEM. Holzapfel, G. A., G. Sommer, M. Auer, P. Regitnig, and
Comput. Methods Appl. Mech. Eng. 194:2859–2896, 2005. R. W. Ogden. Layer-specific 3D residual deformations of
16
Gasser T. C. and G. A. Holzapfel. Physical and numerical human aortas with non-atherosclerotic intimal thickening.
modeling of dissection propagation in arteries caused by Ann. Biomed. Eng., doi 10.1007/s10439-006-9252-z, in
balloon angioplasty. In Hamza, M. H. (ed.) Proceedings of press.
32
the 3rd IASTED International Conference on Biome- Holzapfel, G. A., G. Sommer, and P. Regitnig. Anisotropic
chanics, Anaheim. ACTA Press, 2005, pp. 229–233. mechanical properties of tissue components in human
17
Gasser T. C. and G. A. Holzapfel. Finite element modeling atherosclerotic plaques. J. Biomech. Eng. 126:657–665,
of balloon angioplasty by considering overstretch of rem- 2004.
33
nant non-diseased tissues in lesions. Comput. Mech., 2006, Holzapfel, G. A., M. Stadler, and T. C. Gasser. Changes in the
online doi 10.1007/s00466-006-0081-6. mechanical environment of stenotic arteries during interac-
18
Gasser, T. C. and G. A. Holzapfel. 3D crack propagation tion with stents: Computational assessment of parametric
in unreinforced concrete. A two-step algorithm for tracking stent design. J. Biomech. Eng. 127:166–180, 2005.
34
3D crack paths. Comput. Methods Appl. Mech. Eng. Holzapfel, G. A., M. Stadler, and C. A. J. Schulze-Bauer. A
195:5198–5219, 2006. layer-specific three-dimensional model for the simulation of
19
Gasser, T. C. and G. A. Holzapfel. Modeling the propa- balloon angioplasty using magnetic resonance imaging and
gation of arterial dissection. Eur. J. Mech. A/Solids 25:617– mechanical testing. Ann. Biomed. Eng. 30:753–767, 2002.
35
633, 2006. Holzapfel, G. A. and H. W. Weizsäcker. Biomechanical
20
Gasser, T. C., R. W. Ogden, and G. A. Holzapfel. behavior of the arterial wall and its numerical character-
Hyperelastic modelling of arterial layers with distributed ization. Comp. Biol. Med. 28:377–392, 1998.
36
collagen fibre orientations. J. R. Soc. Interface 3:15–35, Huber, M. S., J. F. Mooney, J. Madison, and M. R.
2006. Mooney. Use of a morphologic classification to predict
21
Haslach, H. W. Rupture mechanisms in circulatory system clinical outcome after dissection from coronary angio-
vascular tissue. In Abstracts of the 5th World Congress of plasty. Am. J. Cardiol. 68:467–471, 1991.
37
Biomechanics. Munich, Germany, 29 July–4 August 2006. Humphrey, J. D. Cardiovascular Solid Mechanics. Cells.
Elsevier, 2006, p. 274. Tissues, and Organs. Springer-Verlag, New York, 2002.
 Modeling Plaque Fissuring and Dissection during Balloon Angioplasty Intervention 723
38 53
Jain, S. P., A. Jain, T. J. Collins, S. R. Ramee, and C. J. Rose, J. H., J. Ferrante, and J. R. Smith. Universal binding
White. Predictors of restenosis: A morphometric and energy curves for metals and bimetallic interfaces. Phys.
quantitative evaluation by intravascular ultrasound. Am. Rev. Lett. 47:675–678, 1981.
54
Heart J. 128:664–673, 1994. Rose, J. H., J. R. Smith, and J. Ferrante. Universal features
39
Kachanov, L. M. Introduction to Continuum Damage of bonding in metals. Phys. Rev. B 28:1835–1845, 1983.
55
Mechanics. Martinus Nijhoff Publishers. Dordrecht, The Saber, R. S., W. D. Edwards, K. R. Bailey, T. W.
Netherlands, 1986. McGovern, R. S. Schwartz, and D. R. Holmes Jr. Coro-
40
Lendon, C. L., M. J. Davies, G. V. R. Born, and P. D. nary embolization after balloon angioplasty or thrombo-
Richardson. Atherosclerotic plaque caps are locally weak- lytic therapy: An autopsy study of 32 cases. J. Am. Coll.
ened when macrophages density is increased. Atheroscle- Cardiol. 22:1283–1288, 1993.
56
rosis 87:87–90, 1991. Salunke, N. V. and L. D. T. Topoleski. Biomechanics of ath-
41
Losordo, D. W., K. Rosenfield, A. Pieczek, K. Baker, erosclerotic plaque. Crit. Rev. Biomed. Eng. 25:243–285, 1997.
57
M. Harding, and J. M. Isner. How does angioplasty work? Schöberl, J. NETGEN – A 3D Tetrahedral Mesh Gen-
Serial analysis of human iliac arteries using intravascular erator – Version 4.2. University Linz, Austria, 2002.
58
ultrasound. Circulation 86:1845–1858, 1992. Sommer G., T. C. Gasser, P. Regitnig, M. Auer, and G. A.
42
MacLean, N. F., N. L. Dudek, and M. R. Roach. The role Holzapfel. Dissection of the human aortic media: An
of radial elastic properties in the development of aortic experimental study. J. Biomech. Eng., in press.
59
dissections. J. Vasc. Surg. 29:703–710, 1999. Stary, H. C. Atlas of Atherosclerosis: Progression and
43
Ogden, R. W. Non-linear Elastic Deformations. Dover, Regression (2nd ed.). The Parthenon Publishing Group
New York, 1997. Limited. Boca Raton London, New York, Washington, D.C.,
44
Oliver, J. Modelling strong discontinuities in solid 2003
60
mechanics via strain softening constitutive equations Part Taylor, R. L. FEAP – A Finite Element Analysis Pro-
1: Fundamentals. Int. J. Numer. Methods Eng. 39:3575– gram. Version 7.4 User Manual. University of California at
3600, 1996. Berkeley, Berkeley, California, 2002.
45 61
Ortiz, M. and A. Pandolfi. Finite-deformation irreversible Tobis, J. M., J. Mallery, D. Mahon, K. Lehmann,
cohesive elements for three-dimensional crack-propagation P. Zalesky, J. Griffith, J. Gessert, M. Moriuchi, M. McRae,
analysis. Int. J. Numer. Methods Eng. 44:1267–1282, 1999. and M. L. Dwyer, et al.. Intravascular ultrasound imaging
46
Patel, D. J., J. S. Janicki, and T. E. Carew. Static aniso- of human coronary arteries in vivo. Analysis of tissue
tropic elastic properties of the aorta in living dogs. Circ. characterizations with comparison to in vitro histological
Res. 25:765–779, 1969. specimens. Circulation 83:913–926, 1991.
47 62
Popma, J. J., M. B. Leon, G. S. Mintz, K. M. Kent, L. F. Toussaint, J. F., J. F. Southern, V. Fuster, and H. L.
Satler, T. J. Garrand, and A. D. Pichard. Results of cor- Kantor. Water diffusion properties of human atheroscle-
onary angioplasty using the transluminal extraction cath- rosis and thrombosis measured by pulse field gradient
eter. Am. J. Cardiol. 70:1526–1532, 1992. nuclear magnetic resonance. Arterioscler. Thromb. Vasc.
48
Purslow, P. P. Positional variations in fracture toughness, Biol. 17:542–546, 1997.
63
stiffness and strength of descending thoracic pig aorta. Toussaint, J. F., J. F. Southern, H. L. Kantor, I. K. Jang,
J. Biomech. 16:947–953, 1983. and V. Fuster. Behavior of atherosclerotic plaque compo-
49
Rhodin, J. A. G. Architecture of the vessel wall. In Bohr, nents after in vitro angioplasty and atherectomy studied by
D. F. A. D. Somlyo, & H. V. Sparks (eds.) Handbook of high field MR imaging. Magn. Reson. Imaging 16:175–183,
Physiology, The Cardiovascular System. 2 American 1998.
64
Physiological Society, Bethesda, Maryland: 1–31, 1980. Weizsäcker, H. W. and J. G. Pinto. Isotropy and anisot-
50
Richardson, P. D.. Biomechanics of plaque rupture: ropy of the arterial wall. J. Biomech. 21:477–487, 1988.
65
Progress, problems, and new frontiers. Ann. Biomed. Eng. Wolinsky, H. and S. Glagov. A lamellar unit of aortic
30:524–536, 2002. medial structure and function in mammals. Circ. Res.
51
Richardson, P. D. Mechanical properties of atherosclerotic 20:90–111, 1967.
66
tissues. In Holzapfel, G. A. & R. W. Ogden (eds.) Mechanics Wriggers, P. Computational Contact Mechanics. John
of Biological Tissue. Springer-Verlag, Heidelberg: 207–223, Wiley & Sons, Chichester, 2002.
67
2006. Zhou, J. and Y. C. Fung. The degree of nonlinearity and
52
Roach, M. R., J. C. He, and R. G. Kratky. Tear propa- anisotropy of blood vessel elasticity. Proc. Natl. Acad. Sci.
gation in isolated, pressurized porcine thoracic aortas. Can. USA 94:14255–14260, 1997.
J. Cardiol. 15:569–575, 1999.

View publication stats