A Review On The Biomechanics of Coronary Arteries

Download as pdf or txt
Download as pdf or txt
You are on page 1of 62

International Journal of Engineering Science 147 (2020) 103201

Contents lists available at ScienceDirect

International Journal of Engineering Science


journal homepage: www.elsevier.com/locate/ijengsci

A review on the biomechanics of coronary arteries


Harry J. Carpenter a, Alireza Gholipour a, Mergen H. Ghayesh a,∗,
Anthony C. Zander a, Peter J. Psaltis b,c
a
School of Mechanical Engineering, University of Adelaide, Adelaide, South Australia 5005, Australia
b
Vascular Research Centre, Lifelong Health Theme, South Australian Health and Medical Research Institute (SAHMRI), Adelaide, South
Australia 5000, Australia
c
Adelaide Medical School, University of Adelaide, Adelaide, South Australia 5005, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Globally, cardiovascular diseases are seen as one of the largest economic burdens on soci-
Received 19 September 2019 ety and the single largest cause of death, with atherosclerosis the leading cause of myocar-
Accepted 5 November 2019
dial infarction (heart attack). Due to the complex interactions in the coronary vasculature,
medical imaging technology is unable to recognise correlations between artery and fluid
Keywords: mechanics and disease initiation and progression, hence, biomechanical analysis of coro-
Biomechanics nary arteries in patient-specific models is necessary to provide critical information in clin-
Coronary artery ical settings; large variability in modelling approaches, parameters and results still, how-
Artery-blood interaction ever, hamper accurate and reliable model development. This review aims to assist in filling
Atherosclerosis that gap by presenting an overview of research efforts to date, from both theoretical and
Cardiovascular disease experimental perspectives, to assist in addressing the challenge of developing a reliable
Fluid-Structure Interaction
and accurate biomechanical model of human coronary arteries. Studies have been cate-
Nonlinear dynamics
gorised primarily on their approach, either purely theoretical, purely experimental/clinical
or a combined theoretical and experimental/clinical approach as well as then divided into
structural, fluid, and fluid–structure interaction (FSI) analysis. From research efforts to date
it is clear that the development of FSI models to incorporate the effects of shear-thinning,
non-Newtonian flows in viscoelastic, realistic artery and plaque morphologies developed
from in vivo, high resolution imaging is critical for accurate determination of disease ini-
tiation/progression, including atherosclerosis, plaque formation, and failure mechanisms.
The inclusion of micro-constituents such as micro-calcification, endothelial cell layer, col-
lagen cross-linking, vascular smooth muscle cell contractility and constitutive blood equa-
tions also affects stress magnitude, distribution and failure mechanisms and should be ade-
quately accounted for. A set of appendices are also provided where studies are summarised
and assessed based on their approach to modelling and are categorised by considering
the imaging modality used, methodology, material and blood properties, type of coronary
artery, approach to the study (theoretical, experimental, in vitro & in vivo), number of
patients/specimens and a general description. It is hoped this review will assist in further-
ing the field of coronary artery biomechanics and contribute to developing accurate and
reliable patient-specific models capable of improving our understanding of cardiovascular


Corresponding author.
E-mail address: [email protected] (M.H. Ghayesh).

https://doi.org/10.1016/j.ijengsci.2019.103201
0020-7225/© 2019 Elsevier Ltd. All rights reserved.
2 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

biomechanics and hence, the initiation/progression of related diseases to address the grow-
ing global morbidity, mortality and economic challenge of cardiovascular disease and my-
ocardial infarction.
© 2019 Elsevier Ltd. All rights reserved.

1. Introduction

Combined, stroke and ischaemic heart disease were responsible for 15.2 million deaths in 2016 accounting for almost
27% of total annual deaths according to the World Health Organisation (2018). They are now outweighing the impacts of
communicable diseases in both developing and developed countries (Bloom et al., 2012). Cardiovascular disease, including
ischaemic heart disease, is now seen as one of the largest economic burdens on global society (American Heart Associa-
tion 2017; American Heart Association 2018; Calvert, 2014; Gheorghe et al., 2018; Leal, Luengo-Fernández, Gray, Petersen, &
Rayner, 2006), with annual medical costs and lost productivity estimates predicted to reach USD$818 billion and USD$275
billion respectively by 2030 in the US, UK and Europe alone (Giedrimiene & King, 2017). Of patients presenting with myocar-
dial infarction, atherosclerotic (Tricerri, Dedè, Gambaruto, Quarteroni, & Sequeira, 2016) plaque disruption leading to artery
(Anand, Kwack, & Masud, 2013; Cyron & Humphrey, 2014) occlusion is often the primary cause (Calvert, 2014). To restore
coronary flow, percutaneous coronary intervention is often used to install a stent guided by in vivo imaging, however stents
are still susceptible to restenosis and thrombosis, especially in artery bifurcations (Antoniadis et al., 2015). Whilst current
medical imaging technology is improving at a rapid rate, the exact nature of plaque constituents, their effect on rupture and
its prediction eludes researchers in clinical settings; hence, the biomechanical (Doraiswamy, Criscione, & Srinivasa, 2016;
Goda & Ganghoffer, 2015; Goda, Rahouadj, Ganghoffer, Kerdjoudj, & Siad, 2016; Joshi & Walton, 2013; Rubin & Solav, 2016;
Shirazi & Ayatollahi, 2014; Švihlová, Hron, Málek, Rajagopal, & Rajagopal, 2016a; Švihlová, Hron, Málek, Rajagopal, & Ra-
jagopal, 2017) modelling of coronary arteries has the potential to contribute to not just the planning and assessment of
interventional procedures, but also in predicting the causes which lead to myocardial infarction and the leading cause of
death globally.

1.1. Coronary artery histology

Coronary arteries are composed of three main layers, from outer to inner layer, known as the adventitia, media and tu-
nica intima (Richardson, 2002). The intima layer contains the endothelium and is made up of a thin endothelial cell layer
which assists in regulating vascular homoeostasis, the formation and dissolution of thrombus, leucocyte adhesion and mi-
gration and prevents the coagulation of blood along its surface making it critical in the initiation and progression of disease
(Edwards, Bouchier, Haslett, & Chilvers, 1991; Fountoulakis, Oikonomou, Lazaros, & Tousoulis, 2018); it is the location of
major changes caused by atherosclerosis (Akyildiz, Speelman, & Gijsen, 2014). Endothelial cells are activated by wall shear
stresses imposed by flowing blood and align themselves with the flow direction under laminar wall shear stress, supporting
vasodilation and preventing coagulation (Jennette & Stone, 2014). The membrane supporting this layer is a combination of
collagen, laminin, fibronectin and extracellular matrix compounds and is connected via an elastic membrane to the media
layer which is primarily made up of smooth muscle cells concentrically layered and reinforced by a matrix of elastin and col-
lagen fibres known as the extracellular matrix. These smooth muscle cells allow for the active contraction of the media layer
during diastole. The adventitia, connected by the external elastic lamina membrane to the media, is made up of primarily
connective tissues such as thin collagen and elastic fibres, fibroblasts and macrophages with the vasa vasorum, a network
of small blood vessels suppling the walls of the larger coronary arteries, also seen in this layer (Tsioufis, Mantzouranis,
Kalos, Konstantinidis, & Tousoulis, 2018). Collagen fibres are seen as wavy when under no load however straighten during
fibre recruitment which results in nonlinear stress-strain relations; further information on nonlinear strains can be found in
Refs. Farokhi and Ghayesh (2015a, 2015b), Farokhi, Ghayesh, and Amabili (2013b), Farokhi, Ghayesh, Gholipour, and Hussain
(2017a), Ghayesh (2018d), Ghayesh and Farokhi (2015b), Ghayesh and Farokhi (2018), Ghayesh, Amabili, and Farokhi (2013c),
Ghayesh, Farokhi, and Amabili (2013e), Ghayesh, Farokhi, and Gholipour (2017b); Ghayesh, Farokhi, and Gholipour (2017a).
With coronary arteries intertwined with the epicardium and myocardium layers of the heart, they are subject to myocar-
dial contraction of muscle fibres made up of myocytes and their primary contraction unit the sarcomere (Edwards et al.,
1991). This complex morphology and interaction on both macro and micro levels makes the study of disease progression
and biomechanical analysis of in vivo conditions difficult.

1.2. Atherosclerosis and myocardial infarction initiators

While the epidemiology of atherosclerosis includes many lifestyle factors, the pathophysiological aspects of atheroscle-
rosis are generally caused by three factors: the mechanical obstruction of flow, thrombosis due to plaque rupture and
aneurysm formation due to weakening of the artery walls (Hanke, Lenz, & Finking, 2001; Kumar, Abbas, Fausto, & Aster,
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 3

2014). As an inflammatory disease (Athanasiou, Fotiadis, & Michalis, 2017; Calvert, 2014; Ross, 1999), increased plasma con-
centrations of primarily low-density lipoprotein (LDL) and cholesterol affect vasoreactivity, leading to increased permeability
of the intima layer, increased adhesiveness to leucocytes (white blood cells) and platelets, causing coagulation (Kumar et al.,
2014; Ross, 1999). Along with endothelial cell injury, this coagulation induces growth factors and smooth muscle cell pro-
gression between the media and intima layers, accelerating atheroma, lesion formation (Ohayon et al., 2017; Tsioufis et al.,
2018) and collagen cell degradation. Endothelial cell dysfunction caused by turbulent flow and oscillating wall shear stresses
leads to further promotion of coagulation and artery vasoconstriction.
All aforementioned processes have been emphasised as important in clinical settings in the development of coronary le-
sions, however, the biomechanical analysis of coronary arteries has begun to highlight the importance of artery and plaque
geometry, tissue properties, principal stresses, plaque morphology (including but not limited to calcifications, fibrous cap
thicknesses and necrotic cores), and fluid induced shear and residual stresses. Plaques prone to rupture are termed vulner-
able (Nguyen & Levy, 2010), however, clinically predicting rupture prone plaques is difficult as it is dependant on multiple
varying parameters. Rupture results in thrombosis formation and coronary occlusion, leading to myocardial ischaemia and
potentially necrosis which could be either transmural (generally constrained to the region of the occluded artery) or sub-
endocardial; of patients who develop myocardial infarction or sudden cardiac death without plaque rupture, erosion is seen
as one of the major factors (Braunwald, 2013). Plaque erosion is often associated with the lack of endothelial layer and
intimal thickening and is also important in coronary thrombosis development (Arbustini et al., 1999).
While a less prominent cause of myocardial infarction, coronary artery aneurysms (Abou Sherif, Ozden Tok, Taşköylü,
Goktekin, & Kilic, 2017) and dissection (Saw, Mancini, & Humphries, 2016) are also being diagnosed more regularly, poten-
tially due to the increasing accuracy and accessibility through coronary intervention and the use of angiography. Atheroscle-
rosis was suggested as a potential initiator of coronary artery aneurysm, disturbing the intima and media layers and causing
degeneration of the elastic muscular tissues similar to connective tissue disorders, weakening the artery wall and leading to
the inability to cope with intramural pressure, hence, predisposing the artery to dilation and aneurysm (Abou Sherif et al.,
2017). Spontaneous coronary artery dissection (SCAD), which does not include iatrogenic dissection, occurs most promi-
nently in women (accounting for over 90% of cases) and causes 20-25% of myocardial infarction cases in women (Saw et al.,
2016). Occurring either within the artery layers or between them, SCAD is often associated with increased pressure and cir-
culatory shear stress coupled with increased myocardial contractility and vasospasm which can affect sites predisposed to
disease such as at endothelial injury or atherosclerosis. With high SCAD recurrence and high resolution imaging necessary
for assessment and clinical intervention, improving the understanding of the initiation and progression of such events will
prove useful in clinical settings.

1.3. Medical imaging modalities

Coronary arteries pose the unique challenge of being both difficult to reach and under constant motion from the cardiac
muscle and pulsatile nature of blood flow. This makes studies on human patients difficult and less common than other ves-
sels such as the carotid artery in the neck. A number of imaging technologies and tests to investigate cardiovascular disease
present a range of useful data. Previous reviews and literature (Athanasiou et al., 2017; Cardoso & Weinbaum, 2014; Jaffer
& Verjans, 2014; Kim et al., 2001; Suter et al., 2011; Taki, Kermani, Ranjbarnavazi, & Pourmodheji, 2017) have examined in
detail the uses, abilities and limitations of various imaging modalities, however a brief overview of some of their capabil-
ities, limitations and applications to biomechanical analysis of coronary arteries is provided here. Coronary angiography is
perhaps the most common catheterisation procedure used for assessing the feasibility of interventional procedures such as
bypass or angioplasty, however it does not allow for recognition of atherosclerosis itself, merely significant lumen diam-
eter changes throughout the vasculature. Non-invasive Magnetic Resonance Angiography (MRA) is more prominently used
to evaluate the abdominal or thoracic aorta (Kim et al., 2001) as well as arteries in the brain or neck due to lower spatial
resolution in coronary settings. Although larger stenosis levels are easier to visually assess, biomechanical analysis of coro-
nary arteries have suggested that stenosis of less than half the lumen diameter are a significant rupture risk, with a 45-50%
stenosis range more at risk than higher ranges (Falk, Shah, & Fuster, 1995; Sorof, 2004), although this is thought to be due
to these plaques being more common in clinical settings.
When assessing plaque and artery morphology, Intravascular Ultrasound (IVUS) along with Optical Coherence Tomog-
raphy (OCT) are two of the most commonly used imaging modalities. While both catheter based probes, OCT allows for
higher resolution images to be taken (less than 10 μm resolution (Suter et al., 2011)), however the use of light limits the
penetration depth compared to IVUS, preventing the assessment of deeper artery and plaque layers. Virtual Histology In-
travascular Ultrasound (VH-IVUS) builds on the grey-scale limitations of IVUS to provide more detailed imaging of plaque
composition by assessing frequency responses to classify basic constituents (Garcìa-Garcìa, Gogas, Serruys, & Bruining, 2011).
Non-invasive techniques such as Computer Tomography Angiography (CTA) produce resolution to 0.5 mm accuracy in clin-
ical settings with short breath holds for the patient, made possible through the injection of an iodine rich contrast, and
both two and three-dimensional imaging is possible through CTA. With many imaging technologies providing complemen-
tary information to each other, multimodal imaging systems that combine two or more technologies may provide a more
comprehensive assessment of coronary pathology, however, the complexity of the cardiovascular system cannot be assessed
purely via visual inspection; the biomechanical modelling of coronary arteries can complement medical imaging and provide
in depth analysis of artery stiffness, stresses or plaque crack initiation; factors medical imaging cannot assess.
4 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

1.4. Biomechanical modelling of coronary arteries

The biomechanical properties of coronary arteries and atherosclerotic plaques (Asgharzadeh Shirazi & Ayatollahi, 2014;
Doraiswamy et al., 2016; Joshi & Walton, 2013; Rubin & Solav, 2016; Švihlová et al., 2017; Švihlová, Hron, Málek, Rajagopal,
& Rajagopal, 2016b) have not been accurately determined, with considerable variation reported amongst studies to date
(Akyildiz et al., 2014); potential factors include variations in the testing environment (in vivo/in vitro/ex vivo), method
(uniaxial/biaxial tensile, inflation), preconditioning methods, time after excision, ambient and material temperature, and
tissue integrity. Constitutive equations to adequately describe soft tissue response using these properties, such as collagen
and elastin, through nonlinear continuum mechanics approaches (Holzapfel, 2001; Holzapfel & Ogden, 2009; Holzapfel &
Ogden, 2014) were developed to predict isotropic/anisotropic (Mazloum & Sevostianov, 2018; Ramírez-Torres et al., 2017;
Shariff, 2017; Wu, Li, Chen, Kang, & Müller, 2018a; Zou & He, 2018) elastic or even finite viscoelastic (Bakhshi Khaniki &
Hosseini-Hashemi, 2017; Ebrahimi & Barati, 2016; Farokhi, Ghayesh, Gholipour, & Tavallaeinejad, 2017b; Ghayesh, 2018c;
Liu, Tang, Yu, & Pipes, 2018; Seyedkavoosi et al., 2017) behaviour, although viscoelastic response has not often been in-
vestigated; see Refs. Ghayesh (2018b, 2019c, 2019d, 2019e, 2019f, 2019g, 2019h), Ghayesh and Moradian (2011), Ghayesh,
Amabili, and Farokhi (2013b) for further information on viscoelastic modelling. The microscopic level properties of liv-
ing tissue require consideration of the cellular interactions, non-homogeneous composition and tissue adaptation whilst
macroscopic level modelling of the biomechanical behaviours has often been carried out using various material models in-
cluding the Mooney-Rivilin, neoHookean, Odgen, and Holzapfel hyperelastic models (Holzapfel, Gasser, & Stadler, 2002) or
fibre-reinforced composite theories to relate shear stress (as a cause of plaque formation) and normal stresses (as the main
cause of aneurysm) (Tricerri et al., 2016) to coronary artery deformation; this makes the development of accurate theoreti-
cal models challenging. With human blood consisting of suspended elements (such as erythrocytes, leucocytes and platelets)
in plasma (Anand et al., 2013), the nonlinear shear-thinning and deformation-dependant properties coupled with complex
flow patterns can further influence disease initiation/progression; platelet function in particular is associated with cardio-
vascular and ischaemic heart disease and is a potential initiator of thrombosis (Siasos, Tsigkou, Oikonomou, Zaromitidou,
& Tousoulis, 2018) under certain physiological conditions such as increased wall shear stress. For biomechanical models of
coronary arteries to accurately and reliably predict the initiation/progression of disease a better understanding of both the
vascular biomechanics and mechanobiology are needed in tandem.

1.5. Contribution of this paper to the field of coronary artery biomechanics

As a result of the complexity discussed in Section 1.4, a number of reviews (Akyildiz et al., 2014; Chen & Kassab, 2016;
Chien, 2003; Cleary et al., 2012; Corrales-Medina, Madjid, & Musher, 2010; Govindaraju, Badruddin, Viswanathan, Ramesh,
& Badarudin, 2013; John, 2009; Katritsis et al., 2007; Li, 2018; Sun & Xu, 2014; VanEpps & Vorp, 2007) were previously
undertaken assessing individual components required for the complete biomechanical modelling of coronary arteries. The
complex, nonlinear correlations between physiological characteristics and the initiation and progression of disease, however,
require all parameters to be considered simultaneously in order to provide accurate, patient-specific information in clinical
settings. The current review paper provides an in-depth review covering all factors of human coronary biomechanics and
artery modelling from both theoretical and experimental standpoints and aims to present a comprehensive overview of re-
search efforts to date to assist in addressing the challenge of developing an accurate and reliable biomechanical model of
coronary arteries. Although medical imaging techniques such as IVUS, OCT and MRI are advancing rapidly, they are unable
to provide the necessary information to accurately assess the coronary vasculature; something biomechanical modelling will
enable by developing three-dimensional, patient-specific models constructed with greater precision for use in the planning
and assessment of coronary interventions such as stenting or balloon angioplasty; eventually, these models could be used
in the assessment and prediction of mechanisms leading to myocardial infarction such as atherosclerosis and plaque rup-
ture. It is hoped this ability will assist in combating the increasing number of deaths, morbidity and economic burden from
cardiovascular diseases.

2. Literature review

This review is a comprehensive overview of research into human coronary artery biomechanics. Studies have been cat-
egorised into three major sections; purely theoretical, purely experimental/clinical and a combination of theoretical and
experimental/clinical investigations, as outlined in Fig. 1. The theoretical section is further divided into structural, fluid, and
fluid–structure-interaction (FSI) (El Baroudi, Razafimahéry, & Rakotomanana, 2014) analysis and covers studies modelling
simplified shapes and constituents of arteries, plaques and the heart (Karšaj, Sorić, & Humphrey, 2010). The experimental
section is divided into “in vitro” and “in vivo” investigations. In vitro studies are those carried out external to the living
body and include fabricated models of arteries or excised portions of the heart and arteries for mechanical testing, such as
tensile tests, in laboratory conditions; for the purpose of this paper, studies investigating “ex vivo” have been grouped with
in vitro as the analysis has been carried out external to physiologically accurate conditions. In vivo studies are carried out
on living patients and include the application of clinical imaging, such as those mentioned in Section 1.3, or surgical inter-
vention techniques. The combined theoretical and experimental section is also classified into in vitro and in vivo analysis
and then further divided into structural, fluid and fluid–structure specific analysis. The biomechanical models developed in
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 5

Appendix
(Secon 5)
Structural
Table 1
(Secon 2.1.1)

Theorecal Fluid
Table 2
(Secon 2.1) (Secon 2.1.2)

Fluid-Structure
Interacon Table 3
(Secon 2.1.3)
Coronary Biomechanics:

In-vitro + Ex-vivo
Study Classificaon

Table 4
(Secon 2.2.1)
Experimental/Clinical
(Secon 2.2)
In-vivo
Table 5
(Secon 2.2.2)

Structural
Table 6
(Secon 2.3.1.1)

In-vitro Fluid
Table 7
(Secon 2.3.1) (Secon 2.3.1.2)

Fluid-Structure
Interacon Table 8
Theorecal + (Secon 2.3.1.3)
Expermental/Clinical
(Secon 2.3) Structural
Table 9
(Secon 2.3.2.1)

Fluid
In-vivo Table 10
(Secon 2.3.2.2)
(Secon 2.3.2)
Fluid-Structure
Interacon Table 11
(Secon 2.3.2.3)

Fig. 1. Classification of reviewed articles on the biomechanical modelling of human coronary arteries.

these studies use inputs from experimental analysis such as geometric parameters, loading and/or boundary conditions as
well as blood characteristics and material properties for a more accurate representation (Akyildiz et al., 2014). Each study
is summarised based on imaging modality, methodology, material and blood property assumptions, artery or vessel section
analysed, number of studies and short descriptors in Appendix A. Fig. 1 outlines the classification of these studies and the
accompanying sections.

2.1. Theoretical investigations

The development of theoretical models simulated through numerical techniques provides insight into the mechanical re-
sponse of healthy and diseased arteries; this is significant for prevention, clinical intervention and on-going treatment of pa-
tients (Chen & Kassab, 2016); for more details on various numerical methods for nonlinear systems please see Refs. Barretta,
Čana d̄ija, Luciano, and de Sciarra (2018), Dehrouyeh-Semnani (2018), Farokhi and Ghayesh (2017), Farokhi and Ghayesh
(2018d), Farokhi, Ghayesh, and Amabili (2013a), Farokhi, Ghayesh, and Hussain (2016), Ghayesh (2012), Ghayesh, Kazemi-
rad, and Darabi (2011), Ghayesh, Kazemirad, and Reid (2012), Ghayesh, Yourdkhani, Balar, and Reid (2010), Jiao and Alavi
(2018), Kazemirad, Ghayesh, and Amabili (2013), Qi, Huang, Fu, Zhou, and Jiang (2018). Methods such as finite element
analysis (FEM) (Arfaoui, Trifa, Mansouri, Karoui, and Renard (2018), Fan, Luo, Yang, and Li, (2018), Freed and Einstein
(2013), Ganghoffer and Sokolowski, (2014), Khakalo, Balobanov, and Niiranen (2018), Kolpakov, Andrianov, Rakin, and Roger-
son (2018), Sevostianov, Levin, and Radi (2016), Shahverdi and Barati (2017), Trofimov, Abaimov, and Sevostianov (2018),
Trofimov, Abaimov, Akhatov, and Sevostianov (2017) have widely been used to simulate arteries and plaques due to its abil-
ity to handle complex geometries and properties; it is carried out by approximating a function over a given domain and then
discretising it into separate elements which can be modelled via simpler equations and then re-assembled and solved by
variational methods. These analyses have typically been solved through two main methods; Lagrangian and Eulerian equa-
tions, however, these methods suffer under large deformation and transient problems respectively. Coupling these methods
to form the Arbitrary Lagrange-Eulerian (ALE) formulations enables the challenges associated with each individual method
6 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

to be overcome with the conservation form of the Arbitrary Lagrange-Eulerian (ALE) equations for mass, momentum and
energy outlined respectively in Eqs. (2.1)–(2.3).
∂ ˜ ˜ ∂
(ρ J ) = J (ρ (w j − υ j )), (2.1)
∂t ∂xj
 
∂ ∂ ∂p
(ρυi˜J ) = ˜J (ρυ j (−υ j + w j )) + ˜J − + ρ bi (2.2)
∂t ∂xj ∂ xi
and
 
∂ ∂ ∂
(ρ e˜J ) = ˜J (ρ e(w j − υ j )) + ˜J ρυ j b j − (ρυ j ) (2.3)
∂t ∂xj ∂xj
expressed in terms of the density ρ , pressure p, velocity υ , body force b, specific energy e, and the Jacobian determinant
˜J, for a moving reference frame w, in each element volume; the relationship between the stress and rate of deformation
of a fluid can then be used to transform the momentum equation to the three-dimensional, compressible Navier-Stokes
equations, responsible for describing the motion of fluids, as follows:
∂ U ∂ F(U ) ∂ G(U ) ∂ H(U ) ∂ V1 (U, Ux ) ∂ V2 (U, Uy ) ∂ V3 (U, Uz ) ∂ W1 (U, Ux )
+ + + = + + +
∂t ∂x ∂y ∂z ∂x ∂x ∂x ∂y
(2.4)
∂ W2 ( U, Uy ) ∂ W3 ( U, Uz ) ∂ Q1 ( U, Ux ) ∂ Q2 ( U, Uy ) ∂ Q3 ( U, Uz )
+ + + + + ,
∂y ∂y ∂z ∂z ∂z
where U is the vector of variables conserved including density, velocity and energy and vectors F, G, H, V, W and Q
are flux vectors based on the fluid temperature, viscosity coefficients, mass flow rate, heat conductivity, and pressure.
Assumptions for geometry, mechanical and blood properties such as axisymmetric or asymmetric arteries, hyperelas-
tic (Bhattacharyya, Sarangi, & Samantaray, 2015; Chebbi, Wali, & Dammak, 2016; Fallah, Ahmadian, & Mohammadi Agh-
dam, 2017b; Galich, Slesarenko, Li, & Rudykh, 2018; Gizzi, Vasta, & Pandolfi, 2014; Heiland, Forsell, Roy, Hedin, & Gasser,
2013; Il’ichev & Fu, 2014) or viscoelastic material models (Attia & Rahman, 2018; Fallah, Ahmadian, & Aghdam, 2017b;
Ghayesh, Farokhi, & Hussain, 2016c; Hashemi, 2016; Khaniki & Hosseini-Hashemi, 2017) and both Newtonian and non-
Newtonian (Janela, Moura, & Sequeira, 2010; Perkowska, Piccolroaz, Wrobel, & Mishuris, 2017) blood (Wu, Aubry, Massoudi,
& Antaki, 2017) models were further applied whilst boundary and loading conditions varied; small size influences (Farokhi &
Ghayesh, 2018a, 2018b, 2018c; Ghayesh, 2017, 2018a, 2019a, 2019b; Ghayesh & Farokhi, 2015a; Ghayesh, Amabili, & Farokhi,
2013a; Ghayesh, Farokhi, & Alici, 2016a, 2016b; Ghayesh, Farokhi, & Amabili, 2013d, 2014; Ghayesh, Farokhi, Gholipour, &
Tavallaeinejad, 2018; Gholipour, Farokhi, & Ghayesh, 2015) have been neglected for macro-level structural elements. Key in-
vestigation parameters included the effect of plaque morphology and hemodynamic properties on wall shear stress, which
are widely recognised as major factors in the progression of atherosclerosis (Katritsis et al., 2007) as well as artery mi-
crostructure effects on damage and crack propagation. Theoretical studies are categorised into structural, fluid, and fluid–
structure-interaction analysis based on the theory and solution approach with these studies summarised in Tables 1–3 of
Appendix A (Section 5).

2.1.1. Structural analysis


Structural analyses of coronary arteries and plaques have been carried out through simplification of their morphology
as summarised in Table 1. Studies were undertaken through finite element analysis of two-dimensional models of plaques
(Akyildiz et al., 2011; Cilla, Peña, Martínez, & Kelly, 2013b; Dolla, House, & Marso, 2012; Lee, Choi, & Cho, 2017; Mohammadi
& Mequanint, 2014) to assess the effect of morphology variations in the lipid pool, necrotic core, intima stiffness and fi-
brous cap thickness. With an assumed cylindrical artery and linear elastic and isotropic material, Dolla et al. (2012) investi-
gated thin-cap fibroatheroma effects on plaque stress finding that stress and lumen diameter were proportional whereas the
thickness of the fibrous cap was inversely proportional to stress; cap thickness was suggested as the most important mor-
phological parameter for peak stress assessment (Akyildiz et al., 2011). Necrotic core size and stress were also proportional
and stress was seen to increase exponentially as the thickness of the necrotic core tended towards the thickness of the
transmural necrotic core, shown by Fig. 2. This, however, conflicted with Mohammadi and Mequanint (2014) who suggested
that plaque rupture may be a local effect at regions of stress concentration, irrespective of fibrous cap thickness or necrotic
core size, using a cylindrical, fibre-reinforced, hyperelastic and anisotropic artery model. Akyildiz et al. (2011) developed a
three-layered artery model to include the effects of the intima (developed with a simplified Neo-Hookean material model)

WNH = C (I1 − 3 ); I1 = λ2r + λ2θ + λ2z , (2.5)


and the media and adventitia layers modelled as hyperelastic and anisotropic materials through the Gasser–Odgen–Holzapfel
strain energy density model (Gasser, Ogden, & Holzapfel, 2005).
k1    
ψAniso = μ[I1 − 3] + exp k2 [κ [I1 − 3] + [1 − 3κ ][I4 − 1]]2 − 1 . (2.6)
k2
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 7

Fig. 2. Necrotic core thickness relationship with peak stress on the thin cap fibroatheroma. Each line colour corresponds to a varying combination of
parameters (Dolla et al., 2012). Reprinted with permission from Elsevier.

Fig. 3. Idealised cross section of geometric parameters studies. (a) positive remodelling, (b) negative remodelling with eccentric plaque & (c) negative
remodelling with concentric plaque (Cilla et al., 2013). Reprinted with permission from Elsevier.

In these functions Ix defined the xth invariant of the strain tensor; μ, k1 and k2 are material constants derived by fit-
ting the model to human coronary artery data; λ2r , λ2θ and λ2z are the radial, circumferential and axial principal stresses
respectively. Results showed that intima stiffness in fact also plays an essential role in cap stress; intermediate to high in-
tima stiffness resulted in a tripling in plaque cap stress when the fibrous cap thickness was reduced from 0.25 to 0.05 mm
while doubling necrotic core thickness increased cap stress by 60 and 90% respectively corresponding to results found by
Dolla et al. (2012). Fibrous cap thickness had less effect than the necrotic core thickness on maximum stress in soft in-
tima models, with necrotic core angle becoming more prominent, showing the importance of a multi-layered artery model,
however only idealised cylindrical shapes were assessed in these studies.
Artery remodelling, caused by atherosclerosis progression, is the effect of artery enlargement respective to plaque area
in order to conserve the lumen diameter. Cilla et al. (2013b) investigated this eccentricity in an incompressible, hyperelastic
and non-linear material model for positive (eccentric) and negative (cylindrical) remodelling as well as eccentric and con-
centric lipid pool sizes as shown in Fig. 3. Positive remodelling resulted in increased stress concentrations in the fibrous
cap, however as the fibrous cap thickness decreased the vulnerability of positively remodelled arteries decreased, again in
contradiction to Dolla et al. (2012). Increased lipid pool size resulted in increased maximum principal stress on the fibrous
cap, with negative remodelled arteries having a higher sensitivity to the lipid pool size. To account for variations in stress
magnitude at rupture, Lee et al. (2017) hypothesised that the stress gradient was a more important measure of vulnerability
than stress magnitude itself; based on plaque classifications this study showed that thin-cap fibroatheroma plaque types
resulted in a 679% and 1568% increase in stress and negative stress gradients respectively compared to standard fibrofatty
plaques, suggesting weaker loading capacity and more vulnerability to rupture. The inclusion of calcification in the thin-cap
fibroatheroma plaque was also investigated, however results showed only an 11% and 110% increase in stress and negative
8 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

stress gradient compared to the fibrofatty plaque; the lower stress increase compared to the case without calcification is
in contradiction to clinical findings suggesting rupture possibility is increased with calcification inclusion (Lee et al., 2017).
Furthermore the cause of plaque rupture at locations that do not contain peak radial stress could not be explained, thus
pointing to other mechanisms of rupture.

Macro/micro-calcification. A potential cause for rupture of plaques at stresses up to half their predicted strength was the
inclusion of calcification, with the calcification-fibrous-plaque interface a potential location for failure (Buffinton & Eben-
stein, 2014; Nguyen & Levy, 2010). Nguyen and Levy (2010) investigated low load decohesion between the calcification-
plaque cap interface modelling the calcification as a rigid inclusion in a remotely loaded, linear-elastic layer interacting with
a nonlinear interface to model the integrin receptor proteins bonding the calcified cells to the extracellular layer. Maximum
circumferential stress increased with calcification size and proximity to the lumen surface; cap tearing initiated from the
calcification surface except when close to the lumen surface which resulted in failure at the lumen surface itself. Nguyen and
Levy (2010) further suggested that decohesion may be brittle, hence explaining the sudden onset of myocardial infarction;
these findings were corroborated with a homogeneous and incompressible finite element model comparing Mooney-Rivilin
hyperelastic and linear orthotropic parameters (Buffinton & Ebenstein, 2014). The orthotropic model showed increased stress
correlated with increased Poisson’s ratio, calcification length and decreased thickness, highlighting the importance of accu-
rate material and geometric properties; however, geometry was considered a better predictor of high stress with arc shaped
calcifications further amplifying these effects.
While imaging of micro-calcification is possible, difficulty in detecting micro-calcification due to medical image reso-
lution could also help explain rupture at locations of non-peak artery stresses (van der Giessen et al., 2011). The effects
of varied geometry was investigated by assessing both idealised and physiologically similar micro-calcification morphology
in slightly extruded, finite element, two-dimensional artery cross sections (Cilla, Monterde, Peña, & Martínez, 2013a) and
three-dimensional (Cardoso, Kelly-Arnold, Maldonado, Laudier, & Weinbaum, 2014) models. Cilla et al. (2013a) showed that
idealised micro-calcifications in the fibrous plaque cap increased maximum principal stress up to 32% and were most in-
fluential for low fibrous cap thicknesses; micro-calcification angle and eccentricity were also important parameters in the
hyperelastic, incompressible and nonlinear model with isotropic plaque. Cardoso et al. (2014) furthered these findings by
modelling complex physiological micro-calcifications as shown in Fig. 4. With similar assumptions to Cilla et al. (2013a),
variations between thick-walled Neo-Hookean (Eq. (2.7)), Mooney–Rivilin (Eq. (2.8)) and Holzapfel (Eq. (2.9)) models were
assessed with the stress concentration factor found to exceed five when micro-calcifications were spaced closely (Fig. 4(b))
and also aligned with the plaque cap tensile stress axis. In other words, when micro-calcifications are present, background
stress only need be one fifth of the plaque rupture threshold to cause failure; this resulted in massively increased vulnera-
bility with a micro-calcification size range of 5–65 μm the most critical for rupture.
The Neo-Hookean type strain energy density function (Cardoso et al., 2014), outlined as:

ψ = D1 (−1 + J )2 + C10 (−3 + I¯1 )
   2  with J = det(F ), (2.7)
I¯1 = J −2/3 I1 , I1 = λ1 + λ2 + λ3
2 2

where C10 and D1 are constant material properties coupled with shear, μ, and the bulk modulus, K, by C10 = μ/2 and D1 =
K/2; I¯1 defines the right Cauchy–Green deformation tensors first invariant (deviatoric part), expressed in terms of J and
I1 , the deformation gradient’s third invariant and the right Cauchy Green deformation tensor’s first invariant respectively,
expressed as a summation of the squares of the first three principal stretches. Similarly, the Mooney-Rivilin strain energy
model used incorporates an extra material property (Cardoso et al., 2014), C01 , related by C10 + C01 = μ/2




ψ = C10 (−3 + I¯1 ) + C01 (−3 + I¯2 ) + D1 (−1 + J )2
   2 2  , (2.8)
I¯2 = J −4/3 I2 , I2 = λ2 λ3 + λ1 λ2 + λ3 λ1
2 2 2 2

and the Holzapfel (Holzapfel, Gasser, & Ogden, 20 0 0) strain energy model is further defined as:

   
k1    2 J2 − 1
ψ = C10 (I¯1 − 3 ) + exp k2 κ I¯1 − 3 + (1 − 3κ ) I¯4(αα ) − 1 − 1 + D1 − ln(J ) , (2.9)
2k2 a=1 2

where C10 describes the isotropic behaviour of the non-collagenous matrix portion of the artery where the anisotropic be-
haviour is taken into account by the constants k1 and k2 ; the collagen fibre orientation was described by cylindrical polar
coordinates dependant on the angle between collagen fibres found from experimental data in the literature. Despite previous
studies suggesting interfacial debonding and decohesion as the major failure mode, failure by cavitation at the calcification
poles was shown to be most prominent (Maldonado, Kelly-Arnold, Cardoso, & Weinbaum, 2013). Compressibility was also
assessed with the Neo-Hookean and Mooney-Rivilin models, showing similar results for both compressible and incompress-
ible materials. The anisotropic Holzapfel model (Holzapfel & Gasser, 2007) showed significantly lower stress concentrations
than isotropic compressible models but higher stress concentrations than isotropic incompressible models; incompressibility
resulted in stress concentration halving which further highlighted the effect of both anisotropic and compressibility proper-
ties as well as the use of the most appropriate model.
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 9

Fig. 4. Stress concentration factors (SCF) for variations in micro-calcification orientation with respect to the horizontal tensile axis. (a & c) physiological
microcalcification models. (b & d) Spherical microcalcifications (Cardoso et al., 2014). Reprinted with permission from Elsevier.

Slightly extruded cross-sections (Cilla et al., 2013a) and three-dimensional models (Cardoso et al., 2014) in the inves-
tigation of micro-calcification allowed three-dimensional parameters, such as fibre orientation, to be accounted for which
greatly influenced the maximum stress distribution. Cilla, Pena, and Martinez (2012b) included circumferential and residual
axial stresses in a Gasser–Odgen–Holzapfel (Gasser et al., 2005) hyperelastic, incompressible and nonlinear finite element
material model which resulted in three parameters being vital to determining the maximum principal stress when residual
stresses was included, namely lipid core length, fibrous cap thickness, and lipid core width. A maximum principal stress
of 385.7 kPa, always located at the fibrous cap, was found without residual stress; with axial residual stress included it
increased to 451.7 kPa and decreased to 254.4 kPa with both axial and circumferential residual stresses. For a lipid core
length of 8 mm, stenosis ratio of 66.6%, fibrous cap thickness of 0.025 mm and width of 90%, maximum principal stress was
found to triple the rupture safety threshold of 247 kPa (Loree, Grodzinsky, Park, Gibson, & Lee, 1994), highlighting the im-
portance of both geometrical factors and residual stress variations. In comparison to two-dimensional models, results from
Dolla et al. (2012) and Cilla et al. (2013b) showed lower maximum principal stress which suggested that two-dimensional
models overestimated the magnitude of maximum principal stresses compared to three-dimensional models (Ohayon, Finet,
Treyve, Rioufol, & Dubreuil, 2005). These studies omitted the pulsatile nature of blood flow and assumed static pressure
loading.

Blood pulsation and artery failure mechanisms. The physiological pulsatile nature of blood flow is an important characteristic
to include in biomechanical models of coronary arteries. Versluis, Bank, and Douglas (2006) and Pei, Wu, and Li (2013)
included pulsatile flow through a pressure pulse to assess its effect on crack location, initiation and propagation in two-
dimensional, finite element models of single layered arteries as shown in Fig. 5. Versluis et al. (2006) used a two-term
Ogden (1972) strain energy function modelling hyperelastic, incompressible and isotropic material and noted that cracks
began at the area of highest stress concentration on the lumen wall; both the number of loadings to plaque rupture and
the crack growth rate were affected by the pulsatile nature of pressure loading, with fewer cycles to rupture for increasing
pressure pulses. Pei et al. (2013) further corroborated this and showed increasing crack growth rates with the first third of
the crack length reached over 60% of the fatigue life and the final third in just 15% by adopting Paris’ Law to determine the
10 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

Fig. 5. Crack propagation in an artery cross section using the finite element method. (a) geometry, (b) von Mises stress contour and (c) meshing using
eight node quadrilateral elements with detailed mesh showing crack tip (Reprinted from Pei et al., 2013, with permission from ASME.).

dependency between the stress intensity factor and fatigue crack growth rate (Pei et al., 2013):
da
= C (
K )m , (2.10)
dN
where a and N are defined as the current crack length and number of cycles to rupture respectively, the change in stress
intensity factor over one stress cycle (one heart beat) is defined by
K and C and m are constants. This was re-arranged
and integrated at each crack growth step to define the direction of crack propagation where the angle of crack propagation
was evaluated as the angle between the crack propagation pathway and the x axis. The maximum circumferential stress
criterion and propagation angle, θ , in terms of the first two modes stress intensity factors, Ki , i = 1, 2, were defined as
(Pei et al., 2013):

KI sin θ + KII (3 cos θ − 1 ) = 0 ⎪
  ⎬
KII 3KII . (2.11)
θ = arctan  − arctan ⎪
KI2 + 8KII2 KI ⎭
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 11

Fig. 6. (a) Number of loading cycles verses crack length. (b) crack location and propagation direction around lumen with number of cycles to rupture for
manually initiated (blue) and automatically initiated (red) cracks (Reprinted from Pei et al., 2013, with permission from ASME.). (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of this article.)

Whilst cracks initiated at the region of maximum circumferential stress showed faster propagation, they did not neces-
sarily correlate to a minimum fatigue life. An increase and decrease in lipid pool size and fibrous cap thickness respectively
were seen as the major factors leading to decreased fatigue life as illustrated in Fig. 6. Cracks that initiated near the lipid
pool ends tended to move towards those inclusions, potentially due to the stress concentration at these locations, whereas
cracks at other sites propagated directly outwards. Pei et al. (2013) further noted that the stress intensity factor methodology
used for assessment considered the stress to be infinite at crack tips under both linear and power law material hardening. As
inflammation, healing and sedimentation are all possible contributing factors to rupture vulnerability (Versluis et al., 2006),
the inclusion of cell recovery and crack closing rates should be considered (Pei et al., 2013) to fully understand rupture
mechanisms.
Whilst crack propagation in isotropic materials assisted in developing an understanding of initiation and propagation
directions under pulsatile pressure, soft tissues subject to large deformations show anisotropic material behaviours with
very few models having been proposed with the inclusion of damage. Artery damage mechanics were assessed in a single
layered, continuum (Mullins effect) model (Maher, Creane, Lally, & Kelly, 2012) and dual layered (Alastrué, Rodríguez, Calvo,
& Doblaré, 2007b) continuum and stochastic models while Holzapfel and Gasser (2007) investigated the stress-deformation
evolution through a fibre-composite, thick-layered artery as a means to assess non-physiological loading on arteries expe-
rienced during balloon angioplasty. A continuum model resulted in a larger damage localisation area and a lower rate of
stress decrease after unloading compared to the stochastic model, which is based on a fibre stretch damage model. The
strain energy density function for stress-softening (Holzapfel & Gasser, 2007) is defined as

ψSS = (1 − Dm )ψm (I¯1 ) + (1 − D j )ψ j (I¯j ), (2.12)
i=4,6

with ψm and ψ j defining the strain energy density for the undamaged isotropic matrix and undamaged anisotropic fibre
family respectively as well as damage variables for fibre direction (D4 and D6 ) and matrix (Dm ) respectively. Stress soften-
ing was seen in both the extracellular matrix and fibres (Maher et al., 2012), however, the fibres seem to have a higher
influence on strain magnitude with shorter fibres failing earlier than longer ones leading to a rapid stiffening of the ma-
terial (Alastrué et al., 2007b). Maher et al. (2012) noted that the limiting factor of their single layered artery was omis-
sion of the adventitia effects which are noted to be greater at higher strain levels; the adventitia was noted as the major
load bearer at higher pressures (Holzapfel & Gasser, 2007) with more damage experienced in this region due to the larger
stretch imposed on the fibres orientated at 40° to the circumferential direction (Alastrué et al., 2007b). With the shift in
the dominant load bearer from media to adventitia at increasing loads, plaque redistribution occurs through radial com-
pression (Holzapfel & Gasser, 2007), an effect seen in clinical scenarios (Furuichi et al., 2012). Further to the importance of
including cell based mechanisms in artery modelling (Pei et al., 2013; Versluis et al., 2006), it is suggested that a param-
eter to represent collagen cross-linking density is needed to more accurately model damage (Maher et al., 2012) as well
as experiments to relate loading conditions to tissue damage (Holzapfel & Gasser, 2007), with improper residual stress and
material parameters a potential cause for the adventitia showing more damage during stenting (Alastrué et al., 2007b) de-
spite more recent attempts to relate residual strains to physiological, homoeostatic artery pressure (Destrade, Liu, Murphy, &
Kassab, 2012).
12 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

Fig. 7. Cell to cell interaction potentials (a) and interaction forces (b) as functions of inter-cell distance (Melnikova et al., 2017). Reprinted with permission
from The Royal Society.

Micro-structure interactions. To include effects such as collagen cross linking and vascular adaptation, a number of
cellular/molecular based models were proposed (Cyron & Humphrey, 2014; Keshavarzian, Meyer, & Hayenga, 2018;
Melnikova, Svitenkov, Hose, & Hoekstra, 2017). Cyron and Humphrey (2014), for instance, adapted Lyapunov’s stability
theory to examine the mechanical consequences of the degradation and deposition of new material in living tissue.
Keshavarzian et al. (2018) further this by coupling a cellular level, agent based model with a three-layered finite element,
continuum based model to investigate the role of three key factors in the maintenance of homoeostasis, namely proteases,
signalling molecules and growth factors; particularly useful in studying intimal hyperplasia and stent restenosis. The macro
(stress/strains) and molecular level (growth factors, collagen, elastin and gelatine mass and extracellular matrix) coupling
allowed simulation of vascular adaptation through assumptions of hyperelastic, incompressible, isotopic and inhomogeneous
materials. Analysis showed the importance of both mechanical and mechano-biological stability with material deposition in-
cluded to avoid decreased artery structural integrity over time (Cyron & Humphrey, 2014). Increased production of collagen
and collagen cross-linking resulted in improved mechano-biological stability (Cyron & Humphrey, 2014) and also increased
stress, whereas decreased wall stress resulted in decreased cell population and collagen mass (Keshavarzian et al., 2018).
Whilst the innovation of the Keshavarzian et al. (2018) study was the inclusion of a three-layered artery with the struc-
tural role of collagen, elastin and gelatine, they were limited by not considering fibre anisotropy and the active properties
of smooth muscle cells. These were addressed by Melnikova et al. (2017) through the development of a cell-based, hexago-
nal close packed lattice model of the tunica media layer to include residual forces; these were included through increased
attraction forces in the outer layers of smooth muscle cells and cell to cell interactions, as shown in Fig. 7 for molecular
dynamics potentials and forces as functions of cell spacing. This model showed comparable results to uniaxial testing data
for the tunica media layer however the inclusion of many key properties such as a three-layered artery, cell-to-cell interac-
tion, collagen cross linking, material growth and remodelling, active smooth muscle cell properties and anisotropic material
behaviour into one model is yet to be realised.

Model reduction techniques. With increasing model complexity, computation times become longer. In order to circumnavi-
gate this challenge a number of constitutive models were proposed to simplify calculation by either decreasing the number
of essential parameters required to maintain result accuracy (Cilla, Martinez, Pena, & Martínez, 2012; Hollander, Durban, Lu,
Kassab, & Lanir, 2011; Liu, Zhang, Wang, & Kassab, 2011), or circumvent using complex strain-energy functions altogether
(Haddad & Samani, 2017). Hollander et al. (2011) developed a hyperelastic and incompressible model to include residual
stresses, media layer swelling, a continuous elastin scaffold, interlamellar struts and concentric lamellar composed of helical
collagen fibres with gradual activation upon straightening which resulted in estimation error of 4.58% compared to com-
monly used phenomenological and structure-motivated models with 9.71% and 8.99% estimation errors respectively. Only
four parameters were needed compared to 7–9 for the phenomenological model when compared against three-dimensional
data for radial inflation, axial extension and twist. Similar to Hollander et al. (2011), Cilla et al. (2012) only required 4 pa-
rameters in their machine learning tool to reach relative errors of 4.14% and 6.76% respectively for their artificial neural
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 13

network and support vector machines compared to a 55% error in linear regression models due to the nonlinearity of the
challenge. These results were however limited to idealised, straight geometries and training of such artificial neural net-
works requires large clinical datasets. To avoid the use of the complex strain-energy functions, Haddad and Samani (2017)
developed a background tissue with embedded fibres distributed in its volume which allowed conventional strain energy
functions to be used while still accounting for anisotropy, hyperelasticity and active fibre contraction; good agreement was
shown with experimental observations, allowing for faster computation times (Hollander et al., 2011) and the avoidance of
custom, non-linear codes or subroutines (Haddad & Samani, 2017). Despite these reduced parameter models, the role of
stress itself in plaque vulnerability must be established (Cilla et al., 2012a) for models to be used in clinical settings.
Many of these valuable studies were limited to simplifying geometries to straight and cylindrical arteries with static
pressure and a constant opening angle to assess residual stress, although clinical data showed it varied with geometry.
The effects of viscoelastic material properties were also not considered due to limited material data for coronary arteries
(Ohayon et al., 2005) or due to the effects being found insignificant at the applied pressure or pulse range (Versluis et al.,
2006); viscoelasticity is, however, considered an important parameter in the analysis of dynamic biomechanics. The consid-
eration of all three layers of the artery was shown to be important with the intima layer critical in disease initiation and
progression. Furthermore the accurate representation of micro-calcifications in patient specific models resulted in consid-
erable stress variations and altered fracture mechanisms, hence should be included. Whilst structural models of coronary
arteries have presented useful information on the various layer mechanics, plaque constituents and damage, in order to de-
velop biomechanical models capable of predicting disease progression or myocardial infarction the fluid dynamics must also
be coupled and considered at the same time. Table 1 in Appendix A summarises theoretical studies analysing the structural
components of coronary arteries and plaques.

2.1.2. Fluid analysis. Early investigations into the hemodynamic conditions of blood were carried out to model myocardial,
epicardial (Arts & Reneman, 1985; Holenstein & Nerem, 1990; Rooz, Wiesner, & Nerem, 1985; Rumberger & Nerem, 1977)
and coronary vasculature blood flow (Beyar, Caminker, Manor, & Sideman, 1993). Prediction of pressure waveforms in par-
ticular resulted in large amplitude, low frequency oscillations of 5-10 Hz in diastole conditions as the flow deteriorated
(Rooz et al., 1985; Rumberger & Nerem, 1977), an effect that could have structural implications on the artery. Artery cur-
vature resulted in wall shear stresses being highest at the external wall and the formation of secondary flows, resulting in
wall shear stress reversal with pulsatile and Newtonian blood flow (i.e. constant coefficient of viscosity (Fung, 2013)) in a
rigid tube (Perktold, Nerem, & Peter, 1991). The finite element analysis of Newtonian flow resulted in secondary flows of just
3-4% of axial velocity (Perktold et al., 1991) although other studies suggested it could reach as high as 25%. With distance
into the tube, flow reversal and secondary flows become more prominent leading to increased wall shear stress pulsatility,
hence, amplifying the oscillatory shear index (Qiu & Tarbell, 1999), and leading to promotion of vessel inflammation and
atherosclerosis initiation/progression. These early findings clearly indicated the importance of secondary flow regimes and
pulsatile blood flow on increased oscillatory wall shear stress index but were likely limited by computation ability at the
time.
With advances in computation ability, the modelling of increasingly complex parameters became possible. Through a
two dimensional finite-volume model, Bark and Ku (2010) showed that surface roughness increased focal shear stresses
59% for a 90% stenosis with 200 μm surface roughness due to local recirculation zones; Shanmugavelayudam, Rubenstein,
and Yin (2010), however, showed that two-dimensional models underestimated shear stresses when compared to three-
dimensional models under both laminar and pulsatile flows, a relationship that further diverged with increased stenosis.
This is in stark contrast to structural model comparisons where two-dimensional models overestimated peak circumferential
stresses (Ohayon et al., 2005). The effect of plaque morphology on hemodynamics was investigated through simple three-
dimensional computational fluid dynamics (CFD) models of helical lesions (Wong et al., 2006) and eccentric and concentric
plaque (Melih Guleren, 2013) in a non-Newtonian fluid; here dynamic viscosity was modelled by a Carreau fluid defined as:

(q−1)/2
μ = μ∞ + (μ0 − μ∞ ) 1 + λ2 γ 2 , (2.13)
with μ0 , μ∞ viscosity at zero and infinite shear respectively; γ the shear rate and constants λ, q; eccentric plaques resulted
in the formation of hairpin vortices with twisted vortex rings downstream (Fig. 8) producing varying wall shear stress
with particularly large variations at the upper wall (opposite side to stenosis), however, despite increased variation the
maximum wall shear stress was similar in both cases. A 17% pressure increase was seen for the concentric plaques however
results suggested that the spiralling of stenosis spreads the increase in wall pressure seen with stenosis along the artery
(Wong et al., 2006), potentially reducing the risk of artery dilation or aneurysm. These studies, however, are yet to consider
the most atheroprone areas of the coronary vasculature: bifurcations.

Fluid Models. Despite studies growing in complexity and depth, one of the two major simplifications in fluid-based coronary
artery models was the simplification of blood to a Newtonian fluid (Bark & Ku, 2010; Chua, Yu, & Xue, 2001; Dash, Jayara-
man, & Mehta, 1999; Frattolin, Zarandi, Pagiatakis, Bertrand, & Mongrain, 2015; Huo et al., 2012; Kachanov & Abedian, 2015;
Keshavarz-Motamed & Kadem, 2011; Migliavacca et al., 2006) despite its shear-thinning nature, due to the suspension of
molecules such as blood cells, lipids and platelets. Reasoning used for the simplification was that wall shear effects were
only present at low shear rates; as the rigid artery conditions used in computational fluid dynamics produced shear stresses
14 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

Fig. 8. Three-dimensional vortex structures for the LAD with concentric (upper) and eccentric (lower) stenosis. VR = vortex ring, HV = horseshoe/hairpin
vortex, TVR = twisted vortex ring (Melih Guleren, 2013). Reprinted with permission from Elsevier.

above the relevant threshold for a non-Newtonian model, fluid was assumed Newtonian. In physiological conditions low wall
shear stress, important in oscillatory shear rates, correlated to inflammation and particulate build-up leading to atheroscle-
rosis, hence should not be neglected. Moreover, Li, Wen, and Li (2001) showed that for a power law, non-Newtonian model,
defined as
η = η0 γ˙ n−1 , (2.14)
for the fluid viscosity η, strain rate γ˙ and constants η0 , n; maximum value for wall shear stress decreased at the stenosis
throat with increased macromolecule concentration near the stenosis compared to Newtonian fluid. Wall permeability was
also dependant on wall shear stress hence wall flux was highest for maximum shear stress demonstrating the importance
of shear-thinning, non-Newtonian fluid properties. In particular, the primary shear-thinning models used to date were the
Power Law (Giannoglou, Soulis, Farmakis, Farmakis, & Louridas, 2002; Wong et al., 2006) and Carreau model (Banerjee et al.,
2007; Jung, Lyczkowski, Panchal, & Hassanein, 2006; Melih Guleren, 2013; Sinha Roy, Back, & Banerjee, 2006), however,
Anand et al. (2013) proposed a new Oldroyd-B model to include a thermodynamic framework and deformation-dependant
viscoelastic behaviour which allowed for smooth transitions of material properties across all shear rates with four parame-
ters (shear thinning viscosity denoted by η and four constant parameters η1 , m, μ, α ), such that:
⎡ ⎤−m
 
2η γ˙
2 2
1
η = α⎣ 3 +  ⎦ . (2.15)
4 μ2
1 + μ γ˙
2 2

4 μ2

This model included shear lag for modelling complex geometries, such as those seen in aneurysms, and was validated
against experimental data as shown in Fig. 9, further highlighting the nonlinearity associated with coronary hemodynam-
ics. The second major simplification of fluid studies was the replacement of pulsatile flow conditions with steady flow
(Giannoglou et al., 2002; Huang et al., 2007; Wong et al., 2006). Over consecutive stenosis, Akherat and Kimiaghalam (2010)
showed the occurrence of secondary plaque was determined by the flow reattachment point and for larger stenosis (64%),
wall shear stress gradient intensified downstream, increasing vulnerability. Methods to scale wall shear stress onto a single
curve for fast clinical application (Chua et al., 2001) showed further that the complex nature of pulsatile blood flow limits
the application of reduction techniques. Both Li et al. (2001) and Akherat and Kimiaghalam (2010) showed the incorporation
of pulsatile flow caused variation and intensification of all parameters, with wall shear stress variation more than tripling
compared to steady flow.
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 15

Fig. 9. Shear stress variation showing lag (hysteresis) at the end of a triangular loop of shear rate (Anand et al., 2013). Reprinted with permission from
Elsevier.

With the abilities of interventional techniques improving, an understanding of the effect that these techniques play on
hemodynaimcs and artery properties is important for both planning and assessment purposes. Coronary catheterisation can
provide a range of information on in vivo properties however the effect of the catheter (Dash et al., 1999) and guidewire
for balloon angioplasty (Sinha Roy et al., 2006) insertion into the artery altered local hemodynamic effects. In a Newtonian
fluid under steady flow catheter insertion increased pressure drop by a factor of 1.60-5.16 for a curved, stenotic artery and
1.74-4.89 in a straight, healthy artery for catheter-vessel ratios of 0.1-0.4 (Dash et al., 1999). In a non-Newtonian fluid model
(Sinha Roy et al., 2006), guidewire insertion led to increased mean hyperaemic pressure drop and flow reserve resulting in
an overestimation of stenosis severity; wall shear stress along the guidewire was found to be 35 and 50% higher in transient
and steady flow respectively than along the artery wall leading to increased flow resistance and elevated viscous losses
(Banerjee et al., 2007). To more accurately model collateral flow through anastomosis (Harmouche, Maasrani, Corbineau,
Verhoye, & Drochon, 2012) a collateral flow index was proposed based on aortic and central venous pressures and pressure
distal to thrombosis which showed increased sensitivity to distal pressure and could be used as a more accurate clinical
evaluation tool in interventional procedures or assessment of restenosis in stenting or shunt insertion (Migliavacca et al.,
2006).
The modelling of the complex parameters of blood showed many challenges. Whilst the geometries of the coronary vas-
culature and plaque morphology influenced wall shear stress and flow recirculation results, no consistent relation between
artery geometry and blood flow was found. Blood flow’s pulsatile profile caused by the heart’s pumping motion resulted in
intensified wall shear stresses (including a tripling of variation in wall shear stress) and flow turbulence; this highlighted the
importance of considering physiologically accurate pulsatile blood flow. Purely fluid based analyses were mainly conducted
through CFD, hence, arteries were modelled as rigid tubes; due to this the simplification of flow to a Newtonian fluid was
justified as often the low wall shear stresses needed to differentiate between a Newtonian and non-Newtonian fluid were
not reached. To account for the complex blood constituents and assess the shear-thinning nature of blood, including platelets
and red blood cells, non-Newtonian fluids should be implemented. Many studies investigated blood in an idealised context;
consideration of blood disease parameter variation may also shed light on the role of blood in artery biomechanics whilst
flow outlet conditions, such as considering capillary flow along the length of the artery, may also influence results however
are yet to be investigated. As two-dimensional fluid and structural models were shown to underestimate and overestimate
peak stresses respectively, and CFD models were unable to consider fluid-induced motion, the use of three-dimensional, fluid–
structure interaction (FSI) modelling techniques should be implemented. Table 2 in Appendix A summarises these theoretical
studies analysing the fluid components of coronary arteries and plaques.
16 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

2.1.3. Fluid–structure interaction (FSI) analysis


The limitations with structural and fluid based analyses carried out in the preceding sections were that key parameters
effecting the inputs to each model were omitted as the boundaries to each problem change as they interact between the
blood flow and artery wall. Due to this, the use of fluid–structure interaction (FSI) (Kachanov & Abedian, 2015; Mandal,
Mukhopadhyay, & Layek, 2012a; Taelman, Degroote, Swillens, Vierendeels, & Segers, 2014) is essential for the accurate
modelling of artery biomechanics. A number of FSI based studies developed models to investigate either carotid arteries
(Janela et al., 2010; Pu, Xiong, Liu, Zhang, & Zhang, 2014; Wenk, Papadopoulos, & Zohdi, 2010), basilar artery (Karšaj et al.,
2010), aorta (Taelman et al., 2014) and vertebral artery (Tricerri et al., 2016), however work on understanding the biome-
chanics of coronary arteries is limited, potentially due to the difficulties associated with including the many complex pa-
rameters associated with heart or respiratory motion and the challenge of extracting reliable data from a region of the
body difficult to image accurately in vivo. FSI models are also numerically complex and moreover, expensive due to real-
time fluid–solid couplings. Modelling hyperelastic coronary arteries under the Mooney-Rivilin and Newtonian blood models
(Pakravan, Saidi, & Firoozabadi, 2017; Rotman, Zaretsky, Shitzer, & Einav, 2017), the morphology of endothelial cells was
predicted under mechanical stimuli from microfilaments, modelled as Voight viscoelastic elements, connecting two compu-
tation nodes (Pakravan et al., 2017). A multiscale simulation to couple artery and cellular scale biomechanics showed that
the myocardial side of bifurcations was most at risk of atherosclerosis, a result later corroborated by Jahromi, Pakravan,
Saidi, and Firoozabadi (2019). Dynamic artery curvature, a characteristic later shown to also be a marker for atherosclerosis
(Gholipour, Ghayesh, & Zander, 2018a), caused increased shape index and reorientation variation in cells, also leading to
atherosclerosis, however the impact of assuming Newtonian flow in these models may have affected accuracy and clinical
applicability of the model.
It is believed that the two studies by Gholipour, Ghayesh, Zander, and Mahajan (2018b) provide the first purely theo-
retical, in depth biomechanical assessment of non-Newtonian flow in coronary arteries in finite element fluid–structure in-
teraction modelling with the inclusion and assessment of the impacts of micro-calcification, blood pulsation, three-layered
arteries with taper, media layer contraction, asymmetric heart motion, plaque morphology and viscoelastic effects; modelled
with the Prony shear relaxation method to determine the Cauchy stress defined as
 
t
de t
d

σ= 2G(t − τ ) dτ + I K (t − τ ) dτ , (2.16)
0 dτ 0 dτ
where
and е represent the volumetric and deviatoric strains respectively, t is time and the identity tensor is defined by I.
The Prony series, G(t), and Bulk relaxation moduli K(t), are then defined respectively as:
  ⎫ 

nG

α exp − G ⎪
t
G(t ) = G0 α + G G ⎪


τi
i ⎬
i=1
   , (2.17)
nK ⎪

K (t ) = K0 α∞
K
+
t
αiK exp − K ⎪

τi ⎭
i=1

where Go and Ko represent the relaxation moduli defined when time is equal to zero, n=5 for the number of terms, relative
moduli are defined as α∞ G and α K respectivly and the relaxation times to develop the material coefficients are defined by

τiG and τiK respectivly. The effect of including all of these parameters was shown clearly in Fig. 10 with a significant in-
crease in von Mises stress of 511%. Furthermore simplifying fluid properties to a Newtonian fluid, compared to a power law
non-Newtonian model (see Eq. (2.14)), resulted in the largest stress variation of any simplification (Fig. 11) highlighting
the magnitude of error potentially associated with many previous assumptions. The presence of micro-calcification was
found to increase the von Mises stress drastically in the artery leading to increased plaque rupture vulnerability; with
an increasing plaque shoulder region, wall shear stress was reduced, however, the structural stress on the plaque fur-
ther increased. The expansion of this model to include the left main artery bifurcation point (Gholipour et al., 2018a),
as shown in Fig. 12 and a known atheroprone area, resulted in the largest risk of rupture resulting from the left ante-
rior descending as well as left main artery plaques, with von Mises stress proportional to increased stenosis. In contra-
diction to earlier studies showing that the potential for downstream plaque formation is increased (Akherat & Kimiagha-
lam, 2010) at the location of flow reattachment, Jahromi et al. (2019) showed that progression of the primary plaque length
was more likely than new plaque formation. Increased artery curvature resulted in an increased risk of rupture; plaque
shoulder width reduced the von Mises stress in the plaque itself and the wall shear stress on the endothelial layer. Whilst
these models provided a leap in biomechanical modelling accuracy, results showed that both the artery and plaque were
prone to rupture, leading to a gap for three-dimensional rupture mechanisms to be assessed in fluid–structure interaction
scenarios.
Literature to date has not managed to provide three-dimensional plaque crack initiation (and hence thrombus forma-
tion) and propagation analysis with the inclusion of fluid–structure interaction in a coronary context. Whilst not focused on
coronary biomechanics, Perkowska et al. (2017) studied a crack driven by viscous fluids which could have applications to
modelling artery failure mechanisms. By assessing multiple fracture modes for hydraulic action of a viscoelastic fluid, clas-
sical theories such as the maximum circumferential stress (MCS) and minimum strain energy density (MSED) resulted in
differences when the position of the initial crack did not align with the orientation of the principal stress in low toughness
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 17

Fig. 10. The effect of including all artery parameters compared to a simplified model (Gholipour et al., 2018). Reprinted with permission from Elsevier. (For
interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 11. The effect of Newtonian and non-Newtonian flow on wall shear stress (Gholipour et al., 2018). Reprinted with permission from Elsevier.

materials. Crack propagation in low toughness and highly viscous dominated regimes, such as modelling the viscous effects
of blood in arteries, were shown to be highly impacted by hydraulic shear forces; a hydraulic shear stress intensity factor
was developed through the linear elastic fracture mechanics approach for the first three fracture modes, with υ Poisson’s
ratio, which resulted in the energy release rate and fracture criterion (Perkowska et al., 2017) respectively as
⎧ ⎡ ⎤ ⎫⎫
⎪ Hydraulic shear stress
⎪⎪

⎨ ⎢ 2
intensity factor
%&'( ⎥ ⎬⎪
⎪⎪

1+υ ⎥ + KI2II ⎪

ε= ( 1 − υ )⎢
⎣KI + KII + 4(1 − υ )KI
2
Kf
⎦ ⎪

E ⎪⎪ ⎪
⎪⎬
⎩ ⎭ . (2.18)


KI2 + KII2 + 4(1 − υ )KI +
1
K 2 = KIC
2 ⎪

Kf
'(%& 1 − υ III ⎪



Hydraulic shear stress ⎭
intensity factor

The ability to model these mixed mode fracture effects could lead to a much stronger understanding of artery and plaque
failure and hence the ability to predict them. Furthermore, the impact of multiple-constituent blood models shown in the
18 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

Fig. 12. The location of the modelled bifurcation and the three-dimensional finite element model (Gholipour et al., 2018). Reprinted with permission from
Elsevier.

fluid analysis section on platelet aggregation, red blood cell and plasma interaction and what role they have in coagulation
leading to atherosclerosis formation or failure initiation requires the coupling of the macro FSI model to cellular level mod-
els. Similarly, the impact of cellular level interaction within the artery walls and their ability for growth and remodelling
should be considered. As the primary source of inflammation and damage which leads to atherosclerosis, endothelial cell
remodelling and alignment are important for the characterisation of vulnerable areas and have been shown to be impacted
by circumferential and longitudinal strains as well as wall shear stress variations (Jahromi et al., 2019). With Jahromi et al.
(2019) also showing the myocardial side of the artery has higher wall shear stress leading to higher probability of athero-
genesis, the development of models to incorporate the biomechanics of the cardiac muscle such as the myocardium and
pericardium shows promise for more accurate predictive capabilities. Table 3 in Appendix A summarises these theoretical
studies which investigate coronary arteries and associated plaques through fluid–structure interaction analysis.
In conclusion, theoretical, biomechanical studies of coronary arteries have contributed greatly to understanding the com-
plex interactions and implications associated with cardiovascular diseases such as atherosclerosis. Focused investigations
of artery macro and micro-structure, disease progression including plaque formation and fracture, plaque morphology in-
cluding micro-calcifications and lipid pool size, artery curvature and motion, pulsatile blood flow as well as variation in the
modelling approach for the artery structure and blood properties resulted in a more in-depth understanding of human coro-
nary biomechanics. In order for the field of coronary artery biomechanics to progress there are two major considerations
to assess; firstly, accurate data and properties for both artery material and blood rheology must be extracted from human
samples and second, patient-specific biomechanical models should be analysed by reconstructing the vasculature from in
vivo clinical imaging. This will result in more accurate biomechanical analysis subsequently ensuring more available data for
clinicians to assess disease progression or surgical interventions; studies investigating these two considerations are assessed
in the following sections.

2.2. Experimental/clinical investigations

Experimental studies have assisted in the determination of mechanical properties of both healthy and diseased arteries.
Whilst studies on animal coronary properties have been more prominent, human studies have begun to develop an under-
standing of the complex behaviour of arteries and plaques in both clinical and lab based conditions. This section reviews
these studies by dividing them into in vitro and in vivo classifications. Early in vitro studies primarily focused on the de-
velopment of models to duplicate the conditions experienced in coronary arteries as well as mechanical testing of arteries
to extract material properties used in the modelling of their biomechanics. In vivo studies were clinically based with imag-
ing of arteries often used to extract parameters useful in the assessment of atherosclerosis progression. Tables 4 and 5 in
Appendix A summarises these studies.

2.2.1. In vitro experimental/clinical studies


A major component of in vitro based experimental studies focused on constructing idealised models of coronary arteries
(Feng et al., 2015; Lin et al., 2006; Mark, Bargeron, Deters, & Friedman, 1985); however replicating the complex viscoelas-
tic behaviour resultant from muscle fibre orientation and plaque constituents proved challenging. Rather than fabricating
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 19

Fig. 13. Human coronary artery before (a) and during (b) uniaxial tensile testing (Karimi et al., 2013). Reprinted with permission from Elsevier.

models to attempt to assess the properties of coronary arteries, Feng et al. (2015) investigated the accuracy of using wave-
form analysis to predict plaque rupture. Using fat and bone to replicate lipid and calcification percentages they showed that
plaque hardness increased pressure variations whilst calcification within the plaque produced local variations in pressure
and flow, detectable through non-invasive arterial waveform analysis. This technique could lead to the development of future
tools for the detection of plaque vulnerability however did not provide any information on material properties. A parallel
plate flow system was developed by Lin et al. (2006) to investigate the response of calcified cells exposed to shear stress and
suggested that calcification actually has no effect on plaque vulnerability as the mineralisation could cause stronger bonds,
in contrast to many later studies (Buffinton & Ebenstein, 2014; Cardoso et al., 2014; Cilla et al., 2013a; Nguyen & Levy, 2010;
van der Giessen et al., 2011). Using a similar parallel plate flow system to investigate shear stress effects, Albuquerque and
Flozak (2001) focused their investigation on endothelial cell growth and wound recovery. Their results showed improved
motility, filament and clumping patterns for β -actin suggesting that dynamic shear stresses improved biological recovery
compared to static flow conditions (Albuquerque & Flozak, 2001). This result showed the potential for enhancing the cellular
level cross-talk during the development of vascular grafts leading to more mechanically and biologically stable interventions.
To better understand the biomechanical properties of arteries, geometric and mechanical properties must be accurately
determined. An early study by Chamiot-Clerc, Copie, Renaud, Safar, and Girerd (1998) focused on discerning the diameter,
wall thickness and reactivity of mammary and radial arteries to assess their suitability for bypass graft surgery. Results
suggested that differences in smooth muscle mass and sensitivity to vasoactive compounds resulted in higher strains and
distensibility in the radial artery but lower elastic modulus compared to the mammary artery; with far greater radial artery
tension due to a seven fold higher vasoconstricting agent sensitivity, graft occlusion due to vasospasm was heightened.
Building on this, the biomechanical properties of coronary arteries were determined through undertaking uniaxial tensile
testing on excised artery segments (Holzapfel, Sommer, Gasser, & Regitnig, 2005; Jankowska, Bartkowiak-Jowsa, & Bedzinski,
2015; Karimi, Navidbakhsh, Shojaei, & Faghihi, 2013b; Karimi, Sera, Kudo, & Navidbakhsh, 2016; Karimi, Shojaei, & Razaghi,
2017b). Material properties of each individual artery layer were determined from healthy left anterior descending arteries
with hysteresis evident in every layer (Holzapfel et al., 2005). Cyclic preconditioning of arteries before testing was important
for result convergence with artery axial stretch determined to be 5% of overall load free length. The adventitia was found
to have tensile stress of three times the media and intima with the adventitia and intima showing higher circumferential
stiffness and the media layer higher longitudinal stiffness. All results showed large variation which highlighted the varia-
tion between specimens. A key material assumption of arteries being a soft matrix material with imbedded collagen fibres
was validated through the non-linear response and stiffening behaviour over finite stretches; increased Young’s modulus
and maximum stress calculated using both Odgen and Mooney-Rivilin models were hypothesised to be related to colla-
gen and elastin content in the artery wall, however, the effect of atherosclerosis on these mechanical parameters was not
investigated.
To determine the effect of atherosclerosis on mechanical properties of arteries a further series of uniaxial tensile tests
were carried out on both healthy and diseased specimens (Jankowska et al., 2015; Karimi et al., 2013b, 2016, 2017b), as
shown in Fig. 13. Results indicated that diseased (atherosclerotic) arteries tolerated 44.55% higher and 34.61% lower stress
and strain respectively, compared to healthy arteries. Both maximum and physiological elastic moduli were also found to
be magnified in diseased arteries (Karimi et al., 2013b); the highest stress and strains were seen in the circumferential
direction (Jankowska et al., 2015), consistent with earlier literature. With many theoretical models assuming incompressible
artery properties, Karimi et al. (2016) investigated this assumption based on Poisson’s ratio. Atherosclerosis progression
seemed to have no effect on Poison’s ratio of diseased vs healthy arteries suggesting that the incompressibility assumption
was valid. Combining uniaxial testing with direct image correlation methods showed highly viscoelastic behaviour for both
healthy and atherosclerotic arteries, with increasing viscoelasticity in diseased arteries (Karimi et al., 2017b). Atherosclerosis
resulted in a significant increase in stress from 18±2.88 to 292±16.14 kPa and slower time to reach a stress balance after
loading compared to healthy arteries suggesting that collagen fibres lost their elasticity and could not recover and release
20 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

energy fast enough, potentially leading to damage due to overload; a similar assumption to Jankowska et al. (2015). Despite
numerous uniaxial tensile tests, biaxial testing would expand the available data (Holzapfel et al., 2005) on the anisotropic
properties of arteries. Pressure testing (Ozolanta, Tetere, Purinya, & Kasyanov, 1998; Zhao et al., 2007) was used to study the
effect of ageing showing that arteries thicken nonuniformly and connective tissue volume increases with age. Right coronary
as well as left anterior descending arteries experienced increased rigidity with age, counterintuitive to the suggestion by
Karimi et al. (2013b) that age does not alter stiffness; elastic modulus and strain energy were seen to increase and decrease,
respectively.
The investigation of the biomechanical properties of arteries used in vein grafts was also carried out in a number of
studies for the radial and mammary arteries (Chamiot-Clerc et al., 1998), and saphenous vein (Montini-Ballarin et al., 2016;
Zhao et al., 2007). Despite not being coronary arteries, the characteristics of these veins and arteries are important for un-
derstanding how they interact in the coronary vasculature due to their use in interventions such as bypass surgery. Pressure
testing was used to determine the limitations of the artery to better circumvent causing trauma during grafting such as the
stiffening of the artery caused by the wall becoming oedematous (Zhao et al., 2007). This could lead to further implications
in the design of synthetic grafts based on the mechanical properties of arteries. Nanofiber mesh (He et al., 2006; Montini-
Ballarin et al., 2016) and biomaterial scaffolds (Sharifpoor, Simmons, Labow, & Santerre, 2011) were investigated for potential
use as vascular grafts and in tissue regeneration. Electro-spun nanofiber mesh showed ideal porosity and surface chemistry
conditions for the infiltration and growth of cells to remould extracellular matrices and maintain the functionality of en-
dothelial cells (He et al., 2006); this highlighted the importance of incorporating microscale mechanics into biomechanical
modelling. The orientation of the mesh fibres dictated the mechanical properties highlighting the importance of fibre align-
ment in biomechanical modelling of coronary arteries. The use of biodegradable scaffolds also showed that uniaxial cyclic
mechanical strain during cell growth drastically improved the proliferation, penetration and contractile function of smooth
muscle cells into the scaffold (Sharifpoor et al., 2011), improving the tensile mechanical properties compared to static cell
growth and highlighting the importance of using accurate biomechanical modelling to replicate the physiological environ-
ment and improve tissue regeneration. Synthetic polymers, segmented poly(ester urethane) (PHD) and poly (L-lactic acid)
(PLLA), have also been used to mimic the biomechanical characteristics of collagen and elastin which were found from in
vitro testing (Montini-Ballarin et al., 2016). The similar behaviour of the biological and synthetic materials also allowed for
burst pressure and suture strength compliance to closely mimic the physiological characteristics of arteries showing promise
for tissue engineered grafts in the future.
The study of coronary biomechanics through experimental/clinical in vitro scenarios focused on discerning the relation-
ships between artery and plaque biomechanical properties and disease progression. Fluid shear stress was shown to be
important in the function of endothelial cell growth and recovery from wounding whilst variations between the mechani-
cal properties of the intima, media and adventitia layer had implications for the biomechanical response and load uptake
of the artery structure. Diseased arteries exhibited significantly higher stress and lower strain than healthy arteries and a
decreased ability to recover from loading which could lead to damage. The vasoconstricting reactivity of arteries was tested
to determine their suitability in grafting, however, no mention was made of changes in vasoreactivity under varying stresses
and strains; the conditions at which many arteries were tested were also not accurately described. Factors such as time
since excision from the body, storage temperature, testing temperature, preconditioning method or the lifestyle parameters
of each patient were rarely outlined in detail. This makes drawing conclusions about the mechanical response of arteries
challenging, hence, consistent testing procedures should be adopted to be able to accurately compare results. Table 4 in
Appendix A summarises these experimental studies analysing the components of coronary arteries and plaques in vitro.

2.2.2. In vivo experimental/clinical studies


A number of early in vivo studies on the human vasculature were undertaken to determine the effect of obesity on car-
diac and contractile function (Carabello & Gittens, 1987) and measure the geometric parameters of the coronary vasculature
(Brinkman, Baker, Newman, Vigorito, & Friedman, 1994), such as the effect of artery wall thickening at bifurcation points
(Friedman & Ding, 1997). The development of a system to measure the geometric parameters of arteries such as curva-
ture, tortuosity, length and branch angles resulted in large parameter variation, however, the left main artery bifurcation
angle was on average much larger than the subsequent bifurcations which has implications for atherosclerosis vulnerability
(Brinkman et al., 1994). The angle of this bifurcation was shown to positively correlate to intima and media thickness and
circumferential asymmetry with increased branch angles leading to increased eccentric intima thickening due to decreased
shear stress and increased flow recirculation, contributing to atherosclerosis (Friedman & Ding, 1997). With developing IVUS
technology, strain in the artery wall and plaques was feasibly measured with results indicating that calcification was asso-
ciated with significantly lower strain values (van der Steen et al., 2002), however, plaque size had also been overestimated
through the inclusion of the media layer in IVUS imaging (Tajaddini, Kilpatrick, & Vince, 2003). Despite this, IVUS imag-
ing in patients both with and without angiographic coronary artery disease showed that a reduced plaque cross-sectional
area during systolic cycle is a major determinant of artery cross-sectional compliance (Shaw et al., 2002); with a stable
catheter required for data acquisition, cardiac and respiratory motion (Sauvée, Noce, Poignet, Triboulet, & Dombre, 2007)
pose challenges to imaging technologies.
The relationship between atheroprone regions and morphological features was investigated in an attempt to determine
geometric risk factors for atherosclerosis (Bulant et al., 2017; Konta & Bett, 2003; Zhu, Ding, Piana, Gehrig, & Friedman,
2009). By classifying varying artery motion into three categories; bend, compression and displacement type motion (Konta
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 21

& Bett, 2003), the compression type category resulted in significantly higher stenosis than the other categories suggesting
that mechanically stressing vascular tissue led to endothelial injury and eventually atherosclerosis. A deeper assessment of
atherogenic risk factors in the distal section of the left anterior descending artery suggested that curvature, tortuosity and
torsion account for over 80% of variability in mechanical stress and were highest in this region (Zhu et al., 2009). Curvature
in particular was shown to be a potential risk factor showing a significant correlation to artery stenosis (Bulant et al.,
2017), corroborating this finding. Results from this study, however, were primarily focused on high risk patients with factors
such as hypertension or dyslipidaemia, potentially impacting results and suggesting artery straightness could be a protective
mechanism rather than an effect of atherosclerosis; no further studies addressed this hypothesis. Relating the morphological
features to biomechanical changes seen in cardiac allograft patients (Lin et al., 2014) with two-dimensional MRI and three-
dimensional angiogram imaging showing that thicker walls and a higher percentage of the vessel wall occupying the total
vessel area indicated higher stiffness, a result similar to that expected in elderly hypertension patients. It was suggested that
endothelial cells could modulate total artery stiffness due to pathological similarities between the progression of lesions and
atherosclerosis, further highlighting the importance of the intima layer on the progression of disease; a result impacting
studies that simplified arteries to single or dual layered vessels.
The use of IVUS for determination of strain distribution in artery walls, whilst overcoming the challenge associated with
catheter motion including continuous pullback and tissue movement, showed promise as an early indicator of plaque insta-
bility (Keshavarz-Motamed et al., 2014; Liang, Zhu, Gehrig, & Friedman, 2008). The transverse strain tensor was determined
to show that heterogeneous plaque morphology resulted in higher strain, thicker lipid pools and lower strain associated
with fibrous plaques; similar to previous studies, radial strain magnitude was concentrated at the plaque shoulders sug-
gesting increased vulnerability (Liang et al., 2008). Alternativly, a Lagrangian speckle model estimator was used to calculate
shear strain elastographs that showed potential as an early indicator for atherosclerosis (Keshavarz-Motamed et al., 2014).
Inflammation in particular was seen to increase the shear strain elastograph reading; as inflammation is a primary cause
of atheroclerosis initiation and progresson, effecting many microstructure factors such as collagen compostion and growth
factors; this may hold strong potential as an early indicator. The method also showed good sensitivity to soft plaque con-
stituents which may be useful in assessing the stresses in plaque shoulders where the softer artery layer undergoes stress
intensifications and is a potential rupture site. The images were segmented and combined through the outlining of the ad-
ventitia and lumen boundaries with a fast-marching model, validated manually by a cardiologist. This image registration
was a key process in the accurate determination of artery biomechanics, hence accurate reconstruction is a critical factor.
The co-registration of images obtained clinically from virtual histology IVUS and OCT was important for the assessment
of disease progression and also plays a vital role in the reconstruction of three-dimensional biomechanical models from two
dimensional images. Changes in parameters such as necrotic core size, which has been shown to impact plaque vulnerability
in specific sectors of a two-dimensional artery image, validated focal relationships between plaque progression and its en-
vironment (Timmins et al., 2013). In order to longitudinally co-register both IVUS and OCT images, a dynamic time warping
technique was developed by Molony, Timmins, Rasoul-Arzrumly, Samady, and Giddens (2016). The challenge of variations
in catheter pullback speeds was met by this method which co-registered images based on lumen eccentricity, calcification
angle and catheter angle resulting in excellent agreement to expert readers at faster speeds.
In vivo experimental/clinical studies have shown the importance of variations in morphological parameters as potential
atherogenic risk factors and the intima layer was highlighted as crucial for the modulation of artery stiffness. Artery tortu-
osity, curvature and torsion were suggested as the largest cause of variation in mechanical stress and the use of dynamic
time warping techniques showed the ability to accurately co-register imaging techniques whilst including catheter pullback
speeds; critical for the development of accurate biomechanical models. While purely experimental/clinical investigations
have uncovered many important coronary artery properties, they are unable to determine potential rupture sites or myocar-
dial ischaemia initiation without the development of theoretical, biomechanical models of coronary arteries. The remainder
of this paper investigates the combination of experimental/clinical techniques with theoretical modelling of coronary artery
biomechanics. Table 5 in Appendix A summarises the experimental/clinical studies analysing the components of coronary
arteries and plaques in vivo.

2.3. Combined theoretical and experimental/clinical investigations

This section of the review contains studies that carried out experimental/clinical investigations to determine artery char-
acteristics which were then used in theoretical modelling of coronary arteries and plaque biomechanics. This section is
again divided into “in vivo” and “in vitro” groups. In “in vitro” studies, experimental/clinical efforts were applied to arter-
ies and plaque after harvesting them from the body for either mechanical or morphological testing. Some studies noting
experiments as “ex vivo” are also presented in this section as tissues are no longer in their normal physiological environ-
ment, hence, the properties and boundary conditions associated with their mechanics are no longer physiologically accurate.
In vivo testing is therefore essential for the accurate representation of artery and plaque properties, however, determining
these is a challenge due to the coronary artery vasculature location and constant motion with the cardiac and respiratory
cycles. Each main group is then again divided into structural, fluid and fluid–structure interaction analysis, similar to the
purely theoretical studies already investigated.
22 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

Fig. 14. Geometry reconstruction procedure for high and low sampling as well as 3D and 2D reconstruction (Nieuwstadt et al., 2013). Reprinted with
permission from Elsevier.

2.3.1. In vitro theoretical and experimental/clinical studies


The in vitro section contains studies that undertook laboratory experiments using both fabricated and human artery
segments. The testing of human artery segments was noted as “ex vivo” in many papers however as these experiments
were conducted on the artery segments outside of the normal physiological environment they have been grouped in this
section. Studies focused on uniaxial and biaxial tensile tests, pressure/inflation testing, fabrication of artery or fluid flow
models, microstructure growth and remodelling and the fabrication of composite arteries or stents through nanofiber mesh.
Properties obtained from the experimental studies were used in the theoretical and simplified models including mechanical
properties, fluid models, artery and plaque morphology and motion. Overviews of these studies are outlined in Tables 6–8
in Appendix A.

Structural Analysis. Two-dimensional images acquired from technology such as IVUS and OCT require reconstruction by co-
registration into three-dimensional artery models for biomechanical modelling. Various techniques, such as ‘warping’ were
analysed to discern sensitivity to various parameters such as material coefficients (Veress, Weiss, Gullberg, Vince, & Rabbitt,
2002); the inclusion of catheter pullback is required for clinical applications due to two-dimensional images being sub-
jected to out of plane motion of up to 1.5 mm. Axial sampling resolution in two and three dimensions was investigated
for the construction of both two and three-dimensional models, as shown in Fig. 14, which outlined that under-sampling
increased error variation in peak plaque cap stress (Nieuwstadt et al., 2013). A ±15.5% variation between high and low sam-
pling error and a larger ±24.0% variation between high three-dimensional sampling rate and two-dimensional geometries
highlighted the importance of not only the correct sampling resolution but also the criticality of using three-dimensional
modelling and imaging to accurately compute plaque cap stresses; this was further highlighted by a 48% variance in peak
cap stress measurements between two and three-dimensional models using hyperelastic and anisotropic material proper-
ties. Undersampling was also seen to lead to omission of key morphological features, such as small necrotic cores, if the
lower resolution limit was not met. A three-layered artery was used to study the effects of manually reconstructing the
outer side of necrotic core areas; a major limitation of OCT’s penetration depth (Kok et al., 2016). Whilst averaged geometric
data resulted in peak cap stresses being underestimated, reconstructing the backside of the necrotic core from averaged
clinical data improved result estimation, showing potential for in vivo application. In contrast to this method, the inclusion
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 23

of IVUS imaging could account for the shortfall in OCT penetration depth and circumvent the need for geometry estima-
tion altogether, however, the largest stress variation between histology and OCT comparisons was seen at the inner lumen
(Chau et al., 2004), the area with the lowest OCT signal attenuation, highlighting the limitation of visual inspection of coro-
nary arteries from medical imaging and the need for biomechanical analysis.
Further building on variations seen between two and three dimensional parameters, Karimi, Navidbakhsh, Faghihi, Sho-
jaei, and Hassani (2013a) used uniaxial testing and direct image correlation to determine the mechanical properties used in
two and three-dimensional investigations of peak plaque cap stresses in calcified, hypocellular and cellular plaques over a
70-90% stenosis range. From the neo-Hookean, hyperelastic and single layered model of an isotropic material and in contrast
to previous literature, results suggested that calcification played a protective role for plaques. With variations in Young’s
modulus of up to 50% across both linear and nonlinear isotropic models and the inclusion of residual strain, stress and
strain variation was less than 10% (Williamson et al., 2003), however, sensitivity increased when shifting from isotropic to
orthotropic material models. Hypocellular plaques resulted in the highest stress of 123.2 kPa compared to 52.52 kPa for calci-
fied plaques over a three-dimensional model of a 70% stenosis, results that further suggested that calcification could have a
stabilisation effect on plaques (Chau et al., 2004); by comparison, the two-dimensional model resulted in a maximum stress
of just 66.59 kPa, further highlighting the effect of two-dimensional simplification. In contrast, through finite element anal-
ysis under uniaxial and biaxial tensile testing, peak stress levels at the edge of calcifications modelled as rigid inclusions
were amplified suggesting that the inclusion of calcification leads to stress intensifications and an increased risk of rup-
ture (Hoshino et al., 2009). The residual stress state of arteries (Speelman et al., 2011) and plaque stress and vulnerability
(Mangalaprakash, Kumar, & Balakrishnan, 2013) were further investigated; accounting for residual stress, plaque cap stress
showed a large variation with a general decrease of 10±15% for plaque caps under 213 μm in thickness and a significant
enlargement of the lumen area when not accounting for residual stress (Speelman et al., 2011). The use of a single layered
artery model, however, limited the results of the study; this is important for residual stress/strain as non-uniform, residual,
transmural strains could result in uniform stress in composite models but not homogeneous ones (Guo, Lu, & Kassab, 2005).
Looking at both of these implications combined, Mangalaprakash et al. (2013) showed that the presence of calcification
close to the lumen surface increased lumen stresses however when it was located deeper in the plaque cap these stresses
were decreased; this could account for the variability of results seen for calcification inclusions to date. Whilst stress was
seen to peak at the plaque shoulders, a result corroborated by literature, Mangalaprakash et al. (2013) noted no direct trend
between calcification and stress; complex heterogeneity of material and geometry were noted as potential causes, although
no shear stress was accounted for in this study. Contrary to this finding, the presence of spherical micro-calcifications in an
isotropic, neo-Hookean, finite element model resulted in a two-fold increase in local stress at inclusion poles; close cluster-
ing resulted in a five-fold stress increase (Maldonado et al., 2012). Stress at these micro-calcifications did not exceed 275 kPa
(under the 300Kpa cap rupture threshold); this suggested that the location of the calcification has the largest effect on
rupture vulnerability through cavitation and interfacial debonding. The variations in plaque morphology and inflammatory
markers that may lead to rupture and erosion of plaques (Campbell et al., 2014) were analysed through a two-dimensional
finite element model based on data from in vitro samples. Inflammatory markers were correlated to increasing strain and
thin cap fibroatheroma in ruptured plaque, however, no correlations were seen in plaque erosion, suggesting the stability of
plaques, mainly thick-cap fibroatheromas, undergoing erosion. The variation in artery geometry from remodelling has still
been minimally studied despite suggestions that positive remodelling can increase plaque rupture vulnerability caused by
calcification inclusions (Mangalaprakash et al., 2013).
To develop an understanding of the failure mechanisms during plaque rupture, a number of crack propagation and dam-
age models were presented (Abdelali et al., 2014; Ferrara & Pandolfi, 2008; Karimi, Rahmati, Sera, Kudo, & Navidbakhsh,
2017a). The fissuring of atherosclerotic plaques in a three-layered, two-fibre anisotropic and hyperelastic artery caused
by overexpansion through a cohesive theory of fracture showed that fractures propagated from the shoulders of plaques
(Ferrara & Pandolfi, 2008). Under the buckling of the plaque cap however, maximum stress shifted from the plaque shoulders
to the cap centre (Abdelali et al., 2014). This result was, however, obtained from two-dimensional, quasistatic modelling of
the Navier–Bresse deformation formulas; previous studies already suggested this underestimated stresses, hence this result
should be validated in three dimensional models. Cracks propagated from plaque shoulders in the radial direction exposing
the media to blood flow shown in Fig. 15, adapted from Ref. Ferrara and Pandolfi (2008); a process which leads to platelet
aggregation and thrombus formation. Cohesive fracture models, unlike linear elastic fracture mechanics theories, can anal-
yse the response of uncracked material while avoiding stress singularities at the crack tip (Sun & Jin, 2012). Cap buckling
resulted in delaminations between the plaque cap and lipid core, with the addition of heterogeneities such as calcifications
increasing the sensitivity to buckling due to partial de-cohesion between components; this may assist in explaining plaque
rupture at areas other than peak cap stresses, an assumption validated theoretically (Cardoso et al., 2014). Increased pressure
also led to cracking of the healthy artery tissue before an eventual decrease in stiffness and a loss of the circular shape of
the artery.
To begin applying fibre-orientation parameters to coronary artery modelling, van der Horst, van den Broek, van de Vosse,
and Rutten (2012) proposed a new constraint for a simplified two-fibre, one-dimensional, single layered, thick-walled artery
with hyperelastic material to assess how fibre orientation effected stress. Parameters were estimated based on both inflation
and extension tests of the left anterior descending artery which resulted in fibre orientation of 36.4±0.3°; a major deter-
minant of stress-strain behaviour. Fibre strength effected crack propagation direction with high strength fibres leading to
diagonal propagation (along average fibre direction) and weaker fibres resulting in longitudinal crack propagation along the
24 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

Fig. 15. Evolution of crack patterns and von Mises stress in MPa at 260 mmHg blood pressure (a) and 380 mmHg (b) (Ferrara & Pandolfi, 2008). Reprinted
by permission of the publisher (Taylor & Francis Ltd, http://www.tandfonline.com).

artery length. This result suggested that collagen remodelling was modulated by principal stress (van der Horst et al., 2012)
as the optimal fibre angle for a fibre wound tube was 35°; as a two-fibre model, the result was a net result and hence
cannot represent actual in vivo morphology. The use of an artificial neural network to better predict collagen alignment was
proposed by Karimi et al. (2017a); both healthy and atherosclerotic arteries (Ooi, Sutcliffe, Davenport, & Maguire, 2014) were
assessed under uniaxial testing with material model parameters determined from a nonlinear minimisation/interior point
optimisation algorithm. At low axial loading, the collagen fibre angle in the healthy artery closely resembled the results
found in van der Horst et al. (2012), however, under increasing axial strains, as shown in Fig. 16(a), the fibre angle dra-
matically increased, with similar, although more steady, results seen for circumferential strain, as shown in Fig. 16(b). The
highly anisotropic behaviour and higher stiffness seen in atherosclerotic arteries was consistent with clinical histological
and microstructure data and the constitutive damage model has promising applications in modelling elastin and collagen
alterations seen during artery remodelling and interventional procedures.
The development of structural biomechanical models of coronary arteries from in vivo imaging technology has high-
lighted the importance of many factors. Accurate image co-registration and reconstruction can be undertaken through vari-
ous techniques but must include accurate catheter pullback speeds and account for out of plane motion to attain accurate
stress and strain results. Two and three-dimensional model comparisons again highlighted the importance of considering
the three-dimensional artery morphology with particular attention paid to residual stress and micro-calcification inclusions;
with the latter causing up to a five-fold increase in stress, although this was limited to a circular geometry and isotropic and
homogenous material without fluid shear acting on the material. Damage models and crack propagation showed differences
in failure modes between cap buckling and plaque rupture with the fibre orientation direction and strength shown to cause
variation in maximum stress and fracture propagation direction; collagen waviness, gradual activation and smooth muscle
cell contraction were not considered but are important parameters and may effect results. Furthermore, despite suggestions
that positive remodelling can increase plaque rupture vulnerability caused by calcification, the growth and/or remodelling
of arteries is yet to be taken into account. Table 6 in Appendix A summarises these combined theoretical and experimental
studies analysing, in vitro, the structural components of coronary arteries and plaques.

Fluid analysis. With limited ability to model the complex nature of the coronary vasculature, many early studies fo-
cused on using vascular casts to assess velocity profile relations to shear stress, intimal thickness and platelet deposition
(Bluestein, Niu, Schoephoerster, & Dewanjee, 1997; Perktold et al., 1997). As the media-intima thickness ratio correlated al-
most directly to the intima thickness alone, it was suggested that the artery wall responds to oscillatory shear rates to a
greater extent than purely shear stress magnitude; thicker intima areas, those relating to lower shear rates, could show the
first initiation and progression of atherosclerosis. Platelet deposition was shown to have strong correlation to wall shear
stress (Bluestein et al., 1997) with their activation potential increasing at regions of high shear stress through the stenosis
throat, suggesting elongation and compressive stresses distal to stenosis (Purvis Jr & Giorgio, 1991) that resulted in increased
activation potential and ligand binding receptivity, as shown in Fig. 17. Finite element analysis using the Galerkin method
also showed increased platelet deposition at regions of flow recirculation and reattachment, highlighting potential rheology
parameters that lead to vulnerable areas for atherosclerosis initiation. With the same assumptions of incompressible and
Newtonian flow, one of the more atheroprone regions, artery bifurcations, was shown to exhibit both flow and shear rever-
sal with artery curvature, further amplifying this finding (Perktold et al., 1997). The most significant flow deviations were
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 25

Fig. 16. Mean collagen fibre angle during axial load bearing vs strain (a) and circumferential load bearing vs strain (b) for both healthy and atherosclerotic
arteries predicted by the artificial neural network algorithm (Karimi et al., 2017). Reprinted with permission of John Wiley & Sons, Inc. (For interpretation
of the references to colour in this figure legend, the reader is referred to the web version of this article.)

seen at times of low flow rate, such as during diastole cycle, with the highest shear stresses occurring at the epicardial wall.
At regions usually prone to lesion formation wall shear stress, both oscillating and time-averaged, was decreased, further
agreeing with findings from earlier studies that investigated the relationship between shear stress and atherogenesis. With
more complex geometries resulting in smoother flow fields, it was also suggested that vascular remodelling and adaptation
occurs to minimise these effects.
With technological advancements the effect of flow patterns, such as flow reversal, on shear stress (Brunette, Mongrain,
Laurier, Galaz, & Tardif, 2008), artery curvature and stiffness can be more accurately assessed (Ai et al., 2010; Ji, Kobayashi,
Morikawa, Tang, & Ku, 2010). A micro-electro-mechanical sensor (MEMS) was used with pulse wave Doppler ultrasound
to measure in real time the flow reversals induced by stenosis (Ai et al., 2010) in a three-dimensional model simulating
an arterial vessel. The convective heat transfer was investigated as a means of assessing stenosis throat flow rates; results
26 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

Fig. 17. Platelet activation level over one streamline for Re = 3600 in both stenosis and straight tube models (Bluestein et al., 1997). Reprinted by permission
from Springer Nature.

Fig. 18. CFD blood flow simulation showing recirculation in (a) carotid bifurcation with minor aneurysm, (b) an idealised left main artery bifurcation
(Martorell et al., 2014). Reprinted by permission of Oxford University Press.

showed it increased across the stenosis before decreasing downstream. Flow reversal was detected using a high frequency
ultrasound MEMS at 45 MHz, demonstrating the potential for real time flow reversal monitoring and, hence, potentially wall
shear stress monitoring. This has particular clinical application for assessing high risk patients with inflammatory biomark-
ers directly correlated to flow recirculation in certain vascular geometries such as bifurcations, as shown in Fig. 18, adapted
from Ref. Martorell et al. (2014). Furthermore it was shown that the size of the flow recirculation area was directly cor-
related with monocyte adhesion to endothelial cells as well as uptake in Oxidised-low-density lipoprotein, increasing the
risk of thrombosis and myocardial infarction; three-dimensional motion produced by the cardiac and respiratory cycle were
not, however, considered in these studies which could affect wall shear stress results. Table 7 in Appendix A summarises
these combined theoretical and experimental studies analysing, in vitro, the fluid components of coronary arteries and
plaques.

Fluid–Structure Interaction (FSI) Analysis. There were limited studies combining fluid–structure interaction analysis with in
vitro testing for human coronary arteries. Two to three dimensional modelling from in vitro experimental/clinical studies has
been investigated (Tang et al., 2004, 2005; Zheng et al., 2005); three-dimensional, finite element fluid–structure interaction
models based on MRI models of unsteady and multi-component plaques were used to assess the critical flow conditions
and mechanical behaviour under varying pressure distributions (Tang et al., 20 04, 20 05). Isotropic, incompressible and ho-
mogeneous material properties in a hyperelastic, Mooney-Rivilin model coupled with laminar, Newtonian, viscous and in-
compressible blood resulted in increased lipid pool size being associated with maximum stretch while decreased fibrous cap
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 27

Fig. 19. Maximum and minimum stress and strain distributions for a straight (a-b) and curved (c-d) artery segment at the same internal pressure and
cyclic flex with position of cut shown. Reprinted from (Tang et al., 2009, with permission from ASME.).

thickness leading to minimum stress-strain levels; up to 300% greater stress and strain concentration was seen due to a thin
plaque cap. When compared to otherwise healthy patients, patients with coronary artery disease (Huang et al., 2014) saw
an average increase of 93.4% in the critical plaque wall stress (at the point of maximum plaque wall stress), suggesting it
as a potentially statistically significant marker for plaque vulnerability assessment. Whilst fatigue effects were not assessed,
the cyclic loading caused by the pulsatile blood flow led to corresponding cyclic stress-strain which could lead to fatigue
failure; as stiffness is seen to influence the maximum stress, the relation between stiffness and fatigue failure could prove
useful for rupture assessment.
Links between plaque cap erosion and increasing maximum strain have also been made, however, it was noted that
erosion did not cause variation in stress. Using the same assumptions as Tang et al. (2005), a 2D FSI model to gradually
pressurise arteries, introduced by Zheng et al. (2005), attempted to locate the point of maximum stress and strain in plaque
vulnerability scores. No consistency of location was found despite maximum stress and strain occurring mostly adjacent
to lipid pools and at the shoulders of plaque. Despite this, biomechanical marker scoring for plaque vulnerability matched
pathologic scores from the scale in 86% of cases, which could assist in the development of a more comprehensive biome-
chanical marker of plaque vulnerability. The two-dimensional cross-sections used by Zheng et al. (2005) and the use of
straight artery sections in three-dimensional modelling (Tang et al., 2004; Tang et al., 2005) did not consider the artery
curvature and cyclic loading experienced from cardiac motion; these are important aspects in the biomechanical modelling
of coronary arteries. The incorporation of anisotropic material properties with cyclic bending of the artery resulted in a 50–
800% maximum-principal-stress increase and an 80–185% increase in maximum strain (Tang et al., 2009). Despite a lower
pressure, Fig. 19 showed that the curvature of the artery increased stress 360%, suggesting the importance, and potential,
of morphological features to predict vulnerable regions. With the inclusion of cyclic flex, flow shear stress increased 15%
for only a 9.8% increase in flow velocity which could lead to peak wall stress exceeding the ultimate material strength and
initiating plaque rupture.
Artery wall thickening is often associated with atherosclerosis progression; Kural et al. (2012) investigated the diseased
state of coronary arteries through comparison of uniaxial and biaxial tensile tests with results suggesting thickening is
associated with increased stiffness and anisotropy direction with the neo-intimal layer playing a major role in this anisotropy
and overall stiffness (Holzapfel et al., 2005). Biaxial testing results showed a significant reduction in both stress and strain
compared to uniaxial testing, with a shift of the location in maximum stress from the thin vessel wall region to the area of
thickening containing calcified plaques and lipid pools. Three different constitutive models previously used were compared;
the Fung-Type (Chuong & Fung, 1983; Fung, 1991):
1
 2 2 2

W = C exp b1 ERR + b2 E + b3 EZZ + 2b4 ERR E + 2b5 EZZ E + 2b6 ERR EZZ + b7 ER2 + b8 E
2 2
Z + b9 ERZ − 1 , (2.19)
2
where Ei j (i, j = R, θ , Z ) are the Green strain components, R, θ , Z are the cylindrical coordinates while C, b1 , . . . b9 define
material parameters; only the circumferential and axial directions were considered in this study, hence b1 , b4 , b6−9 = 0.
28 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

Fig. 20. Collagen fibres and their orientation (shown in blue) for a dissected fibrous cap specimen (Pagiatakis et al., 2015). Reprinted by permission from
Springer Nature. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Choi and Vito (1990) outlined as



ψ = C ( eQ1 + eQ2 + eQ3 − 3 ),
, (2.20)
2
Q1 = bz EZZ , Q2 = bθ Eθ2θ , Q3 = 2bθ z EZZ Eθ θ

and a modified Mooney–Rivilin model similar to that outlined in Eq. (2.8) in Section 2.1.1; using the finite element analysis
with a 3D model constructed from IVUS images from a single patient, the Choi-Vito model was found to show the best
agreement to experimental data (Kural et al., 2012), although the fits were generally better for carotid arteries than for
coronary ones. This may have been due to the more complex interactions associated with coronary arteries including heart
motion and contractile functions, whereas the carotid artery was less impacted by such interactions; multiaxial constitutive
relations are needed for accurate comparison to patient specific data. Furthermore the principal stress orientation was found
to be axially orientated at peak stenosis levels which was found to correlate well to collagen alignment, as shown in Fig. 20,
adapted from Pagiatakis, Galaz, Tardif, and Mongrain (2015). This could lead to weakening in the circumferential direction
and eventually the failure and rupture of plaque as the fibres remodel and align themselves with the principal stress in
the artery, although for a more accurate assessment the inclusion of residual stress and the shear-thinning nature of blood
should be considered.
In conclusion, combining experimental/clinical investigations with theoretical modelling provided new insight into the
biomechanics of coronary arteries. Combining multiple parameters, such as anisotropic material parameters and cyclic bend-
ing, resulted in large variations in stress and strain responses highlighting the nonlinear nature of biomechanical analysis
and showing the need to consider all parameters in modelling. Plaque cap erosion was considered for the first time and fur-
ther variations in material models for soft tissue were used and compared against uniaxial and biaxial tensile testing which
resulted in differences in maximum stress magnitude and location in biomechanical models. This also shows the importance
of obtaining accurate material data for use in theoretical modelling. Studies in the preceding section do not however, assess
the characteristics of non-Newtonian blood which is noted as an important modelling parameter and the use of histology for
input of morphology and material properties in models could also limit the accuracy of theoretical models; hence, obtaining
imaging and data from in vivo is required. Table 8 in Appendix A summarises these combined theoretical and experimental
studies of coronary arteries and plaques using fluid–structure interaction analysis, in vitro.

2.3.2. In vivo theoretical and experimental/clinical studies


In vivo studies have been undertaken with the properties and data used in experimental modelling taken from living hu-
man patients. Intravascular Ultrasound (IVUS), Fractional Flow Reserve (FFR), X-ray angiography, Magnetic Resonance Imag-
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 29

Fig. 21. Flexing in the left anterior descending artery over two cardiac cycles (Gross & Friedman, 1998). Reprinted with permission from Elsevier.

ing (MRI), and Optical Coherence Tomography (OCT) were predominantly used for the re-construction and development
of accurate, three-dimensional, patient specific models that targeted the gold standard for the prediction of atheroscle-
rotic initiation, plaque rupture and myocardial infarction. Overviews of these studies can be found in Tables 9–11 in
Appendix A.

Structural analysis. The motion and geometry of arteries have been hypothesised as biological markers for the initiation and
progression of atherosclerosis (Ding & Friedman, 20 0 0; Gross & Friedman, 1998). By using biplane angiogram images ob-
tained from the left anterior descending arteries, a time interpolation technique was used to co-register and reconstruct the
two dimensional images into a three-dimensional model of the artery to assess artery mechanics (Gross & Friedman, 1998).
The dynamic changes in artery curvature, known as artery flex, as shown in Fig. 21 over two cardiac cycles, were sug-
gested to be more influential than artery curvature itself. To determine if artery motion, such as flex, could be a con-
tributing factor to atherosclerosis, Ding and Friedman (20 0 0) developed a system to track displacement, strain, torsion and
curvature in vivo through de-warping, matching templates and time interpolation to reconstruct three-dimensional arterial
models. The left anterior descending artery (compared to the right coronary artery) experienced twice the torsion mag-
nitude, increased variability and less displacement. With decreased displacement, torsion and flex were seen to be more
influential, possibly producing atherogenic conditions. Further parameters such as axial torsion and radial expansion are
noted to also be potentially biomechanically important, however, these were not able to be characterised in these early
studies.
To better optimise the morphology of the artery and plaques and enhance the determination of Young’s modulus for each
section, Le Floc’h et al. (2009) generated modulograms, such as those adapted from Baldewsing et al. (2008) as shown in
Fig. 22, and applied them to seven in vivo IVUS morphologies. A preconditioning model was used to asses Young’s modulus
for each section of the plaque morphology including the necrotic core, fibrosis and calcification resulting in mean errors
of 32%, 2% and 43% respectively. Peak cap stress showed the largest variations with changes in the fibrous cap thickness
and the necrotic core’s Young’s modulus, potentially demonstrating the effect of small structural changes on the stabil-
ity of plaque. The significant variation could also suggest the relative insignificance of the cap thickness on peak stress
compared to other parameters such as the necrotic core or calcification inclusions. Whilst this method detected calcified
deposits accurately, elasticity was found to be underestimated by a factor of two, potentially due to catheter position which
caused an overestimation of cap thickness by 58% when located close to the necrotic core. Whilst limitations such as two-
dimensional analysis, static loading conditions and the omission of pulsatile pressure/flow, viscoelastic effects, anisotropic
30 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

Fig. 22. (a) Echogram. (b) Corrected elastogram. (c) Reconstructed compounded modulogram from (b) (Baldewsing et al., 2008). Reprinted, with permission,
from IEEE Transactions on Information Technology in Biomedicine.

material and residual stresses were discussed, perhaps a more prominent limitation was rotational tissue misalignment in
IVUS imaging. This rotation could have invalidated strain elastograms due to significant drops in signal correlation with
small tissue misalignment caused by rotation of the artery during IVUS imaging (Danilouchkine, Mastik, & van der Steen,
2009).
Plaque morphological features have been used as a measure of vulnerability in the literature, with two-dimensional finite
element models used to determine stress distributions throughout cross-sections from in vivo images (Costopoulos et al.,
2017; Ohayon et al., 2008). Using the ultimate tensile stress threshold of 300 kPa from the literature, Ohayon et al. (2008)
showed that increased necrotic core size and decreased cap thickness increased the overall peak cap stress; this was the
case both with and without the plane strain assumption often used in two-dimensional analysis; cap stress magnitude varied
only 25% under this assumption. Relatively small plaque stenosis with large necrotic cores were found to exhibit high rupture
vulnerability, in agreement with clinical evidence, however assuming that only the necrotic core was a major factor in low
stenosis rupture was not possible without assessing the influence of parameters such as calcification, pulsatile pressure
from physiological blood flow and omitted residual stresses. Furthermore it was suggested that high imaging resolution
(<65 μm) is critical to detecting the crucial morphology that influences rupture vulnerability; the criticality of high imaging
resolution is further outlined when determining the effect of micro-calcifications on stress intensification, with average size
just 28±13 μm in later studies (Maldonado et al., 2012). The relationship with increasing necrotic core size and increased
plaque structural stress was also shown by Costopoulos et al. (2017) through the use of both VH-IVUS and OCT for more
accurate finite element morphology representations, including calcification, and the investigation of previously ruptured
sites; it was found the dense calcification actually reduced stress when it represented more that 10% of the cross-section in
their hyperleastic and isotropic material model. A reduction in fibrous and fibrofatty tissue coupled with a structural stress
greater than 135 kPa was recognised as a good predictor of rupture in higher risk regions. This stress value was less than
half of the previously recognised critical threshold of 300 kPa, possibly due to the more accurate and complex morphology
of fatigue seen in previously ruptured sites; although the noted overestimation of stress by two-dimensional artery models
(Ohayon et al., 2005) further amplifies this difference.
The comparison of two and three dimensional finite element models (Ohayon et al., 2005) reconstructed from IVUS
imaging resulted in not only the magnitude of stress being overestimated by two-dimensional modelling, but also the loca-
tion of peak circumferential stress could not be accurately determined unless three-dimensional modelling was used which
is of particular importance for determination of rupture sites. The linear elastic and transversely isotropic model, with ra-
dial and circumferential isotropy and varying axial properties, assumed no residual stresses and only considered a dual
layered artery; for comparison Fig. 23, adapted from Alastrué, Peña, Martínez, and Doblaré (2007a), highlighted the influ-
ence of residual stress on the circumferential stress across a dual layered artery, particularly the discontinuity across the
media-adventitia layer boundary. A similar dual layered, left main artery with the omission of the intima layer was used
to assess the effects of myocardial contraction and stiffness distributions throughout the arteries bifurcation (Ohayon et al.,
2011). Stiffening was thought to be critical in atherosclerosis formation in this region; MRI reconstruction of a nonlinear
finite element model with hyperelastic and anisotropic material compared against known atherosclerotic formation in pa-
tients resulted in high luminal wall stretch and stiffness directly correlating to plaque sites (34.7±1.6% and 442.4±113.0 kPa
respectively). Artery stiffness and mechanical properties were found to have an almost linear relationship across previous
property ranges (Holzapfel et al., 2005).
The theoretical modelling of coronary artery structures using in vivo data is a step towards developing an accurate
biomechanical model for real-time analysis. By using patient-specific data, results suggested that artery dynamics, partic-
ularly the flex and torsion are more important than pure geometry, such as artery curvature. Obtaining data from in vivo
conditions presented challenges such as the importance of understanding catheter position during imaging which led to
overestimation in morphological features, such as plaque cap thickness, by over 50%; resolution was also important, espe-
cially when one of the largest stress intensifiers, micro-calcifications, averaged less than 30 μm in size. A number of im-
portant features such as the effect of viscoelastic material models, microstructure alignment and the inclusion of residual
stresses are still to be included in many studies. Coronary arteries are almost always simplified to a two-layered model, often
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 31

Fig. 23. Circumferential stress variations across the artery wall thickness for steady internal pressure with the applied residual stresses and the pressure
only case without residual stress applied for varying artery radii (Alastrué et al., 2007). Reprinted by permission from Springer Nature.

neglecting the intima layer; modelling of all three layers of coronary arteries is an important consideration and should be
included for accurate biomechanical analysis of coronary biomechanics in the future. Table 9 in Appendix A summarises
these combined theoretical and experimental studies analysing, in vivo, the structural components of coronary arteries and
plaques.

Fluid analysis. Early fluid specific investigations into the coronary vasculature based on in vivo experimental data were
focused around the pulsatile physiological patterns of blood flow (Zeng, Ding, Friedman, & Ethier, 2003). Modelling blood
as a non-Newtonian fluid based on the Carreau model in the left anterior descending artery’s bifurcation showed abrupt
velocity changes, flow reversal and a decrease in wall shear stress on wall’s opposite the bifurcation apex when flow was
modelled as a non-Newtonian Carreau fluid; this supported clinical evidence for the outer wall being a known atheroprone
site. Shear stress oscillated with the acceleration and deceleration of flow and was particularly prominent when flow was
decelerating in the pulsatile flow profiles. Despite the use of a viscous Hagen-Poiseuille flow model, Zeng et al. (2003)
further highlighted the variation caused by pulsatile flow in a moving right coronary artery re-constructed from biplane
angiogram imaging. When comparing the impact of pulsatile flow and artery motion on wall shear stress at the inner
artery wall, pulsatile flow was seen to dominate which is highlighted in Fig. 24, however, the same did not hold at the
outer wall. Here the wall shear stress results were comparable between steady flow in a dynamic artery and pulsatile
flow in a static artery, as outlined in Fig. 24(d); no correlation was drawn here due to the assumption of purely global
displacement and no artery deformation or torsion; a relationship that could only be identified through fluid–structure
interaction.
According to the literature, the parameter of most clinical importance in atherosclerosis initiation, progression and my-
ocardial infarction is that of wall shear stress, even more evident through the large number of studies (Chaichana, Sun,
& Jewkes, 2011; García, Crespo, Goicolea, Sanmartín, & García, 2006; Moreno & Bhaganagar, 2013; Soulis, Farmakis, Gi-
annoglou, & Louridas, 2006; Soulis, Giannoglou, Parcharidis, & Louridas, 2007; Timmins et al., 2015) focused on developing
an understanding of its implications, magnitude and distribution in the coronary artery vasculature. Low wall shear stress
was associated with artery wall thickening and impacted endothelial cell regulation in atheroprone regions, such as bifurca-
tion (Soulis et al., 2006); thickening in areas of plaque instability was also associated with growth amplifying atherosclerosis
progression. Leucocyte migration and coagulation was caused by low wall shear stress (Soulis et al., 2007) as well as a signif-
icant increase in low-density-lipoproteins (correlation as shown in Fig. 25) hence, identification of the transport and concen-
tration of micro-constituents in blood is important in assessing atherosclerosis initiation and progression (Soulis, Fytanidis,
Papaioannou, & Giannoglou, 2010). Local hemodynamic variations caused by varied wall shear stress were also shown to in-
duce additional dissipation across stenosis leading to Fractional Flow Reserve (FFR) alterations of as much as 80% in clinical
settings (Melchionna et al., 2013) due to red blood cell size and its globular nature. Comparison of the effects of red blood
cell aggregation and deformability (Zhang & Kuang, 20 0 0), haematocrit and Newtonian or Non-Newtonian blood models
(Johnston, Johnston, Corney, & Kilpatrick, 2004; Johnston, Johnston, Corney, & Kilpatrick, 2006; Katranas, Antoniadis, Kelekis,
& Giannoglou, 2015) through pulsatile flow were also assessed. Many studies using Newtonian fluid models (Chaichana et al.,
2011; García et al., 2006; Moreno & Bhaganagar, 2013; Prosi, Perktold, Ding, & Friedman, 2004; Timmins et al., 2015) jus-
tified their assumption by citing fluid shear stresses being above the threshold for viscous shear thinning to have effect,
32 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

Fig. 24. Wall shear stress (WSS) at the inner (a, b) and outer (c, d) walls along the artery length for normalised time over one cardiac cycle (Zeng et al.,
2003). Reprinted by permission from Springer Nature.

Fig. 25. Normalised lumen surface low-density-lipoprotein concentration vs wall shear stress (Soulis et al., 2010). Reprinted with permission from Elsevier.
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 33

Fig. 26. Wall shear stress at low inlet velocity of 0.02 m/s over the artery curvature length (Johnston et al., 2004). Reprinted with permission from Elsevier.

although considering blood as non-Newtonian is important as micro-constituent blood inclusions, such as haematocrit, can
significantly affect viscosity and potentially the red blood cell aggregative strength (Zhang & Kuang, 20 0 0). To highlight
the variances between Newtonian/non-Newtonian blood models, both steady state (Johnston et al., 2004) and pulsatile flow
models (Johnston et al., 2006) were developed from right coronary artery biplane angiogram images, ignoring side branches,
from centreline and radius extraction methods. With five different non-Newtonian models assessed, including the previously
outlined Carreau (Eq. (2.13)) and Power law (Eq. (2.14)) as well as the Walburn-Schneck, Casson and Generalised Power Law
models defined by Eqs. (2.21), (2.22) and (2.23), respectively:

2
μ = C1 eC2 H eC4 (T PMA/H ) (γ˙ )−C3 H , (2.21)

⎧ * 2  + ⎫
⎪ 1/4 −1/2 1/2 −1/2
 ⎪
⎨μ = (n J2 ) + 2 τy
⎪ ⎪
2
J2 ⎬
with  , (2.22)
⎪  ⎪
⎩ |γ˙ | = 2 J , τ =
0.1(0.625H )3 , and η = η (1 − H )−2.5⎪
⎪ ⎭
2 y 0

  ⎫
η = λ |γ˙ |n−1 , where ⎪



 ,   −b - ⎪

( ) ⎪

|γ˙ |
− 1+ a e |γ˙ | ⎬
λ(γ˙ ) = μ∞ +
μ e . (2.23)
 ⎪
- ⎪
,  
( −d ) ⎪⎪
γ˙
− 1+ | c | e |γ˙ | ⎪

n(γ˙ ) = μ∞ +
μ e ⎪

At low wall shear stress values, the Generalised Power law showed the strongest fit when compared against Newto-
nian fluids which is highlighted in Fig. 26, however, at medium to high shear stresses there was little variance between
34 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

Fig. 27. Stenosis induced by shear stress and the thickness of the artery over time for both physiological ad reduced shear stress (Rodríguez et al., 2007).
Reprinted with permission from Elsevier.

non-Newtonian and Newtonian fluids (Johnston et al., 2004). At low fluid velocities, WSS was underestimated by the
Newtonian model and overestimated by the Power law model, however, when a pulsatile velocity profile was introduced
(Johnston et al., 2006) the non-Newtonian fluid model dominated approximately 30% of the cardiac cycle, with the prox-
imal end of the artery the primary low wall shear stress location; the Newtonian model predicted results well over the
remaining time. With atherosclerosis initiation and progression linked to low wall shear stress regions, the use of non-
Newtonian fluid models are crucial to better understand the micro-constituent particle interaction and biomechanics of
disease.
Accurate wall shear stress calculations are critical for the understanding and future assessment of disease initiation and
progression, hence assessment of artery model re-construction accuracy was needed. Reconstruction of angiogram based,
finite element CFD models using the Galerkin technique (Berthier, Bouzerar, & Legallais, 2002) showed that the incorrect
setup of boundary conditions, such as a lack of mesh refinement at critical locations, led to wall shear stress being halved
with up to a quadrupling of flow velocity between simplified and patient specific modelling. These results, however, are
not likely to hold for fluid–structure interaction analysis with artery wall deformation smoothing velocity profiles. In an
effort to improve the accuracy of models by including parameters that were otherwise overlooked due to processing sen-
sitivity (Rivolo et al., 2014), Multi-slice Computer Tomography (MSCT) and three-dimensional Angiography were fused with
IVUS (Gijsen et al., 2014) and OCT (Li et al., 2015), respectively. Artery centrelines were extracted from MSCT (with spa-
tial resolution 0.32 × 0.32 × 0.5 mm) and IVUS images then registered to cross sectional reconstructions perpendicular to the
artery centreline and stacked to rebuild the 3D geometry. The lumen boundary for both main artery and side branches
was manually delineated and merged with the first step by scaling the MSCT contour images to IVUS which resulted in
the co-localisation of 75.6±7.7% of the wall shear stress zones; three-dimensional angiogram imaging was fused with OCT
by using artery side branches as markers for longitudinal and rotational image matching. Accounting for outflow pressure
conditions resulted in shear stress variation from -60.71 to 7.47 Pascals in the true model, overestimation in 61.7% and un-
derestimation in 9.9% of cases with a significant underestimation in the distal coronary pressure to aortic pressure in the
classical models without side branch outflow; uncertainty in clinical measurements resulted in the average errors of 28.4%
when compared to literature and 4.6% when compared to patient specific results for both oscillatory and time averaged
shear index and shear stress (Tran, Schiavazzi, Ramachandra, Kahn, & Marsden, 2017), well within expected uncertainties.
Table 10 in Appendix A summarises these combined theoretical and experimental studies analysing, in vivo, the fluid com-
ponents of coronary arteries and plaques.

Fluid–structure interaction (FSI) analysis. There are limited studies that developed FSI models from in vivo clinical data,
possibly due to the difficulty in reconstructing accurate models from imaging modalities such as IVUS (Asanuma et al., 2013;
Guo et al., 2017; Wang et al., 2015) and angiography (Dong, Sun, Inthavong, & Tu, 2015; Liu et al., 2017; Wu et al., 2018).
With the remodelling of coronary arties seen as a major preceding factor to atherosclerosis, the inclusion of local shear
forces acting on the anisotropic intima layer is seen to effect the artery remodelling and thickness (Rodríguez, Goicolea, &
Gabaldón, 2007). Using the finite element method in a 3D, nonlinear artery with the inclusion of anisotropic collagen fibres
in a hyperelastic material, the inclusion of the local shear stress factors caused by blood flow resulted in the reduction
of the artery radius due to negative growth; at shear stresses below the equilibrium point the equivalent stenosis of 64%
was reached approximately 400 days after shear equilibrium was reached. In the first few days of remodelling a reduction
in the artery thickness was seen, as shown in Fig. 27, which may be contributed to by balancing stresses between the
adventitia and media layer; increased von Mises stress was also later found to lead to positive (expansive) remodelling
(Liu et al., 2017). Whilst shear stresses were shown to influence artery remodelling, previous studies suggested its effects
were minimal when compared to structural stresses. Asanuma et al. (2013) showed that these tissue stresses could be 10 0 0
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 35

Fig. 28. Comparison of FSI and rigid artery models based on CT angiogram images with the wall shear stress at peak diastole (a) and wall shear stress
variation over the cardiac cycle (b) (Dong et al., 2015). Reprinted by permission of the publisher (Taylor & Francis Ltd, http://www.tandfonline.com).

to 10,0 0 0 times larger than shear stresses for von Mises and principal stresses respectively, suggesting the importance of
shear stress in the promotion and progression of atherosclerosis, but its limited affects in plaque rupture; this was, however,
obtained from a single layered artery without the inclusion of any cardiac motion or cyclic bending.
The inclusion of artery motion caused by the cardiac cycle was shown to significantly alter stress and strain results
for the artery walls as well as flow characteristics inside the artery (Fan et al., 2014; Yang et al., 2009). To more accu-
rately model arteries from in vivo IVUS and angiogram imaging, Yang et al. (2009) included the artery cyclic bending and
axial stretch whilst comparing the effects of both anisotropic and isotropic material property assumptions. Through X-ray
angiogram movies, the motion of the myocardium was plotted to allow for the inclusion of cyclic bending in the fluid–
structure interaction model. Whilst an important step, the model was limited to two-dimensional cyclic motion, hence a
second imaging plane should be used to couple out of plane motion for a fully three-dimensional analysis. A more recent
study by Wu et al. (2018) proposed a one-way fluid–structure interaction model based on a ‘global displacement method’
to reduce computation time by directly inputting artery deformation, however one-way coupling would result in important
interactions between the fluid and artery wall being missed. Results did however highlight the importance of including
cyclic bending, with up to a 102% increase in maximum strain compared to no artery bending; the flow in the artery was
36 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

Fig. 29. The FSI model analysis process from reconstruction of the right coronary artery using coronary CT Angiography to assess the artery lumen surface
(Liu et al., 2017). Reprinted by permission of the publisher (Taylor & Francis Ltd, http://www.tandfonline.com).
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 37

only mildly affected with a 15% decrease in flow maximum shear stress and a 5% decrease in maximum velocity, results
that Fan et al. (2014) corroborated with mechanical data obtained from ex vivo tensile tests after imaging, albeit with much
smaller variations in stress and strain. The greatest variances were seen when comparing isotropic and anisotropic material
parameters; anisotropic models produced 63% higher maximum principal stress compared to isotropic models. When cou-
pled with cyclic bending and then also axial stretch, stress was further increased to 126 and 345% of the maximum principal
stress, highlighting the importance of accurate material parameters, however, the inclusion of atherosclerotic progression in
the arteries has also not been considered.
The use of VH-IVUS was used to re-create a three-dimensional, nonlinear, anisotropic artery based on the finite element
Mooney-Rivilin model with isotropic plaque constituents and Newtonian, incompressible flow (Guo et al., 2017; Wang et al.,
2015). With the inclusion of cyclic bending, it was attempted to investigate the progression of plaques based on increases
in the artery wall thickness, plaque cap thickness and lipid depth by correlating them with measured wall shear stress
and strain (Wang et al., 2015). Previously noted as a potential marker for atherosclerosis progression, wall shear stress
correlations were seen to vary more significantly than other markers, with plaque progression best predicted by the coupling
of plaque wall stress and the thickness of the artery. Plaque cap thickness and wall stress both showed a negative correlation
to wall shear stress, potentially leading to plaques with higher vulnerability rupturing by tensile wall stresses (Wang et al.,
2015). When comparing the mechanical data used in three-dimensional modelling, plaque cap strains were found to be 150-
180% higher and corresponding stress 50-75% lower when mechanical data from in vivo tests were used suggesting increased
artery stiffness in ex vivo segments altered results, further highlighting plaque sensitivity to artery stiffness (Guo et al.,
2017). In an effort to include residual stresses in the model, a 5% axial and circumferential shrinkage were accounted for in
the VH-IVUS model; exclusion of these shrinkages led to an overestimation of stress and strain and an overexpansion of the
artery past its in vivo dimensions.
Similar modelling from CT angiogram images was used to investigate bifurcation branch angle effects on the local hemo-
dynamic forces (Dong et al., 2015). The nine parameter Mooney-Rivilin model resulted in large variances in stress, particu-
larly at the apex of the bifurcation, with increasing bifurcation angle having a negative correlation with principal stress. With
increasing bifurcation angle, however, the stress at the Left Circumflex side of the bifurcation resulted in a positive correla-
tion to principal stress, lowered oscillatory shear stress and higher tensile stress. This supports clinical findings and shows
potential for assisting in clinical examination of atherosclerotic risk. Comparisons were made between the fluid–structure
interaction model and a rigid model from the same patient data, as shown in Fig. 28 adapted from Dong et al. (2015);
whilst generally similar trends were seen throughout the cardiac cycle, wall shear stress magnitudes were largely overes-
timated by the rigid artery, similar to what was previously modelled with computational fluid dynamics (Chaichana et al.,
2011; Soulis et al., 2006), highlighting the importance of fully utilising fluid–structure interaction capabilities. Building on
the capabilities of CT angiogram modelling, Liu et al. (2017) developed an elastic model of the right coronary artery based
on patients who had acute coronary syndrome, as shown in Fig. 29, to predict the progression of plaques in clinical settings
over a 12 month period. In doing to, correlations were drawn between low baseline von-Mises stress and both decreased lu-
men area and decreased plaque and both low baseline wall shear stress and decreased minimal lumen area were correlated
to an increase in plaque burden.
Despite the increasing model complexity and accuracy from the re-constructed in vivo imaging, no current in vivo based
study incorporates the full three-layered coronary artery histology with intima, media and adventitia layers. The interaction
between the three layers, particularly the variation in the uptake of strains and the discontinuities in residual stresses was
shown to affect the resultant maximum stress and distribution of stresses throughout the artery. Axial and circumferential
shrinkage were outlined as important parameters by Guo et al. (2017) in their IVUS based study; axial shrinkage was esti-
mated at 5% and circumferential shrinkages were estimated based on the assumption that a diseased artery is stiffer than
a healthy one as no data were available to determine this. The variation seen in Dong et al. (2015) between the rigid and
compliant artery model could be further exacerbated by the late uptake of artery loading forces through the outer layers,
namely the adventitia and media, and further amplified by the media’s active contraction throughout the cardiac cycle. Cou-
pling of these model parameters with in vivo IVUS and OCT imaging shows the most promise for accurate, patient-specific
results. The use of OCT is also lacking in human in vivo modelling despite its higher resolution which is able to account for
the shortfalls in IVUS imaging resolution and could also assist in accurately determining the plaque compositions required
for more accurate stress distribution analysis.
Biomechanical modelling of coronary arteries has the potential to provide critical information on disease initiation and
progression that could lead to prediction of myocardial infarction. By coupling in vivo experimental/clinical investigations
with theoretical models, patient specific models can be developed capable of enhancing our knowledge of the coronary
artery vasculature. Whilst these investigations have provided insight into the important effects of plaque compositions,
residual stress, simplifications between modelling approaches and variations in in vivo imaging methodologies, the use of
non-Newtonian fluid models and viscoelastic material properties have been seldom applied; these are important both in the
modelling of physiologically accurate soft tissue and blood flow; in particular developing relations to assess the effect of
artery remodelling and plaque growth over time would prove useful in clinical settings. Table 11 in Appendix A summarises
these combined theoretical and experimental studies, in vivo, of coronary arteries and coronary plaques using fluid–structure
interaction.
38 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

3. Conclusions

In conclusion, biomechanical modelling of coronary arteries can discern the critical links between artery and fluid me-
chanics and disease initiation and progression that medical imaging technology cannot do. The highly nonlinear correlations
between key biomechanical factors and disease require the assessment of all parameters simultaneously in order to pro-
vide useful and accurate data to clinicians. This review provides an in-depth assessment of all factors of human coronary
biomechanics and artery modelling from both theoretical and experimental standpoints, aiming to present a comprehensive
overview of research efforts to date to assist in addressing the challenge of developing an accurate and reliable biome-
chanical model of coronary arteries. With cardiovascular diseases the largest cause of death and economic burden globally,
biomechanical analysis of coronary arteries can provide information to clinicians that medical imaging cannot; this infor-
mation could predict the initiation and progression of disease such as atherosclerosis, and eventually, predict myocardial
infarction in advance. Studies reviewed were classified based on their approach to modelling such as imaging modality
used, methodology, material and blood properties, type of coronary artery, approach to the study (theoretical, experimental,
in vitro & in vivo), number of patients/specimens and a general description, provided in Appendix A. Divided into purely
theoretical, purely experimental/clinical and a combined theoretical and experimental/clinical methodology and further di-
vided into structural, fluid and fluid–structure interaction (FSI) approaches, key implications for the biomechanical analysis
of coronary arteries were assessed.
Results highlighted that the complex interactions seen in the human coronary vasculature cannot be simplified as static,
purely fluid or structural models without compromising accuracy; hence, FSI approaches must be used in three-dimensional,
patient specific models in order to best replicate physiological conditions. By comparing against FSI based results, rigid artery
models overestimated the wall shear stresses, presenting a possible explanation for the assumptions of Newtonian fluid
models being applied due to shear stress magnitude not being low enough for non-Newtonian models to impact results.
Comparisons between two and three-dimensional artery models highlighted that under purely fluid analyses, shear stresses
were underestimated, however, in purely structural analysis, two-dimensional models overestimated peak circumferential
stress. Variations in the material and fluid models used resulted in differences in most results; comparisons between a New-
tonian fluid model with constant viscosity and a number of non-Newtonian models (Carreau, Power law, Walburn-Schnech,
Casson, generalised Power Law) resulted in the non-Newtonian shear-thinning effects being dominant over approximately
30% of the cardiac cycle with Newtonian fluids approximating the shear effects well over the remaining 70%. The gener-
alised Power Law suggested by Johnston et al. (2004) showed the best fit at low wall shear stress values while pulsatile
flow resulted in increased variation and intensity of all flow parameters compared to steady flow, hence, the physiological
nature of blood flow must be considered; as wall shear stress played an important part in disease progression, considering
non-Newtonian fluid is critical in furthering biomechanical analysis of coronary arteries.
A number of constitutive material models were utilised in biomechanical analysis of human coronary arteries. The most
prominently used was the hyperelastic Mooney-Rivilin model (see Eq. (2.8) in Section 2.1.1); however, the number of model
parameters varied considerably; uniaxial and biaxial tensile tests as well as pressure-inflation tests were prominently used
to determine these material parameters, however, many of these studies lacked consistent excision protocols, with vari-
ances in post mortem excision time, storage and testing temperatures, preconditioning procedures or mention of detailed
lifestyle factors, making comparison of the large variances seen in soft tissue data from the coronary vasculature difficult.
This was highlighted through direct comparison of biomechanical modelling results with in vivo data resulting in 150-180%
higher plaque cap strain and 50-75% lower stress (Guo et al., 2017) compared to in vitro data suggesting increased stiff-
ness in excised artery portions and high sensitivity of biomechanical modelling to stiffness. Holzapfel’s soft tissue strain
energy model (Holzapfel et al., 20 0 0) (Eq. (2.9) Section 2.1.1) included both anisotropic and isotropic non-collagenous ma-
trix behaviour and collagen fibre orientation which was shown to influence fracture propagation directions during plaque
rupture.
Although highlighted as important in soft tissue biomechanics, the effects of tissue viscoelasticity (see Eq. (2.16),
Section 2.1.3) were rarely considered; one of the most complete theoretical biomechanical artery models developed by
Gholipour et al. (2018b) highlighted the importance of considering all relevant physiological effects, critically including
three-dimensional heart motion, artery taper, media layer contraction and viscoelastic effects; this resulted in over 500%
variation in von Mises stress compared to simplified models. The intima in particular plays an essential role in disease ini-
tiation and progression and also in plaque cap stress, with intermediate to high intima stiffness resulting in a tripling of
plaque cap stress as the cap thickness decreased; the inclusion of low-density-lipoprotein transport, as a cause of artery
growth and remodelling, could assist in more accurately predicting the initiation of damage and then disease in the in-
tima layer. Furthermore the outermost layer, the adventitia, resulted in nonlinear uptake in lumen pressure, becoming the
major load bearer at higher pressure; this resulted in greatest artery damage occurring in this layer which led to plaque
redistribution through radial compression. Residual stresses, highlighted as causing critical variation in biomechanical artery
response, further showed the importance of including multiple artery layers with stress discontinuities occurring across
artery boundaries, as shown previously in Fig. 23 from Section 2.3.2 Structural analysis. Inclusion of both axial and circum-
ferential residual stresses resulted in a 43% decrease in the fibrous cap’s maximum principal stress when compared to only
considering axial residual stress. This highlights the importance of considering all three artery layers, data matched material
models, anisotropic micro-constituents, residual stress and viscoelastic effects in biomechanical analysis of coronary arteries
which is crucial to understanding the complex interactions.
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 39

Stress-strain magnitude and distribution and, hence, the plaque rupture mechanism, were largely influenced by plaque
morphology. Stress gradient was suggested as a more important parameter than stress magnitude alone in thin-cap fi-
bratheroma plaque types (Lee et al., 2017) compared to standard fibrofatty plaques, leading to increased rupture vulnerabil-
ity. Peak stresses were seen at the plaque shoulders, confirming clinical evidence of rupture occurring at this region, with
increasing shoulder length resulting in increased structural stress and decreased wall shear stress. The inclusion of micro-
calcification resulted in significant contradictory results, however when considered as rigid inclusions, stress concentration
factors surrounding micro-calcifications was found to exceed a value of five (Cardoso et al., 2014), with most influence at low
fibrous cap thicknesses, particularly when spaced closely together; the critical threshold noted for plaque rupture was seen
to vary from 247 – 300 kPa. This stress concentration was a potential cause for plaque rupture at non-peak stress locations,
however, the small size (5-65 μm) makes in vivo imaging of micro-calcification challenging; these inclusions were suggested
to fail through various methods such as brittle decohesion and interfacial debonding, however cavitation was suggested as
the most prominent (Maldonado et al., 2013). Paris’ law was adapted to relate crack propagation to the number of cardiac
cycles, with increased lipid pool size and thinner fibrous cap thickness resulting in decreased fatigue life; collagen fibre
strength also effected crack propagation direction. The effects of fluid driven crack propagation has yet to be considered
in a coronary artery setting, however, the inclusion of a hydraulic shear stress intensity factor into the fracture criterion
(Perkowska et al., 2017) seems an intuitive next step.
Whilst current imaging technology is improving rapidly, visual inspection of the coronary vasculature alone is not suf-
ficient to consider all the aforementioned interactions; hence, the biomechanical analysis of coronary arteries can provide
much needed support and predictive capabilities to clinical settings. Whilst computation ability still limits the assessment of
some characteristics, future biomechanical investigations should consider some of the following key areas: (1) Microstructure:
Endothelial cells in the intima layer are critical for cellular transport between the media and blood flow and impact on wall
shear stress, fluid-artery interaction, pressure, growth factors, molecular viscosity and wall shear stress gradient. Future
modelling should investigate how changes in microstructure can be accounted for with specific attention to fluid leakage,
localised weak points in fibres and inelastic damage of arteries in biomechanical settings to accurately capture the effects
on the initiation of disease or plaque rupture. (2) Fluid properties: Whilst the use of non-Newtonian, pulsatile flow is critical,
further investigation is needed into the effect of outlet profiles into smaller arteries such as the capillaries to discern pres-
sure or flow changes throughout the artery. Furthermore, standard or ‘healthy’ properties of blood have been prominently
investigated; the effect of blood disorders or age as well as variations in the pulsatile flow profile to include variations in
heart rate or cardiac valve anomalies could be investigated. (3) Material data: The mechanical testing of soft tissue in ‘ex
vivo’ conditions poses challenges to obtaining accurate data; standardisation of testing protocol to determine accurate data
or assessment of these properties in vivo is crucial for accurate biomechanical analysis. (4) Heart and respiratory effects: The
three-dimensional motion of the heart and respiratory motion were rarely considered in previous studies; furthermore the
effect of the heart wall motion and properties, including epicardium and myocardium layers on artery motion and biome-
chanics, should be considered for the accurate development of biomechanical models of the coronary vasculature. With
the continuing development of more accurate, patient specific biomechanical models from in vivo data, providing critical
information in the clinical settings that was otherwise unavailable will assist in the prediction of mechanisms leading to
myocardial infarction such as atherosclerosis and plaque rupture. It is hoped this ability will assist in combating the in-
creasing number of deaths, morbidity and economic burden from cardiovascular diseases.

Appendix A. Classification of reviewed articles


40
Table 1
Theoretical studies on coronary arteries (Structural analysis).

Study Methodology Material property Blood Model Vessel type In vivo/ in vitro Description

(Nguyen & Levy, 2010) Nonlinear boundary value problem, Elastic — Coronary artery Idealised model Interface model, calcification,
rigid inclusion de-bonding, brittle interface failure

H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201
(Dolla et al., 2012) 2D Finite element, linear, near Isotropic, elastic — Coronary artery Idealised model Thin-cap fibroatheroma, Necrotic core,
incompressible plaque rupture
(Cilla et al., 2013) Nonlinear finite element, hyperelastic, isotropic plaque/ — Coronary artery Idealised model Positive & Negative remodelling, fibre
incompressible anisotropic artery orientation, eccentric & concentric
plaque
(Mohammadi & Finite element Anisotropic & isotropic, — Coronary artery Idealised model Fibre-reinforced, fibrous cap & necrotic
Mequanint, 2014) hyperelastic core thickness, stress intensity factor
(Lee et al., 2017) Nonlinear 2D finite element, Hyperelastic, — Coronary artery Idealised model Plaque classification, stress gradient
incompressible, Mooney-Rivilin
model
(Akyildiz et al., 2011) 2D nonlinear finite element, large Homogeneous, anisotropic & — Coronary artery Idealised model Intima thickness variation, necrotic core
deformation, incompressible isotropic, hyperelastic
(Buffinton & Nonlinear finite element, Homogeneous, orthotropic, — Coronary artery Idealised model Calcification geometry, stress
Ebenstein, 2014) incompressible Mooney-Rivilin hyperelastic
model
(Cilla et al., 2013) Finite element, nonlinear, Hyperelastic, isotropic plaque/ — Coronary artery Idealised model Fibre orientation, microcalcification
incompressible anisotropic artery
(Cardoso et al., 2014) Nonlinear finite element, Hyperelastic, homogeneous, — Coronary artery Idealised model Holzapfel models, cell level
Neo-Hookean & Mooney-Rivilin, isotropic, elastic calcification microcalcification
thick-walled
(Maldonado et al., 2013) Incompressible 2D finite element, Isotropic, hyperelastic Homogeneous Coronary artery Idealised model Rupture mechanics, cavitation &
neo-Hookean model, pressure interfacial debonding, high resolution
homogeneous micro-CT, fibrous cap voids
(Holzapfel & Gasser, 2007) Two-layer finite element, thick Anisotropic, inelastic — Left anterior Theoretical model Fibre-matrix composite, balloon
walled, circular tube descending angioplasty
(Cilla et al., 2012) 3D nonlinear finite element, Hyperelastic, isotropic & — Coronary artery Idealised model Residual stress, fibre orientation,
incompressible anisotropic Gasser-Odgen-Holzapfel model
(Versluis et al., 2006) Nonlinear, incompressible, Odgen Isotropic, hyperelastic Homogeneous Coronary artery Theoretical model Crack propagation, plaque rupture,
strain, finite element analysis pressure fatigue
(Pei et al., 2013) Linear finite element Elastic — Coronary artery Idealised model Fracture mechanics, crack propagation,
growth rate
(Alastrué et al., 2007) Nonlinear, incompressible, finite Anisotropic/ compressible & — Coronary artery Theoretical model Deterministic, stochastic, fibre
element damage model hyperelastic & elastic orientation,
(Cyron & Humphrey, 2014) Incompressible numerical model Homogeneous, elastic + inelastic, — Blood Vessels Idealised Model Material degradation and deposition,
smooth muscle cell, collagen fibres
(Melnikova et al., 2017) Cell based model Anisotropic — Coronary Artery Idealised model Hexagonal close packed lattice,
intercellular repulsion/attraction, stent
deployment
(Hollander et al., 2011) Nonlinear microstructure model, Hyperelastic, — Left Anterior Idealised Model Passive media layer, descriptive
incompressible Descending estimation error, residual stress,
swelling
(Cilla et al., 2012) Nonlinear, incompressible, Artificial Hyperelastic — Coronary artery Idealised model Support Vector Machines, plaque
Neural Network vulnerability
This Table relates to Section 2.1.1 of this review and outlines purely theoretical, structural studies on coronary arteries.
Table 2
Theoretical studies on coronary arteries (fluid analysis).

Study Methodology Material property Blood model Vessel type In vivo/ in vitro Description

(Anand et al., 2013) Oldroyd-B model Viscoelastic Shear thinning Blood Vessels, Carotid Idealised numerical Thermodynamic framework, rate-type fluid
Artery methods model model and low-shear rate criterion inclusion
(Arts & Thick walled, incompressible Soft tissue — Idealised coronary Idealised, numerical Intramyocardial pressure, myocardial perfusion
Reneman, 1985) vessel model due to stenosis, coronary circulation

H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201
(Holenstein & Linear Three-parameter Collapsible artery — Coronary artery Idealised model Intramyocardial circulation, three-layered,
Nerem, 1990) Windkessel model prolonged diastole
(Rumberger & Nonlinear, incompressible, Elastic Newtonian Left coronary artery Idealised model Method of characteristics, 1-D unsteady flow,
Nerem, 1977) thick-walled, straight wave oscillations
tapered tube
(Perktold et al., 1991) Galerkin finite element Rigid artery Viscous, homogeneous, Left main coronary Idealised model 3-D Pulsatile flow, curved tapered artery, full
method, Incompressible, Newtonian artery Navier-Stokes equations, secondary flow &
wall shear stress
(Bark & Ku, 2010) Incompressible finite volume Rigid artery Newtonian, laminar, Coronary artery Idealised model 2D steady & pulsatile flow, wall shear stress,
CFD viscous platelet aggregation/activation, surface
roughness
Rigid, impermeable Laminar & turbulent, Left coronary artery Idealised model 2D and 3D model comparison, pulsatile flow,
(Shanmugavelayudam et al., artery viscous shear stress, velocity distribution
2010)
(Wong et al., 2006) Incompressible Rigid artery Non-Newtonian, Coronary artery Mathematical model Steady flow, helical stenosis lesion, irrotational
mathematical model, homogeneous flow
Power law,
(Melih Guleren, 2013) 3D finite element, Carreau — Laminar & Turbulent, Left anterior Idealised model Pulsatile flow, eccentric & concentric plaque,
model, Large Eddy Non-Newtonian descending, right heart load
Simulation coronary artery
(Chua et al., 2001) Axisymmetric, finite element Rigid artery Newtonian, viscous, Coronary artery Idealised model Steady and pulsatile flow, normalised wall shear
CFD stress, Womersley number,
(Dash et al., 1999) Incompressible Rigid artery Newtonian, laminar Coronary artery Idealised model Steady flow, catheter insertion, artery curvature,
secondary streamlines, wall shear stress,
Reynolds number
(Migliavacca et al., Non-linear, incompressible, Rigid artery Newtonian, viscous Coronary artery Idealised model Hypoplastic left heart syndrome, Norwood &
2006) 3-D finite volume CFD, shunts Sano operation
multiscale lumped
parameter model
(Frattolin et al., 2015) Incompressible, Finite — Newtonian, viscous Left main coronary Idealised model Coronary steal, side branch flows, Medina
element CFD artery bifurcation classification
(Kachanov & Rigid particle suspension — Newtonian, anisotropic Blood flow Idealised model Effective viscosity, ellipsoidal particles, thin
Abedian, 2015) viscosity platelet, non-interaction approximation (NIA)
(Li et al., 2001) Macromolecule transport Permeable wall Newtonian and Coronary artery Idealised model Steady and pulsatile flow, wall flux, wall shear
model, Womersley non-Newtonian, stress
number viscous
(Giannoglou et al., 2-D finite element CFD Smooth, inelastic, Non-Newtonian, RCA & LCA arterial Idealised model Steady flow, parabolic inlet velocity profile,
2002) impermeable laminar trees artery wall thickening, wall shear stress,
wall molecular viscosity
(Jung et al., 2006) Multiphase flow, Finite Rigid wall Non-Newtonian, Right coronary artery Idealised model Particulate migration, Red Blood cells,
element CFD viscous, Newtonian shear-thinning
Plasma
(Sinha Roy et al., Finite element CFD, Galerkin Smooth, rigid Non-Newtonian, viscous Coronary artery Idealised model Guidewire, balloon angioplasty, steady &
2006) method, Carreau model plaque pulsatile flow, hyperaemic
(continued on next page)

41
42
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201
Table 2 (continued)

Study Methodology Material property Blood model Vessel type In vivo/ in vitro Description

(Banerjee et al., 2007) Carreau model Smooth, rigid Non-Newtonian, viscous Left anterior Mathematical model Lesion flow coefficient, viscous & momentum
plaque descending change loss, percutaneous transluminal
coronary angioplasty,
(Huang et al., 2007) Semi-elastic artery non-newtonian Coronary artery Mathematical model Plasma viscosity, ligand-receptor disassociation,
(Akherat & Incompressible — Laminar, Newtonian & Coronary arteries Idealised model Wall shear stress (WSS) and velocity
Kimiaghalam, 2010) non-Newtonian distribution patterns, 2D, Pulsatile blood flow
(Harmouche et al., Collateral flow index Pressure, — Left main coronary Idealised model Anastomotic vessels, collateral flow estimation,
2012) artery, LAD, LCx, pressure distal to thrombosis
RCA
(Mandal, Incompressible finite — Inhomogeneous, Coronary artery Mathematical model Axisymmetric 2D Pulsatile flow, haematocrit
Mukhopadhyay, & difference method viscous, variation, wall shear stress
Layek, 2012)
(Wu et al., 2017) Non-linear Viscoelastic Viscous, shear-thinning, Blood flow Idealised model Plasma, red blood cells, platelets, mixture
Multi-component, theory, transport flux, micro-channel, crevice,
Oldroyd-B fluid,
non-uniform distribution
This Table relates to Section 2.1.2 of this review and outlines purely theoretical, fluid based studies on coronary arteries.
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201
Table 3
Theoretical studies on coronary arteries (fluid–structure interaction analysis).

Study Methodology Material property Blood Model Vessel type In vivo/ in vitro Description

(Pakravan et al., 2017) Incompressible, Hyperelastic/ viscoelastic Newtonian, viscous, Left main bifurcation to left Idealised model FSI, cell scale model, endothelial cell
Mooney-Rivilin, Voigt microfilaments laminar, circumflex and left morphology
viscoelastic microfibres anterior descending
(Rotman et al., 2017) Compressible, Hyperelastic, Viscous, laminar, Coronary artery Idealised model Normal and Hyperaemic flow, pressure –
incompressible homogeneous, Newtonian distensibility relationship
Numerical FSI
(Jahromi et al., 2019) Incompressible, single Isotropic, homogeneous, Newtonian, viscous Left main bifurcation to left Idealised model FSI, endothelial cell alignment, secondary
layer, Mooney-Rivilin hyperelastic circumflex and left plaque formation, unsteady blood flow
model anterior descending
(Gholipour et al., 2018) Non-linear, 3 layered Viscoelasticity and Non-Newtonian blood Left main, left circumflex, Idealised model Bifurcation, blood pulsation, heart
artery FSI hyperelastic left anterior descending, motion, active media contraction,
microcalcification
(Gholipour et al., 2018) Non-linear, 3 layered, Viscoelasticity, Non-Newtonian blood Coronary artery Idealised model Blood pulsation, microcalcification
tapered artery FSI hyperelastic
This Table relates to Section 2.1.3 of this review and outlines purely theoretical, fluid–structure interaction (FSI) studies on coronary arteries.

43
44
Table 4

H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201
Experimental studies on arteries and plaque (in vitro analysis).

Study Imaging modality Methodology Material property Blood Model Vessel type In vivo/ in vitro Number of Description
samples

(Mark et al., 1985) — Quasi-steady — Blood mimicking Left coronary In vitro — Velocity profiles,
working fluid artery system pulsatile flow, flow
(eugenol) rate, flux wave,
average shear stress
(Lin et al., 2006) Histology Calcification effect on — — Fabricated, parallel In vitro — Beta-glycerophos-phate
fluid shear stress plate model (BGP)/ calcifying
vascular cells nodules/
detachment patterns
(Feng et al., 2015) — Waveform analysis, — Fat and bone to Fabricated model In vitro — Plaque composition and
flow rate and model lipid and rupture risk
pressure calcification in
distribution plaque
(Albuquerque & Microscopy Endothelial cell Endothelial cell — Human coronary In vitro — Intimal wound & healing
Flozak, 2001) culture, Parallel monolayer on endothelial cells through wall shear
plate, continuous collagen-coated stress,
flow chamber coverslip β -actin-motility
(Chamiot-Clerc et al., — Incompressible, — — Mammary and Ex vivo 32 patients Vasoreactivity,
1998) cylindrical, constant radial arteries vasorestricting agents,
length noradrenaline,
stiffness, bypass graft,
sensitivity
(Holzapfel et al., 2005) — Nonlinear, uniaxial Hysteresis, anisotropic, — Left anterior In vitro 13 (78 segments) Intima, Media,
tension test, descending Adventitia mechanical
least-squares artery properties, stiffness
constitutive model
(Karimi et al., 2013) — Stress, strain, — — Coronary artery In vitro 22 Force-displacement,
deformation, elastic atherosclerosis vs
modulus healthy artery
(Jankowska et al., 2015) — Polynomial- Isotropic + orthotropic — Left and right In vitro 18 patients 31 Stress-strain, histological
exponential model, coronary samples structure
uni-axial tensile arteries
testing, One-layer
artery,
incompressible
(continued on next page)
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201
Table 4 (continued)

Study Imaging modality Methodology Material property Blood Model Vessel type In vivo/ in vitro Number of Description
samples

(Karimi et al., 2016) Direct Image Uniaxial strain testing, Poisson’s ratio, — Coronary artery In vitro 17 Healthy and
Correlation incompressible isotropic atherosclerotic
(DIC) artery wall, arteries
(Karimi et al., 2017) Direct Image Quasilinear, prony Viscoelastic — Coronary arteries In vitro 10 Stress-relaxation curve,
Correlation series, uniaxial ramp/hold model
(DIC) tensile test
(Ozolanta et al., 1998) — Mechanical properties, — — Right coronary In vitro 205 Age, sex, artery
biochemical artery, Left thickness, connective
composition, anterior tissue
structure descending
(Zhao et al., 2007) — Length, wall — — Saphenous vein In vitro 12 Coronary artery bypass
thickness + area, graft graft, manual
stress-strain, elastic distension,
properties pressurized &
zero-stress state
(Montini-Ballarin et al., — 2 layered artery Anisotropic, — Coronary arteries, In vitro — Vascular graft, synthetic
2016) (adventitia, media), bioresorbable, saphenous vein polymer
segmented mimicking
poly(ester
urethane),
poly(L-lactic acid),
(He et al., 2006) Electron Morphology, porosity, Nanofiber mesh of — Coronary artery In vitro — Collagen coated,
Microscopy mechanical poly(L-lactic endothelial cells endothelial cell
properties, surface acid)-co-poly(e- function, tissue
chemistry caprolactone) engineered grafts
(Sharifpoor et al., 2011) — Cyclic Mechanical Biodegradable — Coronary artery In vitro — Vascular Smooth Muscle
Strain, tensile polyurethane cells cell proliferation
properties scaffold
This Table relates to Section 2.2.1 of this review and outlines purely experimental/clinical studies on coronary arteries undertaken in vitro. For classification purposes, studies outlined as “ex vivo” have also been
classified in this section as investigations were undertaken outside of the physiologically accurate environment.

45
46
Table 5
Experimental studies on arteries and plaque (in vivo analysis).

Study Imaging modality Methodology Material Blood Vessel type In vivo/ in vitro Number of samples Description
property Model

H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201
(Carabello & Angiography Blood pressure, cardiac — — Left Ventricle + Aorta In vivo 14 Obesity, cardiac loading, contractile
Gittens, 1987) volumes & stress function comparison to lean physique
(Brinkman et al., Coronary Distance, tortuosity, curvature — — Left circumflex, Left In vivo 30 Geometric parameters, age, branch
1994) angiography anterior angles,
descending, and
Left main artery
(Friedman & Multiplane contrast Branch angles, Intimal — — Left anterior In vivo 15 (23 branch points) Daughter vessel relationship, Intima &
Ding, 1997) angiogram thickening descending Media thickness, mechanical forces
(van der Steen et al., IVUS elastography Calcification, strain, pressure — — Left anterior In vivo 12 Plaque vulnerability, strain in the arterial
2002) descending, right wall
coronary artery,
and circumflex
(Shaw et al., 2002) IVUS Artery dilation, compliance, — — Left anterior In vivo 61 Coronary angioplasty, plaque morphology
plaque compression descending and
circumflex
(Sauvée et al., 2007) Endoscope imaging Calibrated landmark tracking, — — Cardiac motion In vivo — Computer vision, on-pump cardiac
surface texture tracking surgery, displacement, acceleration
(Bulant et al., 2017) Computed Artery curvature, geometry — — Left anterior In vivo 48 Point wise curvature, geometric risk
Tomography descending artery factor, distal curvature, artery lesions,
Angiogram centerlines
(Zhu et al., 2009) Biplane angiography Curvature, torsion, tortuosity — — Right coronary artery In vivo 67 Images (52 Geometric atherogenesis risk factors, end
and Left anterior patients) diastole, variability
descending
(Konta & Bett, 2003) Angiogram Artery compression, bend, — — Left circumflex artery In vivo 100, 673 segments Coronary artery movement, lesion
displacement and left anterior severity
descending
(Lin et al., 2014) Magnetic Resonance Coronary dispensability index — — Left main, proximal In vivo 27 (98 segments) Heart transplant patient, allograft
Angiogram, MRI (CDI), wall thickness, wall left anterior
area, lumen area descending, right
coronary artery
(Liang et al., 2008) IVUS 2D, strain tensor measurement — — Left anterior In vivo 1 Radial & circumferential strain,
descending continuous pullback
(Keshavarz- IVUS Lagrangian Speckle Model — — Coronary artery In vivo 12 Shear-strain elastography, directional
Motamed et al., Estimator coronary atherectomy
2014)
(Timmins et al., 2013) Virtual Circumferentially co-register — — Coronary artery In vivo 636 paired images Plaque progression & focal mechanical
Histology-IVUS VH-IVUS datasets, artery (5 patients) forces, plaque morphology &
thickness constituents
(Molony et al., 2016) IVUS, OCT Longitudinal & circumferential — — Coronary artery In vivo 12 Catheter pullback speeds & angle, Lumen
co-registration eccentricity, calcification
This Table relates to Section 2.2.2 of this review and outlines purely experimental/clinical studies on coronary arteries undertaken in vivo.
Table 6
Experimental/clinical together with theoretical studies on arteries and plaque in vitro (structural analysis).

Study Imaging modality Methodology Material property Blood model Vessel type In vivo/ in vitro Number of Description
samples

(Veress et al., 2002) IVUS Nonlinear, finite Isotropic, — Left anterior Ex vivo + theoretical — Warping, plaque strain, image
element, hyperelastic descending model registration, noise sensitivity
constitutive
analysis

H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201
(Nieuwstadt et al., — Homogeneous, Hyperelastic, — Coronary artery In vitro + theoretical 4 Axial sampling resolution, peak
2013) incompressible, anisotropic model plaque cap stress, plaque
rupture
(Kok et al., 2016) — Homogeneous, — — Coronary arteries In vivo + theoretical 73 Necrotic core thickness, peak
incompressible, model cap stress
MATLAB,
Neo-Hookean
model
(Chau et al., 2004) OCT + Histology 2D FEA, plane strain, Isotropic — Coronary artery Ex vitro + theoretical Stress & strain comparison
Mooney-Rivilin model between OCT and histology
model
(Karimi et al., 2013) DIC Nonlinear 2D & 3D Isotropic, — Coronary arteries In vitro + theoretical 23 Peak plaque stress, plaque
finite element hyperelastic, model rupture, 70-90% Stenosis,
model, tensile testing,
Neo-Hookean
model,
homogeneous,
incompressible
(Williamson et al., Histology 2D, Linear & Isotropic & — Coronary artery Ex vivo + theoretical Stress & strain sensitivity,
2003) nonlinear, anisotropic, model fibrous plaque, lipid pool,
incompressible FEA elastic calcification, residual stress
model
(Hoshino et al., 2009) IVUS Incompressible, rigid Hyperelastic, — Bovine smooth Ex vivo + theoretical — Effect of lipid and rigid
inclusion, FEA, muscle model inclusion mechanical failure,
uniaxial + biaxial cells + Human thin solid
loading coronary artery
(Speelman et al., 2011) — Finite element Isotropic, — Coronary arteries In vitro + idealised 50 cross-sections Necrotic core, cap thickness,
analysis, dual hyperelastic, model of 9 plaques initial stress state, backward
layered artery, incremental method
homogeneous
(Mangalaprakash et al., — Nonlinear, — — Left anterior In vitro _ theoretical 1 Plaque rupture, calcification,
2013) homogeneous, descending model multiple lipid pools,
finite element artery
analysis, uniaxial
tensile test
(Maldonado et al., Micro-CT, 3D FEA, Neo-Hookean, Isotropic Constant LAD, RCA, Left Ex vivo + theoretical 92 Micro-calcification & plaque
2012) Histology incompressible pressure circumflex model rupture, stress intensity
(Campbell et al., 2014) Histology Linear, 2D FEA Elastic, isotropic, — Coronary artery Ex vivo + Theoretical 74 Inflammation, plaque erosion
model and rupture, thin & thick cap
fibroatheroma, physiological
markers
(Ferrara & MRI Finite element model, Hyperelastic, — Coronary artery In vitro + theoretical — Plaque rupture, fracture, crack
Pandolfi, 2008) 3 layered artery anisotropic, model propagation, fibre cohesive
isotropic strength
plaque

47
(continued on next page)
48
Table 6 (continued)

Study Imaging modality Methodology Material property Blood model Vessel type In vivo/ in vitro Number of Description

H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201
samples

(Abdelali et al., 2014) IVUS Hyperelastic, quasi- Isotopic — Right coronary In vitro + idealised 12 Hydrogel bilayer phantom
incompressible, artery model + in-vivo models, plaque cap buckling
nonlinear, finite validation
element model
(Karimi et al., 2017) — Nonlinear, artificial Isotropic (Neo- — Coronary artery In vitro + theoretical 16 Collagen fibre
neural network, Hookean) + model remodelling + angles,
constitutive anisotropic mechanical properties
damage modelling (Fung),
hyperelastic
(van der Horst et al., — 1D, thick-walled, Hyperelastic, — Porcine + Human In vitro + idealised 7 Porcine, 2 Fibre-reinforced, pulsatile
2012) single layered left anterior model Human pressure test, netting
artery, two-fibre descending analysis, collagen orientation
model,
incompressible
(Ooi et al., 2014) — — Hyperelastic — Coronary artery In vitro + theoretical 55 Endothelin-1, artery
model distensibility, contractile
response, smooth muscle
tone, pre-stretch
(Wang, Zhang, & — Incompressible wall, Isotropic — Left anterior In vitro + theoretical — Parameter sensitivity,
Kassab, 2008) nonlinear + linear descending, model stress-strain equation, helical
parameters, right coronary collagen alignment
Generalised artery + Porcine
Hooke’s Law, coronary artery
exponential
constitutive model
(Pazos, Mongrain, & IVUS Incompressible, Isotropic„ — Coronary arteries In vitro + idealised — Single and dual layered, lipid
Tardif, 2010) homogeneous, hyperelastic, model pool,
finite element Polyvinyl
analysis alcohol artery
(Akyildiz et al., 2016) — Plane strain, Hyperelastic, — Left anterior In vitro + theoretical 13 Geometric irregularities
incompressible, 2D isotropic descending, first model
finite element diagonal branch,
analysis, large right coronary,
deformation, left circumflex
uniaxial tensile
test, neo-Hookean
model
This Table relates to Section 2.3.1 Structural analysis of this review and outlines theoretical together with experimental/clinical studies on coronary arteries focusing on structural based analysis undertaken in
vitro.
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201
Table 7
Experimental/clinical together with theoretical studies on arteries and plaque in vitro (fluid analysis).

Study Imaging Methodology Material property Blood model Vessel type In vivo/ in vitro Number of Description
modality samples

(Bluestein et al., PIV Galerkin finite element, — Newtonian, viscous Artery In vitro + idealised — Poiseuille flow, platelet transport
1997) incompressible, model and activation potential
steady & turbulent
(Perktold et al., — Finite element CFD, Vascular cast Sodium thiocyanate blood Left anterior In vitro + idealised — Wall shear stress, velocity profile
1997) incompressible, mimicking fluid, descending artery model
Newtonian, first bifurcation
homogeneous
(Brunette et al., IVUS, PIV 3D phantom model Injection moulded Glycerol (63%) + water Coronary artery In vitro + idealised — Mild stenosis flow regime,
2008) reconstructed from silicone blood mimicking fluid, model secondary flow patterns & wall
PIV Newtonian shear stress determination from
2D PIV data
(Ai et al., 2010) – Micro-electric- — Dextran 40 dissolved in Fabricated coronary In vitro + theoretical — Steady + pulsatile flow, convective
mechanical system, saline working fluid, artery model heat transfer, flow reversal, wall
incompressible, fully laminar, Newtonian shear stress
developed
(Ji et al., 2010) — — Polyvinyl alcohol Water as working fluid Fabricated coronary In vitro — Pulsatile flow, curvature, stenosis,
hydrogel model artery stiffness
(Martorell et al., — CFD Polydimethylsiloxane Oscillatory + pulsatile flow Left main artery and In vitro + theoretical — Oxidized low-density lipoprotein
2014) (PDMS) scaffolds common carotid model uptake, monocyte adhesion,
seeded with artery bifurcations recirculation, endothelial cell
Smooth Muscle function
Cells
This Table relates to Section 2.3.1 Fluid analysis of this review and outlines combined theoretical and experimental/clinical studies on coronary arteries focusing on fluid based analysis undertaken in vitro.

49
50
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201
Table 8
Experimental/clinical together with theoretical studies on arteries and plaque in vitro (fluid–structure interaction analysis).

Study Imaging Methodology Material property Blood model Vessel type In vivo/ in vitro Number of Description
modality samples

(Tang et al., 2004) MRI Incompressible, Hyperelastic, isotropic, Laminar, viscous, Coronary + Carotid In vitro theoretical 6 Plaque structure, lipid size, cap
2D + 3D model homogenous Newtonian artery model thickness, calcification, pulsatile
flow
(Tang et al., 2005) MRI Incompressible Hyperelastic, isotropic, Laminar, viscous, Coronary + Carotid In vitro theoretical — Pulsatile pressure, calcification,
homogenous Newtonian artery model necrotic core, fibrotic plaques
(Zheng et al., 2005) MRI Nonlinear, Hyperelastic, isotropic, Newtonian + viscous Coronary arteries Ex vivo + idealised FSI 22 Pulsatile flow, stenosis, fibrous cap,
incompressible homogeneous fluid model lipid pool, fibrocalcific, necrotic
wall and fluid, core
Mooney-Rivlin
model
(Huang et al., 2014) MRI Incompressible, Hyperelastic, isotropic, Laminar, Newtonian, Coronary artery Ex vivo + theoretical 12 Plaque wall stress, ex-vivo MRI,
Mooney-Rivlin homogeneous viscous model formalin set artery
model, 2D + 3D
(Tang et al., 2009) MRI Incompressible Hyperelastic, Laminar, viscous, Coronary artery In Vitro + theoretical 6 Cyclic bending, plaque
isotropic + Newtonian model vulnerability
anisotropic,
homogenous
(Kural et al., 2012) IVUS Nonlinear, Pseudoelastic, — Coronary and Carotid In vitro _ theoretical — Stiffness, stress location, 3 model
incompressible, anisotropic artery model comparison (Fun-type, Choi-Vito,
biaxial + uniaxial Mooney-Rivilin)
testing, finite
element model
(Pagiatakis et al., — Linear, Incompressible Elastic, isotropic Newtonian, viscous, Left main, left In vitro + theoretical 2 FSI, collagen fibre orientation,
2015) laminar anterior model principle stress orientation,
descending, left anisotropic map
circumflex
This Table relates to Section 2.3.1 Fluid-Structure Interaction (FSI) Analysis of this review and outlines combined theoretical and experimental/clinical studies on coronary arteries focusing on fluid–structure
interaction (FSI) based analysis undertaken in vitro.
Table 9
Experimental/clinical together with theoretical studies on arteries and plaque in vivo (structural analysis).

Study Imaging modality Methodology Material property Blood Model Vessel type In vivo/ in vitro Number of Description
samples

(Ohayon et al., 2005) IVUS 2D & 3D FEA, finite Elastic, transverse Constant Coronary artery In vivo + idealised — 2D & 3D comparisons, peak
deformation, isotropic (varied pressure model circumferential stress, plaque

H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201
linear, 2 layered radial) rupture location
artery
(Gross & Biplane Time interpolation Artery curvature, — Left anterior In vivo + theoretical 1 Artery flexure, time
Friedman, 1998) cineangiogram 3D acceleration & descending model interpolation,
reconstruction velocity
(Ding & Biplane Time interpolation Torsion, — Left anterior In vivo + theoretical 12 Template matching technique,
Friedman, 2000) cineangiogram 3D displacement, descending, right model strain, motion trajectories
reconstruction curvature coronary artery
(Le Floc’h et al., 2009) IVUS Finite element, Isotropic, — Coronary artery In vivo + theoretical 7 Cap thickness, necrotic core &
plane strain, Model calcium effect on mechanical
constrained properties & plaque
motion, quasi- vulnerability
incompressible
(Baldewsing et al., IVUS 2D, Linear, Elastic, isotropic, — Coronary artery In vivo + in — Heterogeneous plaque
2008) incompressible, homogeneous vitro + theoretical morphology, soft tissue
plane strain constituents model strain, Young’s modulus,
elastograph, modulograph
(Danilouchkine et al., IVUS Linear, Elastic, isotropic — LAD, RCA, Left In vivo + theoretical 16 patients Coronary artery motion, IVUS
2009) incompressible, circumflex model pullback, tissue rotation,
finite element, dense optical flow
plane stress
(Ohayon et al., 2008) IVUS 2D, single layered Anisotropic — Coronary arteries In vivo + theoretical 24 Necrotic core area & positive
FEA, virtually model remodelling, plaque
incompressible, vulnerability
plane strain
(Costopoulos et al., VH-IVUS, OCT 2D finite element, Isotropic, Coronary artery In vivo + idealised 64 2D Plaque structural stress,
2017) Mooney-Rivilin hyperelastic, model plaque composition
model, incompressible,
nonlinear homogeneous
(Alastrué et al., 2007) CT Multilayered, fibre Anisotropic, Constant Coronary artery + Iliac In vivo + theoretical — Residual stress effects, opening
reinforced, FEA hyperleastic pressure artery model angle estimation,
deformation gradient tensor
decomposition
(Ohayon et al., 2011) CT angiogram, MRI 3 layered, finite Anisotropic, Constant Left main coronary In vivo + theoretical 8 Circular cross section, constant
element, hyperelastic pressure bifurcation model thickness, stiffness &
nonlinear, stress/strain, plaque
incompressible, formation site
finite strain
(Baldewsing, Schaar, IVUS 2D, linear, near Isotropic, elastic — Coronary artery In vivo + in 5 Young’s modulus from
Mastik, Oomens, & incompressible, vitro + theoretical Elastogram, strain imaging,
van der Steen, plane strain model minimisation algorithm
2005) parametric FEA
(Wang et al., 2016) IVUS 3D finite element, Anisotropic, — Coronary artery In vivo + theoretical 1 Residual stress, axial and
Mooney-Rivilin isotropic plaque model circumferential shrinkage
This Table relates to Section 2.3.2 Structural analysis of this review and outlines combined theoretical and experimental/clinical studies on coronary arteries focusing on structural based analysis undertaken in
vivo.

51
Table 10

52
Experimental/clinical together with theoretical studies on arteries and plaque in vivo (fluid analysis).

Study Imaging modality Methodology Material property Blood Model Vessel type In vivo/ in vitro Number of Description
samples

(Zeng et al., 2003) Biplane 3D finite element, Moving artery Hagen-Poiseuille, Right coronary In vivo + idealised 1 Wall shear stress from artery
cineangiogram unsteady, viscous artery model motion, steady and pulsatile
incompressible flow
(García et al., 2006) IVUS + angiogram 3D finite element, Rigid artery Newtonian, Right coronary In vivo + idealised 7 Stent restenosis & wall shear

H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201
incompressible, viscous, laminar artery model stress, wall growth instability
quasi-steady
(Soulis et al., 2006) Angiogram 3D finite element Rigid artery Non-Newtonian, Left main coronary In vivo + idealised — Artery centreline & radius
CFD, Power law laminar artery & model reconstruction, steady flow, wall
model bifurcations shear stress variation
(Soulis et al., 2007) Angiogram 3D finite element Rigid artery Non-Newtonian, Left main coronary In vivo + idealised — Wall pressure, wall shear stress,
CFD, Power law laminar artery & model molecular viscosity and
model bifurcations stress/pressure gradient relation
to atherogenesis
(Chaichana et al., CT angiogram 3D CFD, Rigid artery Newtonian, Left main coronary In vivo + idealised 4 Variation in bifurcation angle
2011) incompressible viscous, laminar artery and model effects on hemodynamics, wall
flow bifurcations shear stress and gradient,
pulsatile flow
(Timmins et al., VH-IVUS, biplane 3D CFD, Rigid artery Newtonian, Left main and left In vivo + idealised 5 Focal association between wall
2015) angiogram incompressible viscous, anterior model shear stress & atherosclerosis
pulsatile flow descending progression
(Moreno & IVUS 3D finite element, Rigid artery Newtonian Coronary artery In vivo + idealised 42 Plaque morphology effects on fluid
Bhaganagar, 2013) incompressible model velocity, laminar-turbulent
transition, peak turbulent
intensity & wall shear stress
(Soulis et al., 2010) IVUS + angiogram 3D steady coupled Rigid, permeable Non-Newtonian Right coronary In vivo + idealised 6 Low-density-lipoprotein
mass transport endothelium (Power Law artery model concentration and transport,
and Model),laminar wall shear stress dependant
Navier-Stokes endothelium permeability and
flow hydraulic conductivity
(Melchionna et al., CT 3D finite Rigid artery Newtonian, Left coronary In vivo + idealised 1 Finite haematocrit, flow of
2013) angiogram + FFR difference, viscous, red artery tree model suspended particles in stenosis,
steady, Lattice blood cell in FFR sensitivity to stenosis
Boltzmann blood plasma severity
technique
(Zhang & — Casson and — Non-Newtonian, Coronary artery In 12 Canine, Blood constitutive equation, red
Kuang, 2000) Power-Law viscous vivo + theoretical 11 human blood cells, haematocrit,
blood model hemorheology
constitutive
equation
(Johnston et al., Biplane angiogram 3D nonlinear finite Rigid artery Newtonian, viscous Right coronary In vivo + idealised 4 Wall shear stress distribution
2004) volume CFD, and artery model changes with blood model,
Generalised non-Newtonian steady inlet velocity variations
Power Law
models
(Johnston et al., Biplane angiogram 3D nonlinear finite Rigid artery Newtonian, Right coronary In vivo + idealised 4 Wall shear stress over the cardiac
2006) volume CFD, viscous, artery model cycle, pulsatile blood, particle
Generalised non-Newtonian flow field
Power Law
model
(continued on next page)
Table 10 (continued)

Study Imaging modality Methodology Material property Blood Model Vessel type In vivo/ in vitro Number of Description

H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201
samples

(Katranas et al., CT angiogram 3D finite volume Rigid artery Non-Newtonian, Left main, LAD, In vivo + idealised 22 subjects, Wall stress & atherosclerosis
2015) CFD, Power Law steady, laminar RCA, left model 28 progression, vascular stiffness,
model circumflex arteries curvature, plaque volume
(Prosi et al., 2004) Diagnostic 3D finite element Dynamic curvature Newtonian Left anterior In vivo + idealised — Influence of dynamic curvature of
catheterization Arbitrary descending model bifurcation on local
Lagrangian- bifurcation hemodynamics, pulsatile flow,
Eulerian, wall shear stress
incompressible
(Berthier et al., Angiogram 3D finite element Rigid wall Newtonian, viscous Right coronary In vivo + idealised 1 Simplified & patient specific
2002) CFD, Galerkin artery model geometry comparison, wall sear
method stress & mesh qaulity
(Rivolo et al., 2014) Fractional Flow Finite difference — — Porcine & human In vivo + signal 10 Human, Coronary Wave Intensity Analysis
Reserve, scheme, left anterior analysis 50 (cWIA), forward & backward
Coronary Flow Savitzky-Golay descending Porcine wave pressure, velocity
Reserve filter
(Gijsen et al., 2014) IVUS + Multi Slice 3D CFD, Rigid artery Non-Newtonian Right coronary and In vivo + idealised 10 3D bifurcation reconstruction,
Computer incompressible left anterior model shear stress comparisions
Tomography descending
(Li et al., 2015) OCT, 3D angiogram 3D nonlinear finite Rigid artery Steady, viscous LAD, RCA, left In vivo + idealised 21 OCT & 3D angiogram fusion and
volume CFD flow circumflex model inclusion of coronary side
branch outflow for shear stress
effect
(Tran et al., 2017) CT 3D nonlinear finite Rigid artery Scaled coronary Aorta and coronary In vivo + idealised — Parameter uncertainty effects on
element, outflow – arteries model oscillatory shear index
Bayesian Murray’s Law
lumped
parameter
network, Monte
Carlo method
(Wolters et al., IVUS Carreau-Yasuda Rigid & dynamic Newtonian and Right coronary Idealised model — Characteristic shear rate, scaled
2001) model artery non-Newtonian, artery viscosity, wall shear stress
viscous, dependence on artery motion
unsteady and flow
(Fujiwara et al., X-ray angiogram 3D Finite volume Artery movement Fully developed, Right coronary In vivo + idealised 1 Vessel dynamics & compliance,
2015) CFD, & cross section unsteady flow artery model oscillatory shear index
Incompressible variation with distribution
flow
This Table relates to Section 2.3.2 Fluid analysis of this review and outlines combined theoretical and experimental/clinical studies on coronary arteries focusing on fluid based analysis undertaken in vivo.

53
54
Table 11

H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201
Experimental/clinical together with theoretical studies on arteries and plaque in vivo (fluid–structure interaction analysis).

Study Imaging modality Methodology Material property Blood Model Vessel type In vivo/ in vitro Number of Description
samples

(Asanuma et al., IVUS 3D FEA Isotropic Newtonian, Left anterior In vivo + theoretical 1 Wall shear and tissue stress,
2013) incompressible descending model stress distribution, Stenting
(Wang et al., 2015) VH-IVUS 3D finite element, Hyperelastic, Newtonian, Coronary artery In vivo + theoretical 3 Artery curvature, plaque
nearly- anisotropic artery, incompressible, model progression & predictive
incompressible, isotropic plaque, laminar parameters
Mooney-Rivilin homogeneous
(Guo et al., 2017) VH-IVUS 3D finite element, Anisotropic artery, Newtonian, Coronary artery In vivo + theoretical 1 In-vivo vs ex-vivo material
nonlinear isotropic plaque incompressible, model properties, axial and
Mooney-Rivilin laminar circumferential stretch effects
(Dong et al., 2015) CT angiogram 3D finite element, 9 Hyperelastic Newtonian, viscous, Left main coronary In vivo + idealised 5 idealised, 1 Coronary bifurcation angle effect
parameter laminar bifurcation model in-vivo on local hemodynamic &
Mooney-Rivilin mechanical force
(Liu et al., 2017) CT angiogram 3D linear finite Elastic, isotropic Newtonian, Right coronary In vivo + theoretical 12 Wall shear & von Mises stress,
element incompressible, artery model artery remodelling & plaque
pulsatile burden
(Wu et al., 2018) Angiogram 3D FEA Isotropic — Coronary artery In vivo + theoretical 16 Single layered shell, superficial
Mooney-Rivilin model wall stress, cardiac contraction
model & Global
displacement
loadings
(Rodríguez et al., Biplane 3D FEA, quasi- Hyperelastic, Incompressible, Right coronary In vivo + theoretical 1 Intima remodelling, volumetric
2007) angiogram + IVUS incompressible, isotropic elastin viscous artery growth, shear stress influence
nonlinear, matrix, on endothelial cells
neo-Hookean anisotropic
(isotropic) collagen fibre
(Yang et al., 2009) IVUS, X-ray 3D finite element, Anisotropic artery, Newtonian, Left anterior In vivo + theoretical 1 Uniaxial test, cyclic bending and
angiogram nonlinear isotropic plaque incompressible, descending anisotropic material effect on
Mooney-Rivilin laminar flow and stress-strain
conditions
(Fan et al., 2014) IVUS + X-ray 3D, Mooney-Rivilin Hyperelastic, Newtonian, laminar, Coronary artery In vivo + theoretical 10 Biaxial tensile testing, plaque
angiogram model anisotropic incompressible model thickness & stress, cyclic
bending
This Table relates to Section 2.3.2 Fluid-structure interaction (FSI) analysis of this review and outlines combined theoretical and experimental/clinical studies on coronary arteries focusing on FSI based analysis
undertaken in vivo.
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 55

References

Australian Institute of Health and Welfare. (2018). Heart, stroke & vascular diseases. Australian Institute of Health and Welfare.
Abdelali, M., Reiter, S., Mongrain, R., Bertrand, M., L’Allier, P. L., Kritikou, E. A., & Tardif, J.-C. (2014). Cap buckling as a potential mechanism of atherosclerotic
plaque vulnerability. Journal of the Mechanical Behavior of Biomedical Materials, 32, 210–224.
Abou Sherif, S., Ozden Tok, O., Taşköylü, Ö., Goktekin, O., & Kilic, I. D. (2017). Coronary artery aneurysms: A review of the epidemiology, pathophysiology,
diagnosis, and treatment. Frontiers in Cardiovascular Medicine, 4, 24.
Ai, L., Zhang, L., Dai, W., Hu, C., Shung, K. K., & Hsiai, T. K. (2010). Real-time assessment of flow reversal in an eccentric arterial stenotic model. Journal of
Biomechanics, 43, 2678–2683.
Akherat, S. M. J. M., & Kimiaghalam, M. (2010). A numerical investigation on pulsatile blood flow through consecutive axi-symmetric stenosis in coronary
artery. In Proceedings of the 10th biennial conference on engineering systems design and analysis (pp. 757–765). American Society of Mechanical Engineers.
Akyildiz, A. C., Speelman, L., & Gijsen, F. J. H. (2014). Mechanical properties of human atherosclerotic intima tissue. Journal of Biomechanics, 47, 773–783.
Akyildiz, A. C., Speelman, L., Nieuwstadt, H. A., van Brummelen, H., Virmani, R., van der Lugt, A., van der Steen, A. F., Wentzel, J. J., & Gijsen, F. J. (2016). The
effects of plaque morphology and material properties on peak cap stress in human coronary arteries. Computer Methods in Biomechanics and Biomedical
Engineering, 19, 771–779.
Akyildiz, A. C., Speelman, L., van Brummelen, H., Gutiérrez, M. A., Virmani, R., van der Lugt, A., van der Steen, A. F., Wentzel, J. J., & Gijsen, F. J. (2011).
Effects of intima stiffness and plaque morphology on peak cap stress. Biomedical Engineering Online, 10, 25.
Alastrué, V., Peña, E., Martínez, M. Á., & Doblaré, M. (2007a). Assessing the use of the “opening angle method” to enforce residual stresses in patient-specific
arteries. Annals of Biomedical Engineering, 35, 1821–1837.
Alastrué, V., Rodríguez, J., Calvo, B., & Doblaré, M. (2007b). Structural damage models for fibrous biological soft tissues. International Journal of Solids and
Structures, 44, 5894–5911.
Albuquerque, M. L. C., & Flozak, A. S. (2001). Patterns of living β -actin movement in wounded human coronary artery endothelial cells exposed to shear
stress. Experimental Cell Research, 270, 223–234.
American Heart Association. (2017). Cardiovascular disease: A costly burden for America projections through 2035. American Heart Association.
Anand, M., Kwack, J., & Masud, A. (2013). A new generalized Oldroyd-B model for blood flow in complex geometries. International Journal of Engineering
Science, 72, 78–88.
Antoniadis, A. P., Mortier, P., Kassab, G., Dubini, G., Foin, N., Murasato, Y., Giannopoulos, A. A., Tu, S., Iwasaki, K., & Hikichi, Y. (2015). Biomechanical modeling
to improve coronary artery bifurcation stenting: Expert review document on techniques and clinical implementation. JACC: Cardiovascular Interventions,
8, 1281–1296.
Arbustini, E., Dal Bello, B., Morbini, P., Burke, A., Bocciarelli, M., Specchia, G., & Virmani, R. (1999). Plaque erosion is a major substrate for coronary throm-
bosis in acute myocardial infarction. Heart, 82, 269–272.
Arfaoui, M., Trifa, M., Mansouri, K., Karoui, A., & Renard, Y. (2018). Three-dimensional singular elastostatic fields in a cracked Neo-Hookean hyperelastic
solid. International Journal of Engineering Science, 128, 1–11.
Arts, T., & Reneman, R. (1985). Interaction between intramyocardial pressure (IMP) and myocardial circulation. Journal of Biomechanical Engineering, 107,
51–56.
Asanuma, T., Higashikuni, Y., Yamashita, H., Nagai, R., Hisada, T., & Sugiura, S. (2013). Discordance of the areas of peak wall shear stress and tissue stress in
coronary artery plaques as revealed by fluid-structure interaction finite element analysis: A Case Study. International Heart Journal, 54, 54–58.
Asgharzadeh Shirazi, H., & Ayatollahi, M. R. (2014). Biomechanical analysis of functionally graded biomaterial disc in terms of motion and stress distribution
in lumbar spine. International Journal of Engineering Science, 84, 62–78.
Athanasiou, L. S., Fotiadis, D. I., & Michalis, L. K. (2017). Atherosclerotic plaque characterization methods based on coronary imaging. Academic Press.
Attia, M. A., & Rahman, A. A. A. (2018). On vibrations of functionally graded viscoelastic nanobeams with surface effects. International Journal of Engineering
Science, 127, 1–32.
Bakhshi Khaniki, H., & Hosseini-Hashemi, S. (2017). Dynamic response of biaxially loaded double-layer viscoelastic orthotropic nanoplate system under a
moving nanoparticle. International Journal of Engineering Science, 115, 51–72.
Baldewsing, R. A., Danilouchkine, M. G., Mastik, F., Schaar, J. A., Serruys, P. W., & van der Steen, A. F. (2008). An inverse method for imaging the local
elasticity of atherosclerotic coronary plaques. IEEE Transactions on Information Technology in Biomedicine, 12, 277–289.
Baldewsing, R. A., Schaar, J. A., Mastik, F., Oomens, C. W., & van der Steen, A. F. (2005). Assessment of vulnerable plaque composition by matching the
deformation of a parametric plaque model to measured plaque deformation. IEEE Transactions on Medical Imaging, 24, 514–528.
Banerjee, R. K., Sinha Roy, A., Back, L. H., Back, M. R., Khoury, S. F., & Millard, R. W. (2007). Characterizing momentum change and viscous loss of a
hemodynamic endpoint in assessment of coronary lesions. Journal of Biomechanics, 40, 652–662.
Bark, D. L., & Ku, D. N. (2010). Wall shear over high degree stenoses pertinent to atherothrombosis. Journal of Biomechanics, 43, 2970–2977.
Barretta, R., Čana d̄ija, M., Luciano, R., & de Sciarra, F. M. (2018). Stress-driven modeling of nonlocal thermoelastic behavior of nanobeams. International
Journal of Engineering Science, 126, 53–67.
Berthier, B., Bouzerar, R., & Legallais, C. (2002). Blood flow patterns in an anatomically realistic coronary vessel: Influence of three different reconstruction
methods. Journal of Biomechanics, 35, 1347–1356.
Beyar, R., Caminker, R., Manor, D., & Sideman, S. (1993). Coronary flow patterns in normal and ischemic hearts: Transmyocardial and artery to vein distri-
bution. Annals of Biomedical Engineering, 21, 435–458.
Bhattacharyya, R., Sarangi, S., & Samantaray, A. K. (2015). Effect of stress-softening on the ballooning motion of hyperelastic strings. International Journal of
Engineering Science, 96, 19–33.
Bloom, D. E., Cafiero, E., Jané-Llopis, E., Abrahams-Gessel, S., Bloom, L. R., Fathima, S., Feigl, A. B., Gaziano, T., Hamandi, A., & Mowafi, M. (2012). The global
economic burden of noncommunicable diseases. Program on the Global Demography of Agin.
Bluestein, D., Niu, L., Schoephoerster, R. T., & Dewanjee, M. K. (1997). Fluid mechanics of arterial stenosis: Relationship to the development of mural
thrombus. Annals of Biomedical Engineering, 25, 344.
Braunwald, E. (2013). Coronary plaque erosion: Recognition and management. JACC: Cardiovascular Imaging.
Brinkman, A., Baker, P. B., Newman, W., Vigorito, R., & Friedman, M. (1994). Variability of human coronary artery geometry: An angiographic study of the
left anterior descending arteries of 30 autopsy hearts. Annals of Biomedical Engineering, 22, 34–44.
Brunette, J., Mongrain, R., Laurier, J., Galaz, R., & Tardif, J. C. (2008). 3D flow study in a mildly stenotic coronary artery phantom using a whole volume PIV
method. Medical Engineering & Physics, 30, 1193–1200.
Buffinton, C. M., & Ebenstein, D. M. (2014). Effect of calcification modulus and geometry on stress in models of calcified atherosclerotic plaque. Cardiovas-
cular Engineering and Technology, 5, 244–260.
Bulant, C. A., Blanco, P. J., Clausse, A., Assunção, A., Jr, Lima, T., Ávila, L., Feijóo, R. A., & Lemos, P. A. (2017). Association between three-dimensional vessel
geometry and the presence of atherosclerotic plaques in the left anterior descending coronary artery of high-risk patients. Biomedical Signal Processing
and Control, 31, 569–575.
Calvert, J. W. (2014). Chapter 5- Ischemic Heart Disease and its Consequences. In M. S. Willis, J. W. Homeister, & J. R. Stone (Eds.), Cellular and molecular
pathobiology of cardiovascular disease (pp. 79–100). San Diego: Academic Press.
Campbell, I. C., Suever, J. D., Timmins, L. H., Veneziani, A., Vito, R. P., Virmani, R., Oshinski, J. N., & Taylor, W. R. (2014). Biomechanics and inflammation in
atherosclerotic plaque erosion and plaque rupture: Implications for cardiovascular events in women. PloS One, 9, e111785.
Carabello, B. A., & Gittens, L. (1987). Cardiac mechanics and function in obese normotensive persons with normal coronary arteries. The American Journal of
Cardiology, 59, 469–473.
56 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

Cardoso, L., Kelly-Arnold, A., Maldonado, N., Laudier, D., & Weinbaum, S. (2014). Effect of tissue properties, shape and orientation of microcalcifications on
vulnerable cap stability using different hyperelastic constitutive models. Journal of Biomechanics, 47, 870–877.
Cardoso, L., & Weinbaum, S. (2014). Changing views of the biomechanics of vulnerable plaque rupture: A review. Annals of Biomedical Engineering, 42,
415–431.
Chaichana, T., Sun, Z., & Jewkes, J. (2011). Computation of hemodynamics in the left coronary artery with variable angulations. Journal of Biomechanics, 44,
1869–1878.
Chamiot-Clerc, P., Copie, X., Renaud, J.-F., Safar, M., & Girerd, X. (1998). Comparative reactivity and mechanical properties of human isolated internal mam-
mary and radial arteries. Cardiovascular Research, 37, 811–819.
Chau, A. H., Chan, R. C., Shishkov, M., MacNeill, B., Iftimia, N., Tearney, G. J., Kamm, R. D., Bouma, B. E., & Kaazempur-Mofrad, M. R. (2004). Mechanical
analysis of atherosclerotic plaques based on optical coherence tomography. Annals of Biomedical Engineering, 32, 1494–1503.
Chebbi, E., Wali, M., & Dammak, F. (2016). An anisotropic hyperelastic constitutive model for short glass fiber-reinforced polyamide. International Journal of
Engineering Science, 106, 262–272.
Chen, H., & Kassab, G. S. (2016). Microstructure-based biomechanics of coronary arteries in health and disease. Journal of Biomechanics, 49, 2548–
2559.
Chien, S. (2003). Molecular and mechanical bases of focal lipid accumulation in arterial wall. Progress in Biophysics and Molecular Biology, 83, 131–151.
Choi, H. S., & Vito, R. (1990). Two-dimensional stress-strain relationship for canine pericardium. Journal of Biomechanical Engineering, 112, 153–159.
Chua, L., Yu, S., & Xue, Q. (2001). Scaling laws for wall shear stress through stenoses under steady and pulsatile flow conditions, Proceedings of the
Institution of Mechanical Engineers. Part H: Journal of Engineering in Medicine, 215, 503–514.
Chuong, C., & Fung, Y. (1983). Three-dimensional stress distribution in arteries. Journal of Biomechanical Engineering, 105, 268–274.
Cilla, M., Martinez, J., Pena, E., & Martínez, M. Á. (2012a). Machine learning techniques as a helpful tool toward determination of plaque vulnerability. IEEE
Transactions on Biomedical Engineering, 59, 1155–1161.
Cilla, M., Monterde, D., Peña, E., & Martínez, M. Á. (2013a). Does microcalcification increase the risk of rupture?, Proceedings of the Institution of Mechanical
Engineers. Part H: Journal of Engineering in Medicine, 227, 588–599.
Cilla, M., Pena, E., & Martinez, M. (2012b). 3D computational parametric analysis of eccentric atheroma plaque: Influence of axial and circumferential
residual stresses. Biomechanics and Modeling in Mechanobiology, 11, 1001–1013.
Cilla, M., Peña, E., Martínez, M., & Kelly, D. (2013b). Comparison of the vulnerability risk for positive versus negative atheroma plaque morphology. Journal
of Biomechanics, 46, 1248–1254.
Cleary, M. A., Geiger, E., Grady, C., Best, C., Naito, Y., & Breuer, C. (2012). Vascular tissue engineering: The next generation. Trends in Molecular Medicine, 18,
394–404.
Corrales-Medina, V. F., Madjid, M., & Musher, D. M. (2010). Role of acute infection in triggering acute coronary syndromes. The Lancet Infectious Diseases, 10,
83–92.
Costopoulos, C., Huang, Y., Brown, A. J., Calvert, P. A., Hoole, S. P., West, N. E., Gillard, J. H., Teng, Z., & Bennett, M. R. (2017). Plaque rupture in coronary
atherosclerosis is associated with increased plaque structural stress. JACC: Cardiovascular Imaging, 10, 1472–1483.
Cyron, C., & Humphrey, J. (2014). Vascular homeostasis and the concept of mechanobiological stability. International Journal of Engineering Science, 85,
203–223.
Danilouchkine, M. G., Mastik, F., & van der Steen, A. F. (2009). A study of coronary artery rotational motion with dense scale-space optical flow in intravas-
cular ultrasound. Physics in Medicine & Biology, 54, 1397.
Dash, R. K., Jayaraman, G., & Mehta, K. N. (1999). Flow in a catheterized curved artery with stenosis. Journal of Biomechanics, 32, 49–61.
Dehrouyeh-Semnani, A. M. (2018). On the thermally induced non-linear response of functionally graded beams. International Journal of Engineering Science,
125, 53–74.
Destrade, M., Liu, Y., Murphy, J. G., & Kassab, G. S. (2012). Uniform transmural strain in pre-stressed arteries occurs at physiological pressure. Journal of
Theoretical Biology, 303, 93–97.
Ding, Z., & Friedman, M. H. (20 0 0). Dynamics of human coronary arterial motion and its potential role in coronary atherogenesis. Journal of Biomechanical
Engineering, 122, 488–492.
Dolla, W. J. S., House, J. A., & Marso, S. P. (2012). Stratification of risk in thin cap fibroatheromas using peak plaque stress estimates from idealized finite
element models. Medical Engineering & Physics, 34, 1330–1338.
Dong, J., Sun, Z., Inthavong, K., & Tu, J. (2015). Fluid–structure interaction analysis of the left coronary artery with variable angulation. Computer Methods in
Biomechanics and Biomedical Engineering, 18, 1500–1508.
Doraiswamy, S., Criscione, J. C., & Srinivasa, A. R. (2016). A technique for the classification of tissues by combining mechanics based models with Bayesian
inference. International Journal of Engineering Science, 106, 95–109.
Ebrahimi, F., & Barati, M. R. (2016). A nonlocal higher-order refined magneto-electro-viscoelastic beam model for dynamic analysis of smart nanostructures.
International Journal of Engineering Science, 107, 183–196.
Edwards, C. R., Bouchier, I. A. D., Haslett, C., & Chilvers, E. (1991). Davidson’s principles and practice of medicine (16 ed.). Churchill Livingstone Edinburgh.
El Baroudi, A., Razafimahéry, F., & Rakotomanana, L. (2014). Fluid–structure interaction within three-dimensional models of an idealized arterial wall.
International Journal of Engineering Science, 84, 113–126.
Falk, E., Shah, P. K., & Fuster, V. (1995). Coronary plaque disruption. Circulation, 92, 657–671.
Fallah, A., Ahmadian, M. T., & Aghdam, M. M. (2017a). Rate-dependent behavior of connective tissue through a micromechanics-based hyper viscoelastic
model. International Journal of Engineering Science, 121, 91–107.
Fallah, A., Ahmadian, M. T., & Mohammadi Aghdam, M. (2017b). Rate-dependent behavior of connective tissue through a micromechanics-based hyper
viscoelastic model. International Journal of Engineering Science, 121, 91–107.
Fan, H., Luo, Y., Yang, F., & Li, W. (2018). Approaching perfect energy absorption through structural hierarchy. International Journal of Engineering Science,
130, 12–32.
Fan, R., Tang, D., Yang, C., Zheng, J., Bach, R., Wang, L., Muccigrosso, D., Billiar, K., Zhu, J., Ma, G., Maehara, A., & Mintz, G. S. (2014). Human coronary
plaque wall thickness correlated positively with flow shear stress and negatively with plaque wall stress: An IVUS-based fluid-structure interaction
multi-patient study. Biomedical Engineering Online, 13, 32.
Farokhi, H., Ghayesh, M., & Amabili, M. (2013a). Nonlinear resonant behavior of microbeams over the buckled state. Applied Physics A, 113, 297–307.
Farokhi, H., Ghayesh, M., & Amabili, M. (2013b). Nonlinear dynamics of a geometrically imperfect microbeam based on the modified couple stress theory.
International Journal of Engineering Science, 68, 11–23.
Farokhi, H., & Ghayesh, M. H. (2015a). Nonlinear dynamical behaviour of geometrically imperfect microplates based on modified couple stress theory.
International Journal of Mechanical Sciences, 90, 133–144.
Farokhi, H., & Ghayesh, M. H. (2015b). Thermo-mechanical dynamics of perfect and imperfect Timoshenko microbeams. International Journal of Engineering
Science, 91, 12–33.
Farokhi, H., & Ghayesh, M. H. (2017). Nonlinear resonant response of imperfect extensible Timoshenko microbeams. International Journal of Mechanics and
Materials in Design, 13, 43–55.
Farokhi, H., & Ghayesh, M. H. (2018a). Nonlinear mechanics of electrically actuated microplates. International Journal of Engineering Science, 123, 197–213.
Farokhi, H., & Ghayesh, M. H. (2018b). On the dynamics of imperfect shear deformable microplates. International Journal of Engineering Science, 133, 264–283.
Farokhi, H., & Ghayesh, M. H. (2018c). Nonlinear mechanical behaviour of microshells. International Journal of Engineering Science, 127, 127–144.
Farokhi, H., & Ghayesh, M. H. (2018d). Supercritical nonlinear parametric dynamics of Timoshenko microbeams. Communications in Nonlinear Science and
Numerical Simulation, 59, 592–605.
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 57

Farokhi, H., Ghayesh, M. H., Gholipour, A., & Hussain, S. (2017a). Motion characteristics of bilayered extensible Timoshenko microbeams. International Journal
of Engineering Science, 112, 1–17.
Farokhi, H., Ghayesh, M. H., Gholipour, A., & Tavallaeinejad, M. (2017b). Nonlinear oscillations of viscoelastic microplates. International Journal of Engineering
Science, 118, 56–69.
Farokhi, H., Ghayesh, M. H., & Hussain, S. (2016). Large-amplitude dynamical behaviour of microcantilevers. International Journal of Engineering Science, 106,
29–41.
Feng, J., Rajeswaran, T., He, S., Wilkinson, F., Serracino-Inglott, F., Azzawi, M., Parikh, V., Miraftab, M., & Alexander, M. (2015). Investigation of the composition
of arterial plaques based on arterial waveforms and material properties. In Proceedings of the 37th annual international conference of the IEEE engineering
in medicine and biology society (EMBC) (pp. 993–996).
Ferrara, A., & Pandolfi, A. (2008). Numerical modelling of fracture in human arteries. Computer Methods in Biomechanics and Biomedical Engineering, 11,
553–567.
Fountoulakis, P., Oikonomou, E., Lazaros, G., & Tousoulis, D. (2018). Chapter 1. 2- Endothelial Function. In D. Tousoulis (Ed.), Coronary artery disease
(pp. 13–30). Academic Press.
Frattolin, J., Zarandi, M. M., Pagiatakis, C., Bertrand, O. F., & Mongrain, R. (2015). Numerical study of stenotic side branch hemodynamics in true bifurcation
lesions. Computers in Biology and Medicine, 57, 130–138.
Freed, A. D., & Einstein, D. R. (2013). An implicit elastic theory for lung parenchyma. International Journal of Engineering Science, 62, 31–47.
Friedman, M. H., & Ding, Z. (1997). Relation between the structural asymmetry of coronary branch vessels and the angle at their origin. Journal of Biome-
chanics, 31, 273–278.
Fujiwara, T., Liang, F., Tsubota, K.-i., Sugawara, M., Fan, Y.-q., & Liu, H. (2015). Effects of vessel dynamics and compliance on human right coronary artery
hemodynamics with/without stenosis. Journal of Biomechanical Science and Engineering, 10 15-0 0 015-0 0 015-0 0 015.
Fung, Y. (1991). What are the residual stresses doing in our blood vessels? Annals of Biomedical Engineering, 19, 237–249.
Fung, Y.-c. (2013). Biomechanics: mechanical properties of living tissues. Springer Science & Business Media.
Furuichi, S., Tobaru, T., Ohta, M., Asano, R., Sumiyoshi, T., & Tomoike, H. (2012). Excessive axial plaque redistribution during coronary stent implantation
confirmed by intravascular ultrasound. Cardiovascular Revascularization Medicine, 13, 144–146.
Galich, P. I., Slesarenko, V., Li, J., & Rudykh, S. (2018). Elastic instabilities and shear waves in hyperelastic composites with various periodic fiber arrange-
ments. International Journal of Engineering Science, 130, 51–61.
Ganghoffer, J.-F., & Sokolowski, J. (2014). A micromechanical approach to volumetric and surface growth in the framework of shape optimization. Interna-
tional Journal of Engineering Science, 74, 207–226.
García, J., Crespo, A., Goicolea, J., Sanmartín, M., & García, C. (2006). Study of the evolution of the shear stress on the restenosis after coronary angioplasty.
Journal of Biomechanics, 39, 799–805.
Garcìa-Garcìa, H. M., Gogas, B. D., Serruys, P. W., & Bruining, N. (2011). IVUS-based imaging modalities for tissue characterization: Similarities and differ-
ences. The International Journal of Cardiovascular Imaging, 27, 215–224.
Gasser, T. C., Ogden, R. W., & Holzapfel, G. A. (2005). Hyperelastic modelling of arterial layers with distributed collagen fibre orientations. Journal of the
Royal Society Interface, 3, 15–35.
Ghayesh, M. (2012). Subharmonic dynamics of an axially accelerating beam. Archive of Applied Mechanics, 82, 1169–1181.
Ghayesh, M., & Moradian, N. (2011). Nonlinear dynamic response of axially moving, stretched viscoelastic strings. Archive of Applied Mechanics, 81, 781–799.
Ghayesh, M. H. (2017). Nonlinear dynamics of multilayered microplates. Journal of Computational and Nonlinear Dynamics, 13 021006-021006-021012.
Ghayesh, M. H. (2018a). Nonlinear vibration analysis of axially functionally graded shear-deformable tapered beams. Applied Mathematical Modelling, 59,
583–596.
Ghayesh, M. H. (2018b). Functionally graded microbeams: Simultaneous presence of imperfection and viscoelasticity. International Journal of Mechanical
Sciences, 140, 339–350.
Ghayesh, M. H. (2018c). Dynamics of functionally graded viscoelastic microbeams. International Journal of Engineering Science, 124, 115–131.
Ghayesh, M. H. (2018d). Mechanics of tapered AFG shear-deformable microbeams. Microsystem Technologies, 24, 1743–1754.
Ghayesh, M. H. (2019a). Nonlinear oscillations of FG cantilevers. Applied Acoustics, 145, 393–398.
Ghayesh, M. H. (2019b). Dynamical analysis of multilayered cantilevers. Communications in Nonlinear Science and Numerical Simulation, 71, 244–253.
Ghayesh, M. H. (2019c). Viscoelastic nonlinear dynamic behaviour of Timoshenko FG beams. The European Physical Journal Plus, 134, 401.
Ghayesh, M. H. (2019d). Resonant vibrations of FG viscoelastic imperfect Timoshenko beams. Journal of Vibration and Control, 25, 1823–1832.
Ghayesh, M. H. (2019e). Asymmetric viscoelastic nonlinear vibrations of imperfect AFG beams. Applied Acoustics, 154, 121–128.
Ghayesh, M. H. (2019f). Mechanics of viscoelastic functionally graded microcantilevers. European Journal of Mechanics - A/Solids, 73, 492–499.
Ghayesh, M. H. (2019g). Viscoelastic mechanics of Timoshenko functionally graded imperfect microbeams. Composite Structures, 225, 110974.
Ghayesh, M. H. (2019h). Viscoelastic dynamics of axially FG microbeams. International Journal of Engineering Science, 135, 75–85.
Ghayesh, M. H., Amabili, M., & Farokhi, H. (2013a). Three-dimensional nonlinear size-dependent behaviour of Timoshenko microbeams. International Journal
of Engineering Science, 71, 1–14.
Ghayesh, M. H., Amabili, M., & Farokhi, H. (2013b). Coupled global dynamics of an axially moving viscoelastic beam. International Journal of Non-Linear
Mechanics, 51, 54–74.
Ghayesh, M. H., Amabili, M., & Farokhi, H. (2013c). Nonlinear forced vibrations of a microbeam based on the strain gradient elasticity theory. International
Journal of Engineering Science, 63, 52–60.
Ghayesh, M. H., & Farokhi, H. (2015a). Chaotic motion of a parametrically excited microbeam. International Journal of Engineering Science, 96, 34–45.
Ghayesh, M. H., & Farokhi, H. (2015b). Nonlinear dynamics of microplates. International Journal of Engineering Science, 86, 60–73.
Ghayesh, M. H., & Farokhi, H. (2018). Nonlinear behaviour of electrically actuated microplate-based MEMS resonators. Mechanical Systems and Signal Pro-
cessing, 109, 220–234.
Ghayesh, M. H., Farokhi, H., & Alici, G. (2016a). Size-dependent performance of microgyroscopes. International Journal of Engineering Science, 100, 99–111.
Ghayesh, M. H., Farokhi, H., & Alici, G. (2016b). Internal energy transfer in dynamical behavior of slightly curved shear deformable microplates. ASME Journal
of Computational and Nonlinear Dynamics, 11, 041002.
Ghayesh, M. H., Farokhi, H., & Amabili, M. (2013d). Nonlinear dynamics of a microscale beam based on the modified couple stress theory. Composites Part
B: Engineering, 50, 318–324.
Ghayesh, M. H., Farokhi, H., & Amabili, M. (2013e). Nonlinear behaviour of electrically actuated MEMS resonators. International Journal of Engineering Science,
71, 137–155.
Ghayesh, M. H., Farokhi, H., & Amabili, M. (2014). In-plane and out-of-plane motion characteristics of microbeams with modal interactions. Composites Part
B: Engineering, 60, 423–439.
Ghayesh, M. H., Farokhi, H., & Gholipour, A. (2017a). Vibration analysis of geometrically imperfect three-layered shear-deformable microbeams. International
Journal of Mechanical Sciences, 122, 370–383.
Ghayesh, M. H., Farokhi, H., & Gholipour, A. (2017b). Oscillations of functionally graded microbeams. International Journal of Engineering Science, 110, 35–53.
Ghayesh, M. H., Farokhi, H., Gholipour, A., & Tavallaeinejad, M. (2018). Nonlinear oscillations of functionally graded microplates. International Journal of
Engineering Science, 122, 56–72.
Ghayesh, M. H., Farokhi, H., & Hussain, S. (2016c). Viscoelastically coupled size-dependent dynamics of microbeams. International Journal of Engineering
Science, 109, 243–255.
Ghayesh, M. H., Kazemirad, S., & Darabi, M. A. (2011). A general solution procedure for vibrations of systems with cubic nonlinearities and nonlin-
ear/time-dependent internal boundary conditions. Journal of Sound and Vibration, 330, 5382–5400.
58 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

Ghayesh, M. H., Kazemirad, S., & Reid, T. (2012). Nonlinear vibrations and stability of parametrically exited systems with cubic nonlinearities and internal
boundary conditions: A general solution procedure. Applied Mathematical Modelling, 36, 3299–3311.
Ghayesh, M. H., Yourdkhani, M., Balar, S., & Reid, T. (2010). Vibrations and stability of axially traveling laminated beams. Applied Mathematics and Computa-
tion, 217, 545–556.
Gheorghe, A., Griffiths, U., Murphy, A., Legido-Quigley, H., Lamptey, P., & Perel, P. (2018). The economic burden of cardiovascular disease and hypertension
in low-and middle-income countries: A systematic review. BMC Public Health, 18, 975.
Gholipour, A., Farokhi, H., & Ghayesh, M. H. (2015). In-plane and out-of-plane nonlinear size-dependent dynamics of microplates. Nonlinear Dynamics, 79,
1771–1785.
Gholipour, A., Ghayesh, M. H., & Zander, A. (2018a). Nonlinear biomechanics of bifurcated atherosclerotic coronary arteries. International Journal of Engineer-
ing Science, 133, 60–83.
Gholipour, A., Ghayesh, M. H., Zander, A., & Mahajan, R. (2018b). Three-dimensional biomechanics of coronary arteries. International Journal of Engineering
Science, 130, 93–114.
Giannoglou, G. D., Soulis, J. V., Farmakis, T. M., Farmakis, D. M., & Louridas, G. E. (2002). Haemodynamic factors and the important role of local low static
pressure in coronary wall thickening. International Journal of Cardiology, 86, 27–40.
Giedrimiene, D., & King, R. (2017). Burden of Cardiovascular Disease (CVD) on Economic Cost. Comparison of Outcomes in US and Europe. Circulation:
Cardiovascular Quality and Outcomes, 10, 207.
Gijsen, F. J. H., Schuurbiers, J. C. H., van de Giessen, A. G., Schaap, M., van der Steen, A. F. W., & Wentzel, J. J. (2014). 3D reconstruction techniques of human
coronary bifurcations for shear stress computations. Journal of Biomechanics, 47, 39–43.
Gizzi, A., Vasta, M., & Pandolfi, A. (2014). Modeling collagen recruitment in hyperelastic bio-material models with statistical distribution of the fiber orien-
tation. International Journal of Engineering Science, 78, 48–60.
Goda, I., & Ganghoffer, J.-F. (2015). 3D plastic collapse and brittle fracture surface models of trabecular bone from asymptotic homogenization method.
International Journal of Engineering Science, 87, 58–82.
Goda, I., Rahouadj, R., Ganghoffer, J.-F., Kerdjoudj, H., & Siad, L. (2016). 3D couple-stress moduli of porous polymeric biomaterials using μCT image stack
and FE characterization. International Journal of Engineering Science, 100, 25–44.
Govindaraju, K., Badruddin, I. A., Viswanathan, G. N., Ramesh, S., & Badarudin, A. (2013). Evaluation of functional severity of coronary artery disease and
fluid dynamics’ influence on hemodynamic parameters: A review. Physica Medica, 29, 225–232.
Gross, M. F., & Friedman, M. H. (1998). Dynamics of coronary artery curvature obtained from biplane cineangiograms. Journal of Biomechanics, 31, 479–484.
Guo, X., Lu, X., & Kassab, G. S. (2005). Transmural strain distribution in the blood vessel wall. American Journal of Physiology-Heart and Circulatory Physiology,
288, H881–H886.
Guo, X., Zhu, J., Maehara, A., Monoly, D., Samady, H., Wang, L., Billiar, K. L., Zheng, J., Yang, C., Mintz, G. S., Giddens, D. P., & Tang, D. (2017). Quantify
patient-specific coronary material property and its impact on stress/strain calculations using in vivo IVUS data and 3D FSI models: A pilot study.
Biomechanics and Modeling in Mechanobiology, 16, 333–344.
Haddad, S. M., & Samani, A. (2017). A computational model of the left ventricle biomechanics using a composite material approach. International Journal of
Engineering Science, 111, 61–73.
Hanke, H., Lenz, C., & Finking, G. (2001). The discovery of the pathophysiological aspects of atherosclerosis–a review. Acta Chirurgica Belgica, 101, 162–169.
Harmouche, M., Maasrani, M., Corbineau, H., Verhoye, J.-P., & Drochon, A. (2012). A more sensitive pressure-based index to estimate collateral blood supply
in case of coronary three-vessel disease. Medical Hypotheses, 79, 261–263.
Hashemi, R. (2016). On the overall viscoelastic behavior of graphene/polymer nanocomposites with imperfect interface. International Journal of Engineering
Science, 105, 38–55.
He, W., Yong, T., Ma, Z. W., Inai, R., Teo, W. E., & Ramakrishna, S. (2006). Biodegradable polymer nanofiber mesh to maintain functions of endothelial cells.
Tissue Engineering, 12, 2457–2466.
Heiland, V. M., Forsell, C., Roy, J., Hedin, U., & Gasser, T. C. (2013). Identification of carotid plaque tissue properties using an experimental–numerical
approach. Journal of the Mechanical Behavior of Biomedical Materials, 27, 226–238.
Holenstein, R., & Nerem, R. M. (1990). Parametric analysis of flow in the intramyocardial circulation. Annals of Biomedical Engineering, 18, 347–
365.
Hollander, Y., Durban, D., Lu, X., Kassab, G. S., & Lanir, Y. (2011). Constitutive modeling of coronary arterial media—comparison of three model classes.
Journal of Biomechanical Engineering, 133, 061008.
Holzapfel, G. A. (2001). Biomechanics of soft tissue. The Handbook of Materials Behavior Models, 3, 1049–1063.
Holzapfel, G. A., & Gasser, T. C. (2007). Computational stress-deformation analysis of arterial walls including high-pressure response. International Journal of
Cardiology, 116, 78–85.
Holzapfel, G. A., Gasser, T. C., & Ogden, R. W. (20 0 0). A new constitutive framework for arterial wall mechanics and a comparative study of material models.
Journal of Elasticity and the Physical Science of Solids, 61, 1–48.
Holzapfel, G. A., Gasser, T. C., & Stadler, M. (2002). A structural model for the viscoelastic behavior of arterial walls: Continuum formulation and finite
element analysis. European Journal of Mechanics - A/Solids, 21, 441–463.
Holzapfel, G. A., & Ogden, R. W. (2009). Biomechanical modelling at the molecular, cellular and tissue levels. Springer Science & Business Media.
Holzapfel, G. A., & Ogden, R. W. (2014). Biomechanics of soft tissue in cardiovascular systems. Springer.
Holzapfel, G. A., Sommer, G., Gasser, C. T., & Regitnig, P. (2005). Determination of layer-specific mechanical properties of human coronary arteries
with nonatherosclerotic intimal thickening and related constitutive modeling. American Journal of Physiology-Heart and Circulatory Physiology, 289,
H2048–H2058.
Hoshino, T., Chow, L. A., Hsu, J. J., Perlowski, A. A., Abedin, M., Tobis, J., Tintut, Y., Mal, A. K., Klug, W. S., & Demer, L. L. (2009). Mechanical stress analysis
of a rigid inclusion in distensible material: A model of atherosclerotic calcification and plaque vulnerability. American Journal of Physiology-Heart and
Circulatory Physiology, 297, H802–H810.
Huang, C.-N., Peng, C.-H., Shih, C.-M., Chiu, W.-T., Chayu, C.-C., Chen, K.-C., Peng, C.-C., & Peng, R. Y. (2007). Fluid mechanical and physicochemical modeling
interprets hypertension to be capable of inducing secondary complications. Medical Hypotheses, 68, 967–978.
Huang, X., Yang, C., Zheng, J., Bach, R., Muccigrosso, D., Woodard, P. K., & Tang, D. (2014). Higher critical plaque wall stress in patients who died of coronary
artery disease compared with those who died of other causes: A 3D FSI study based on ex vivo MRI of coronary plaques. Journal of Biomechanics, 47,
432–437.
Huo, Y., Finet, G., Lefevre, T., Louvard, Y., Moussa, I., & Kassab, G. S. (2012). Which diameter and angle rule provides optimal flow patterns in a coronary
bifurcation? Journal of Biomechanics, 45, 1273–1279.
Il’ichev, A., & Fu, Y. (2014). Stability of an inflated hyperelastic membrane tube with localized wall thinning. International Journal of Engineering Science, 80,
53–61.
Jaffer, F. A., & Verjans, J. W. (2014). Molecular imaging of atherosclerosis: Clinical state-of-the-art. Heart, 100, 1469–1477.
Jahromi, R., Pakravan, H. A., Saidi, M. S., & Firoozabadi, B. (2019). Primary stenosis progression versus secondary stenosis formation in the left coronary
bifurcation: A mechanical point of view. Biocybernetics and Biomedical Engineering, 39, 188–198.
Janela, J., Moura, A., & Sequeira, A. (2010). Absorbing boundary conditions for a 3D non-Newtonian fluid–structure interaction model for blood flow in
arteries. International Journal of Engineering Science, 48, 1332–1349.
Jankowska, M. A., Bartkowiak-Jowsa, M., & Bedzinski, R. (2015). Experimental and constitutive modeling approaches for a study of biomechanical properties
of human coronary arteries. Journal of the Mechanical Behavior of Biomedical Materials, 50, 1–12.
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 59

Jennette, J. C., & Stone, J. R. (2014). Chapter 11- diseases of medium-sized and small vessels. In M. S. Willis, J. W. Homeister, & J. R. Stone (Eds.), Cellular
and molecular pathobiology of cardiovascular disease (pp. 197–219). San Diego: Academic Press.
Ji, J., Kobayashi, S., Morikawa, H., Tang, D., & Ku, D. N. (2010). Diastolic predominant flow in compliant coronary stenosis model. Journal of Biomechanical
Science and Engineering, 5, 303–313.
Jiao, P., & Alavi, A. H. (2018). Buckling analysis of graphene-reinforced mechanical metamaterial beams with periodic webbing patterns. International Journal
of Engineering Science, 131, 1–18.
John, L. C. (2009). Biomechanics of coronary artery and bypass graft disease: Potential new approaches. The Annals of Thoracic Surgery, 87, 331–
338.
Johnston, B. M., Johnston, P. R., Corney, S., & Kilpatrick, D. (2004). Non-Newtonian blood flow in human right coronary arteries: Steady state simulations.
Journal of Biomechanics, 37, 709–720.
Johnston, B. M., Johnston, P. R., Corney, S., & Kilpatrick, D. (2006). Non-Newtonian blood flow in human right coronary arteries: Transient simulations.
Journal of Biomechanics, 39, 1116–1128.
Joshi, S., & Walton, J. R. (2013). Reconstruction of the residual stresses in a hyperelastic body using ultrasound techniques. International Journal of Engineering
Science, 70, 46–72.
Jung, J., Lyczkowski, R. W., Panchal, C. B., & Hassanein, A. (2006). Multiphase hemodynamic simulation of pulsatile flow in a coronary artery. Journal of
Biomechanics, 39, 2064–2073.
Kachanov, M., & Abedian, B. (2015). On the isotropic and anisotropic viscosity of suspensions containing particles of diverse shapes and orientations.
International Journal of Engineering Science, 94, 71–85.
Karimi, A., Navidbakhsh, M., Faghihi, S., Shojaei, A., & Hassani, K. (2013a). A finite element investigation on plaque vulnerability in realistic healthy and
atherosclerotic human coronary arteries. Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine, 227, 148–
161.
Karimi, A., Navidbakhsh, M., Shojaei, A., & Faghihi, S. (2013b). Measurement of the uniaxial mechanical properties of healthy and atherosclerotic human
coronary arteries. Materials Science and Engineering: C, 33, 2550–2554.
Karimi, A., Rahmati, S. M., Sera, T., Kudo, S., & Navidbakhsh, M. (2017a). A combination of constitutive damage model and artificial neural networks to
characterize the mechanical properties of the healthy and atherosclerotic human coronary arteries. Artificial Organs, 41, E103–E117.
Karimi, A., Sera, T., Kudo, S., & Navidbakhsh, M. (2016). Experimental verification of the healthy and atherosclerotic coronary arteries incompressibility via
digital image correlation. Artery Research, 16, 1–7.
Karimi, A., Shojaei, A., & Razaghi, R. (2017b). Viscoelastic mechanical measurement of the healthy and atherosclerotic human coronary arteries using DIC
technique. Artery Research, 18, 14–21.
Karšaj, I., Sorić, J., & Humphrey, J. D. (2010). A 3-D framework for arterial growth and remodeling in response to altered hemodynamics. International journal
of engineering science, 48, 1357–1372.
Katranas, S. A., Antoniadis, A. P., Kelekis, A. L., & Giannoglou, G. D. (2015). Insights on atherosclerosis by non-invasive assessment of wall stress and arterial
morphology along the length of human coronary plaques. The International Journal of Cardiovascular Imaging, 31, 1627–1633.
Katritsis, D., Kaiktsis, L., Chaniotis, A., Pantos, J., Efstathopoulos, E. P., & Marmarelis, V. (2007). Wall shear stress: Theoretical considerations and methods of
measurement. Progress in Cardiovascular Diseases, 49, 307–329.
Kazemirad, S., Ghayesh, M., & Amabili, M. (2013). Thermo-mechanical nonlinear dynamics of a buckled axially moving beam. Archive of Applied Mechanics,
83, 25–42.
Keshavarzian, M., Meyer, C. A., & Hayenga, H. N. (2018). Mechanobiological model of arterial growth and remodeling. Biomechanics and Modeling in
Mechanobiology, 17, 87–101.
Keshavarz-Motamed, Z., & Kadem, L. (2011). 3D pulsatile flow in a curved tube with coexisting model of aortic stenosis and coarctation of the aorta. Medical
Engineering & Physics, 33, 315–324.
Keshavarz-Motamed, Z., Saijo, Y., Majdouline, Y., Riou, L., Ohayon, J., & Cloutier, G. (2014). Coronary artery atherectomy reduces plaque shear strains: An
endovascular elastography imaging study. Atherosclerosis, 235, 140–149.
Khakalo, S., Balobanov, V., & Niiranen, J. (2018). Modelling size-dependent bending, buckling and vibrations of 2D triangular lattices by strain gradient
elasticity models: Applications to sandwich beams and auxetics. International Journal of Engineering Science, 127, 33–52.
Khaniki, H. B., & Hosseini-Hashemi, S. (2017). Dynamic response of biaxially loaded double-layer viscoelastic orthotropic nanoplate system under a moving
nanoparticle. International Journal of Engineering Science, 115, 51–72.
Kim, W. Y., Danias, P. G., Stuber, M., Flamm, S. D., Plein, S., Nagel, E., Langerak, S. E., Weber, O. M., Pedersen, E. M., & Schmidt, M. (2001). Coronary magnetic
resonance angiography for the detection of coronary stenoses. New England Journal of Medicine, 345, 1863–1869.
Kok, A. M., Speelman, L., Virmani, R., van der Steen, A. F., Gijsen, F. J., & Wentzel, J. J. (2016). Peak cap stress calculations in coronary atherosclerotic plaques
with an incomplete necrotic core geometry. Biomedical Engineering Online, 15, 48.
Kolpakov, A. G., Andrianov, I. V., Rakin, S. I., & Rogerson, G. A. (2018). An asymptotic strategy to couple homogenized elastic structures. International Journal
of Engineering Science, 131, 26–39.
Konta, T., & Bett, J. H. N. (2003). Patterns of coronary artery movement and the development of coronary atherosclerosis. Circulation Journal, 67, 846–
850.
Kumar, V., Abbas, A. K., Fausto, N., & Aster, J. C. (2014). Robbins and Cotran pathologic basis of disease, professional edition e-book. Elsevier Health Sciences.
Kural, M. H., Cai, M., Tang, D., Gwyther, T., Zheng, J., & Billiar, K. L. (2012). Planar biaxial characterization of diseased human coronary and carotid arteries
for computational modeling. Journal of Biomechanics, 45, 790–798.
Le Floc’h, S., Ohayon, J., Tracqui, P., Finet, G., Gharib, A. M., Maurice, R. L., Cloutier, G., & Pettigrew, R. I. (2009). Vulnerable atherosclerotic plaque elasticity
reconstruction based on a segmentation-driven optimization procedure using strain measurements: Theoretical framework. IEEE Transactions on Medical
Imaging, 28, 1126–1137.
Leal, J., Luengo-Fernández, R., Gray, A., Petersen, S., & Rayner, M. (2006). Economic burden of cardiovascular diseases in the enlarged European Union.
European Heart Journal, 27, 1610–1619.
Lee, W., Choi, G. J., & Cho, S. W. (2017). Numerical study to indicate the vulnerability of plaques using an idealized 2D plaque model based on plaque
classification in the human coronary artery. Medical & Biological Engineering & Computing, 55, 1379–1387.
Li, W. (2018). Biomechanical property and modelling of venous wall. Progress in Biophysics and Molecular Biology, 133, 56–75.
Li, X.-y., Wen, G.-b., & Li, D. (2001). Computer simulation of non-newtonian flow and mass transport through coronary arterial stenosis. Applied Mathematics
and Mechanics, 22, 409–424.
Li, Y., Gutiérrez-Chico, J. L., Holm, N. R., Yang, W., Hebsgaard, L., Christiansen, E. H., Mæng, M., Lassen, J. F., Yan, F., Reiber, J. H. C., & Tu, S. (2015). Impact
of side branch modeling on computation of endothelial shear stress in coronary artery disease. Journal of the American College of Cardiology, 66, 125–
135.
Liang, Y., Zhu, H., Gehrig, T., & Friedman, M. H. (2008). Measurement of the transverse strain tensor in the coronary arterial wall from clinical intravascular
ultrasound images. Journal of Biomechanics, 41, 2906–2911.
Lin, K., Lloyd-Jones, D. M., Taimen, K., Liu, Y., Bi, X., Li, D., & Carr, J. C. (2014). The detection of coronary stiffness in cardiac allografts using MR imaging.
European Journal of Radiology, 83, 1402–1407.
Lin, T. C., Tintut, Y., Lyman, A., Mack, W., Demer, L. L., & Hsiai, T. K. (2006). Mechanical response of a calcified plaque model to fluid shear force. Annals of
Biomedical Engineering, 34, 1535–1541.
Liu, X., Tang, T., Yu, W., & Pipes, R. B. (2018). Multiscale modeling of viscoelastic behaviors of textile composites. International Journal of Engineering Science,
130, 175–186.
60 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

Liu, X., Wu, G., Xu, C., He, Y., Shu, L., Liu, Y., Zhang, N., & Lin, C. (2017). Prediction of coronary plaque progression using biomechanical factors and vascular
characteristics based on computed tomography angiography. Computer Assisted Surgery, 22, 286–294.
Liu, Y., Zhang, W., Wang, C., & Kassab, G. (2011). A linearized and incompressible constitutive model for arteries. Journal of Theoretical Biology, 286, 85–91.
Loree, H. M., Grodzinsky, A. J., Park, S. Y., Gibson, L. J., & Lee, R. T. (1994). Static circumferential tangential modulus of human atherosclerotic tissue. Journal
of Biomechanics, 27, 195–204.
Maher, E., Creane, A., Lally, C., & Kelly, D. J. (2012). An anisotropic inelastic constitutive model to describe stress softening and permanent deformation in
arterial tissue. Journal of the Mechanical Behavior of Biomedical Materials, 12, 9–19.
Maldonado, N., Kelly-Arnold, A., Cardoso, L., & Weinbaum, S. (2013). The explosive growth of small voids in vulnerable cap rupture: Cavitation and interfacial
debonding. Journal of Biomechanics, 46, 396–401.
Maldonado, N., Kelly-Arnold, A., Vengrenyuk, Y., Laudier, D., Fallon, J. T., Virmani, R., Cardoso, L., & Weinbaum, S. (2012). A mechanistic analysis of the role
of microcalcifications in atherosclerotic plaque stability: Potential implications for plaque rupture. American Journal of Physiology-Heart and Circulatory
Physiology, 303, H619–H628.
Mandal, M. S., Mukhopadhyay, S., & Layek, G. (2012a). Pulsatile flow of an incompressible, inhomogeneous fluid in a smoothly expanded vascular tube.
International Journal of Engineering Science, 57, 1–10.
Mandal, M. S., Mukhopadhyay, S., & Layek, G. C. (2012b). Pulsatile flow of an incompressible, inhomogeneous fluid in a smoothly expanded vascular tube.
International Journal of Engineering Science, 57, 1–10.
Mangalaprakash, A., Kumar, R. K., & Balakrishnan, K. R. (2013). Effect of calcification on plaque stresses and vulnerability: A finite element study. Mechanics
of Advanced Materials and Structures, 20, 309–315.
Mark, F. F., Bargeron, C. B., Deters, O. J., & Friedman, M. H. (1985). Nonquasi-steady character of pulsatile flow in human coronary arteries. Journal of
Biomechanical Engineering, 107, 24–28.
Martorell, J., Santoma, P., Kolandaivelu, K., Kolachalama, V. B., Melgar-Lesmes, P., Molins, J. J., Garcia, L., Edelman, E. R., & Balcells, M. (2014). Extent of flow
recirculation governs expression of atherosclerotic and thrombotic biomarkers in arterial bifurcations. Cardiovascular Research, 103, 37–46.
Mazloum, A., & Sevostianov, I. (2018). Connections between anisotropic tensors of thermal conductivity and thermal expansion coefficients. International
Journal of Engineering Science, 122, 1–13.
Melchionna, S., Amati, G., Bernaschi, M., Bisson, M., Succi, S., Mitsouras, D., & Rybicki, F. J. (2013). Risk assessment of atherosclerotic plaques based on global
biomechanics. Medical Engineering & Physics, 35, 1290–1297.
Melih Guleren, K. (2013). Numerical flow analysis of coronary arteries through concentric and eccentric stenosed geometries. Journal of Biomechanics, 46,
1043–1052.
Melnikova, N., Svitenkov, A., Hose, D., & Hoekstra, A. (2017). A cell-based mechanical model of coronary artery tunica media. Journal of The Royal Society
Interface, 14, 20170028.
Migliavacca, F., Balossino, R., Pennati, G., Dubini, G., Hsia, T.-Y., de Leval, M. R., & Bove, E. L. (2006). Multiscale modelling in biofluidynamics: Application to
reconstructive paediatric cardiac surgery. Journal of Biomechanics, 39, 1010–1020.
Mohammadi, H., & Mequanint, K. (2014). Effect of stress intensity factor in evaluation of instability of atherosclerotic plaque. Journal of Mechanics in Medicine
and Biology, 14, 1450072.
Molony, D. S., Timmins, L. H., Rasoul-Arzrumly, E., Samady, H., & Giddens, D. P. (2016). Evaluation of a framework for the co-registration of intravascular
ultrasound and optical coherence tomography coronary artery pullbacks. Journal of Biomechanics, 49, 4048–4056.
Montini-Ballarin, F., Calvo, D., Caracciolo, P. C., Rojo, F., Frontini, P. M., Abraham, G. A., & Guinea, G. V. (2016). Mechanical behavior of bilayered small-diam-
eter nanofibrous structures as biomimetic vascular grafts. Journal of the Mechanical Behavior of Biomedical Materials, 60, 220–233.
Moreno, C., & Bhaganagar, K. (2013). Modeling of stenotic coronary artery and implications of plaque morphology on blood flow. Modelling and Simulation
in Engineering, 2013, 14.
Nguyen, C. M., & Levy, A. J. (2010). The mechanics of atherosclerotic plaque rupture by inclusion/matrix interfacial decohesion. Journal of Biomechanics, 43,
2702–2708.
Nieuwstadt, H., Akyildiz, A., Speelman, L., Virmani, R., van der Lugt, A., van der Steen, A., Wentzel, J., & Gijsen, F. (2013). The influence of axial image
resolution on atherosclerotic plaque stress computations. Journal of Biomechanics, 46, 689–695.
Ogden, R. W. (1972). Large deformation isotropic elasticity–on the correlation of theory and experiment for incompressible rubberlike solids. Proceedings of
the Royal Society of London. A. Mathematical and Physical Sciences, 326, 565–584.
Ohayon, J., Finet, G., Gharib, A. M., Herzka, D. A., Tracqui, P., Heroux, J., Rioufol, G., Kotys, M. S., Elagha, A., & Pettigrew, R. I. (2008). Necrotic core thickness
and positive arterial remodeling index: Emergent biomechanical factors for evaluating the risk of plaque rupture. American Journal of Physiology-Heart
and Circulatory Physiology, 295, H717–H727.
Ohayon, J., Finet, G., Treyve, F., Rioufol, G., & Dubreuil, O. (2005). A three-dimensional finite element analysis of stress distribution in a coronary atheroscle-
rotic plaque: In-vivo prediction of plaque rupture location. Biomechanics Applied to Computer Assisted Surgery, 37, 225–241.
Ohayon, J., Gharib, A. M., Garcia, A., Heroux, J., Yazdani, S. K., Malvè, M., Tracqui, P., Martinez, M.-A., Doblare, M., Finet, G., & Pettigrew, R. I. (2011). Is arterial
wall-strain stiffening an additional process responsible for atherosclerosis in coronary bifurcations?: An in vivo study based on dynamic CT and MRI.
American Journal of Physiology-Heart and Circulatory Physiology, 301, H1097–H1106.
Ohayon, J., Yazdani, S. K., Malvè, M., Gharib, A. M., Garcia, A., Finet, G., & Pettigrew, R. I. (2017). Chapter 9- Arterial Wall Stiffness and Atherogenesis in
Human Coronaries. In Y. Payan, & J. Ohayon (Eds.), Biomechanics of living organs (pp. 193–213). Oxford: Academic Press.
Ooi, C. Y., Sutcliffe, M. P., Davenport, A. P., & Maguire, J. J. (2014). Changes in biomechanical properties of the coronary artery wall contribute to maintained
contractile responses to endothelin-1 in atherosclerosis. Life Sciences, 118, 424–429.
Ozolanta, I., Tetere, G., Purinya, B., & Kasyanov, V. (1998). Changes in the mechanical properties, biochemical contents and wall structure of the human
coronary arteries with age and sex. Medical Engineering & Physics, 20, 523–533.
Pagiatakis, C., Galaz, R., Tardif, J.-C., & Mongrain, R. (2015). A comparison between the principal stress direction and collagen fiber orientation in coronary
atherosclerotic plaque fibrous caps. Medical & Biological Engineering & Computing, 53, 545–555.
Pakravan, H. A., Saidi, M. S., & Firoozabadi, B. (2017). A multiscale approach for determining the morphology of endothelial cells at a coronary artery.
International Journal for Numerical Methods in Biomedical Engineering, 33, e2891.
Pazos, V., Mongrain, R., & Tardif, J.-C. (2010). Mechanical characterization of atherosclerotic arteries using finite-element modeling: Feasibility study on
mock arteries. IEEE Transactions on Biomedical Engineering, 57, 1520–1528.
Pei, X., Wu, B., & Li, Z.-Y. (2013). Fatigue crack propagation analysis of plaque rupture. Journal of Biomechanical Engineering, 135, 101003.
Perkowska, M., Piccolroaz, A., Wrobel, M., & Mishuris, G. (2017). Redirection of a crack driven by viscous fluid. International Journal of Engineering Science,
121, 182–193.
Perktold, K., Hofer, M., Rappitsch, G., Loew, M., Kuban, B., & Friedman, M. (1997). Validated computation of physiologic flow in a realistic coronary artery
branch. Journal of Biomechanics, 31, 217–228.
Perktold, K., Nerem, R. M., & Peter, R. O. (1991). A numerical calculation of flow in a curved tube model of the left main coronary artery. Journal of
Biomechanics, 24, 175–189.
Prosi, M., Perktold, K., Ding, Z., & Friedman, M. H. (2004). Influence of curvature dynamics on pulsatile coronary artery flow in a realistic bifurcation model.
Journal of Biomechanics, 37, 1767–1775.
Pu, L., Xiong, H., Liu, X., Zhang, H., & Zhang, Y.-T. (2014). Quantifying effect of blood pressure on stress distribution in atherosclerotic plaque. In Proceedings
of the international conference on health informatics (pp. 216–219). Springer.
Purvis Jr, N., & Giorgio, T. (1991). The effects of elongational stress exposure on the activation and aggregation of blood platelets. Biorheology, 28, 355–367.
H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201 61

Qi, L., Huang, S., Fu, G., Zhou, S., & Jiang, X. (2018). On the mechanics of curved flexoelectric microbeams. International Journal of Engineering Science, 124,
1–15.
Qiu, Y., & Tarbell, J. M. (1999). Numerical simulation of pulsatile flow in a compliant curved tube model of a coronary artery. Journal of Biomechanical
Engineering, 122, 77–85.
Ramírez-Torres, A., Rodríguez-Ramos, R., Merodio, J., Penta, R., Bravo-Castillero, J., Guinovart-Díaz, R., Sabina, F. J., García-Reimbert, C., Sevostianov, I., &
Conci, A. (2017). The influence of anisotropic growth and geometry on the stress of solid tumors. International Journal of Engineering Science, 119, 40–49.
Richardson, P. D. (2002). Biomechanics of plaque rupture: Progress, problems, and new frontiers. Annals of Biomedical Engineering, 30, 524–536.
Rivolo, S., Asrress, K. N., Chiribiri, A., Sammut, E., Wesolowski, R., Bloch, L. O., Grondal, A. K., Honge, J. L., Kim, W. Y., Marber, M., Redwood, S., Nagel, E.,
Smith, N. P., & Lee, J. (2014). Enhancing coronary Wave Intensity Analysis robustness by high order central finite differences. Artery Research, 8, 98–109.
Rodríguez, J., Goicolea, J. M., & Gabaldón, F. (2007). A volumetric model for growth of arterial walls with arbitrary geometry and loads. Journal of Biome-
chanics, 40, 961–971.
Rooz, E., Wiesner, T., & Nerem, R. (1985). Epicardial coronary blood flow including the presence of stenoses and aorto-coronary bypasses—I: Model and
numerical method. Journal of Biomechanical Engineering, 107, 361–367.
Ross, R. (1999). Atherosclerosis—an inflammatory disease. New England Journal of Medicine, 340, 115–126.
Rotman, O. M., Zaretsky, U., Shitzer, A., & Einav, S. (2017). Pressure drop and arterial compliance–two arterial parameters in one measurement. Journal of
Biomechanics, 50, 130–137.
Rubin, M. B., & Solav, D. (2016). Unphysical properties of the rotation tensor estimated by least squares optimization with specific application to biome-
chanics. International Journal of Engineering Science, 103, 11–18.
Rumberger, J. A., & Nerem, R. M. (1977). A method-of-characteristics calculation of coronary blood flow. Journal of Fluid Mechanics, 82, 429–448.
Sauvée, M., Noce, A., Poignet, P., Triboulet, J., & Dombre, E. (2007). Three-dimensional heart motion estimation using endoscopic monocular vision system:
From artificial landmarks to texture analysis. Biomedical Signal Processing and Control, 2, 199–207.
Saw, J., Mancini, G. B. J., & Humphries, K. H. (2016). Contemporary review on spontaneous coronary artery dissection. Journal of the American College of
Cardiology, 68, 297–312.
Sevostianov, I., Levin, V., & Radi, E. (2016). Effective viscoelastic properties of short-fiber reinforced composites. International Journal of Engineering Science,
100, 61–73.
Seyedkavoosi, S., Zaytsev, D., Drach, B., Panfilov, P., Gutkin, M. Y., & Sevostianov, I. (2017). Fraction-exponential representation of the viscoelastic properties
of dentin. International Journal of Engineering Science, 111, 52–60.
Shahverdi, H., & Barati, M. R. (2017). Vibration analysis of porous functionally graded nanoplates. International Journal of Engineering Science, 120, 82–99.
Shanmugavelayudam, S. K., Rubenstein, D. A., & Yin, W. (2010). Effect of geometrical assumptions on numerical modeling of coronary blood flow under
normal and disease conditions. Journal of Biomechanical Engineering, 132, 8 061004.
Shariff, M. H. B. M. (2017). On the spectral constitutive modelling of transversely isotropic soft tissue: Physical invariants. International Journal of Engineering
Science, 120, 199–219.
Sharifpoor, S., Simmons, C. A., Labow, R. S., & Santerre, J. P. (2011). Functional characterization of human coronary artery smooth muscle cells under cyclic
mechanical strain in a degradable polyurethane scaffold. Biomaterials, 32, 4816–4829.
Shaw, J. A., Kingwell, B. A., Walton, A. S., Cameron, J. D., Pillay, P., Gatzka, C. D., & Dart, A. M. (2002). Determinants of coronary artery compliance in subjects
with and without angiographic coronary artery disease. Journal of the American College of Cardiology, 39, 1637–1643.
Shirazi, H. A., & Ayatollahi, M. (2014). Biomechanical analysis of functionally graded biomaterial disc in terms of motion and stress distribution in lumbar
spine. International Journal of Engineering Science, 84, 62–78.
Siasos, G., Tsigkou, V., Oikonomou, E., Zaromitidou, M., & Tousoulis, D. (2018). Chapter 3. 6- Novel Antiplatelet Agents. In D. Tousoulis (Ed.), Coronary artery
disease (pp. 391–415). Academic Press.
Sinha Roy, A., Back, L. H., & Banerjee, R. K. (2006). Guidewire flow obstruction effect on pressure drop-flow relationship in moderate coronary artery
stenosis. Journal of Biomechanics, 39, 853–864.
Sorof, S. (2004). Intravascular atheroma monitoring: Past, present and future of identifying vulnerable plaques. Applications in Imaging-Cardiac interventions,
34–39.
Soulis, J. V., Farmakis, T. M., Giannoglou, G. D., & Louridas, G. E. (2006). Wall shear stress in normal left coronary artery tree. Journal of Biomechanics, 39,
742–749.
Soulis, J. V., Fytanidis, D. K., Papaioannou, V. C., & Giannoglou, G. D. (2010). Wall shear stress on LDL accumulation in human RCAs. Medical Engineering &
Physics, 32, 867–877.
Soulis, J. V., Giannoglou, G. D., Parcharidis, G. E., & Louridas, G. E. (2007). Flow parameters in normal left coronary artery tree. Implication to atherogenesis.
Computers in Biology and Medicine, 37, 628–636.
Speelman, L., Akyildiz, A., Den Adel, B., Wentzel, J., Van der Steen, A., Virmani, R., Van der Weerd, L., Jukema, J., Poelmann, R., & Van Brummelen, E. (2011).
Initial stress in biomechanical models of atherosclerotic plaques. Journal of Biomechanics, 44, 2376–2382.
Sun, C. T., & Jin, Z. H. (2012). Chapter 9- Cohesive Zone Model. In C. T. Sun, & Z. H. Jin (Eds.), Fracture mechanics (pp. 227–246). Boston: Academic Press.
Sun, Z., & Xu, L. (2014). Computational fluid dynamics in coronary artery disease. Computerized Medical Imaging and Graphics, 38, 651–
663.
Suter, M. J., Nadkarni, S. K., Weisz, G., Tanaka, A., Jaffer, F. A., Bouma, B. E., & Tearney, G. J. (2011). Intravascular optical imaging technology for investigating
the coronary artery. JACC: Cardiovascular Imaging, 4, 1022–1039.
Švihlová, H., Hron, J., Málek, J., Rajagopal, K., & Rajagopal, K. (2016a). Determination of pressure data from velocity data with a view toward its application
in cardiovascular mechanics. Part 1. Theoretical considerations. International Journal of Engineering Science, 105, 108–127.
Švihlová, H., Hron, J., Málek, J., Rajagopal, K. R., & Rajagopal, K. (2016b). Determination of pressure data from velocity data with a view toward its application
in cardiovascular mechanics. Part 1. Theoretical considerations. International Journal of Engineering Science, 105, 108–127.
Švihlová, H., Hron, J., Málek, J., Rajagopal, K. R., & Rajagopal, K. (2017). Determination of pressure data from velocity data with a view towards its application
in cardiovascular mechanics. Part 2. A study of aortic valve stenosis. International Journal of Engineering Science, 114, 1–15.
Taelman, L., Degroote, J., Swillens, A., Vierendeels, J., & Segers, P. (2014). Fluid–structure interaction simulation of pulse propagation in arteries: Numerical
pitfalls and hemodynamic impact of a local stiffening. International Journal of Engineering Science, 77, 1–13.
Tajaddini, A., Kilpatrick, D. L., & Vince, D. G. (2003). A novel experimental method to estimate stress-strain behavior of intact coronary arteries using
intravascular ultrasound (IVUS). Journal of Biomechanical Engineering, 125, 120–123.
Taki, A., Kermani, A., Ranjbarnavazi, S. M., & Pourmodheji, A. (2017). Chapter 4 - overview of different medical imaging techniques for the identifica-
tion of coronary atherosclerotic plaques. In Computing and visualization for intravascular imaging and computer-assisted stenting (pp. 79–106). Academic
Press.
Tang, D., Yang, C., Kobayashi, S., Zheng, J., Woodard, P. K., Teng, Z., Billiar, K., Bach, R., & Ku, D. N. (2009). 3D MRI-based anisotropic FSI models with cyclic
bending for human coronary atherosclerotic plaque mechanical analysis. Journal of Biomechanical Engineering, 131, 061010.
Tang, D., Yang, C., Zheng, J., Woodard, P. K., Saffitz, J. E., Sicard, G. A., Pilgram, T. K., & Yuan, C. (2005). Quantifying effects of plaque structure and material
properties on stress distributions in human atherosclerotic plaques using 3D FSI models. Journal of Biomechanical Engineering, 127, 1185–1194.
Tang, D., Yang, C., Zheng, J., Woodard, P. K., Sicard, G. A., Saffitz, J. E., & Yuan, C. (2004). 3D MRI-based multicomponent FSI models for atherosclerotic
plaques. Annals of Biomedical Engineering, 32, 947–960.
Timmins, L. H., Molony, D. S., Eshtehardi, P., McDaniel, M. C., Oshinski, J. N., Samady, H., & Giddens, D. P. (2015). Focal association between wall shear stress
and clinical coronary artery disease progression. Annals of Biomedical Engineering, 43, 94–106.
62 H.J. Carpenter, A. Gholipour and M.H. Ghayesh et al. / International Journal of Engineering Science 147 (2020) 103201

Timmins, L. H., Suever, J. D., Eshtehardi, P., McDaniel, M. C., Oshinski, J. N., Samady, H., & Giddens, D. P. (2013). Framework to co-register longitudinal virtual
histology-intravascular ultrasound data in the circumferential direction. IEEE Transactions on Medical Imaging, 32, 1989–1996.
Tran, J. S., Schiavazzi, D. E., Ramachandra, A. B., Kahn, A. M., & Marsden, A. L. (2017). Automated tuning for parameter identification and uncertainty
quantification in multi-scale coronary simulations. Computers & Fluids, 142, 128–138.
Tricerri, P., Dedè, L., Gambaruto, A., Quarteroni, A., & Sequeira, A. (2016). A numerical study of isotropic and anisotropic constitutive models with relevance
to healthy and unhealthy cerebral arterial tissues. International Journal of Engineering Science, 101, 126–155.
Trofimov, A., Abaimov, S., Akhatov, I., & Sevostianov, I. (2017). Effect of elastic contrast on the contribution of helical fibers into overall stiffness of a
composites. International Journal of Engineering Science, 120, 31–50.
Trofimov, A., Abaimov, S., & Sevostianov, I. (2018). Inverse homogenization problem: Evaluation of elastic and electrical (thermal) properties of composite
constituents. International Journal of Engineering Science, 129, 34–46.
Tsioufis, C., Mantzouranis, E., Kalos, T., Konstantinidis, D., & Tousoulis, D. (2018). Chapter 1. 4- Risk Factors of Atherosclerosis: Pathophysiological Mecha-
nisms. In D. Tousoulis (Ed.), Coronary artery disease (pp. 43–66). Academic Press.
van der Giessen, A. G., Gijsen, F. J., Wentzel, J. J., Jairam, P. M., van Walsum, T., Neefjes, L. A., Mollet, N. R., Niessen, W. J., van de Vosse, F. N., & de
Feyter, P. J. (2011). Small coronary calcifications are not detectable by 64-slice contrast enhanced computed tomography. The International Journal of
Cardiovascular Imaging, 27, 143–152.
van der Horst, A., van den Broek, C. N., van de Vosse, F. N., & Rutten, M. C. M. (2012). The fiber orientation in the coronary arterial wall at physiological
loading evaluated with a two-fiber constitutive model. Biomechanics and Modeling in Mechanobiology, 11, 533–542.
van der Steen, A. F. W., de Korte, C. L., Mastik, F., Doyley, M. M., Bom, N., Serruys, P. W., & Carlier, S. G. (2002). Morphological and mechanical information
of coronary arteries obtained with intravascular elastography. Feasibility Study in Vivo. European Heart Journal, 23, 405–413.
VanEpps, J. S., & Vorp, D. A. (2007). Mechanopathobiology of atherogenesis: A review. Journal of Surgical Research, 142, 202–217.
Veress, A. I., Weiss, J. A., Gullberg, G. T., Vince, D. G., & Rabbitt, R. D. (2002). Strain measurement in coronary arteries using intravascular ultrasound and
deformable images. Journal of Biomechanical Engineering, 124, 734–741.
Versluis, A., Bank, A. J., & Douglas, W. H. (2006). Fatigue and plaque rupture in myocardial infarction. Journal of Biomechanics, 39, 339–347.
Wang, C., Zhang, W., & Kassab, G. S. (2008). The validation of a generalized Hooke’s law for coronary arteries. American Journal of Physiology-Heart and
Circulatory Physiology, 294, H66–H73.
Wang, L., Wu, Z., Yang, C., Zheng, J., Bach, R., Muccigrosso, D., Billiar, K., Maehara, A., Mintz, G. S., & Tang, D. (2015). IVUS-based FSI models for human
coronary plaque progression study: Components, correlation and predictive analysis. Annals of Biomedical Engineering, 43, 107–121.
Wang, L., Zhu, J., Samady, H., Monoly, D., Zheng, J., Guo, X., Maehara, A., Yang, C., Ma, G., Mintz, G. S., & Tang, D. (2016). Effects of residual stress, axial
stretch, and circumferential shrinkage on coronary plaque stress and strain calculations: A modeling study using IVUS-based near-idealized geometries.
Journal of Biomechanical Engineering, 139, 11 014501.
Wenk, J. F., Papadopoulos, P., & Zohdi, T. I. (2010). Numerical modeling of stress in stenotic arteries with microcalcifications: A micromechanical approxi-
mation. Journal of Biomechanical Engineering, 132, 091011.
Williamson, S., Lam, Y., Younis, H., Huang, H., Patel, S., Kaazempur-Mofrad, M., & Kamm, R. (2003). On the sensitivity of wall stresses in diseased arteries
to variable material properties. Journal of Biomechanical Engineering, 125, 147–155.
Wolters, B. J. B. M., Slager, C. J., Gijsen, F. J. H., Wentze, J. J., Krams, R., & Vossel, F. N. v. d. (2001). On the numerical analysis of coronary artery wall shear
stress. Computers in Cardiology, 2, 169–172.
Wong, K., Mazumdar, J., Pincombe, B., Worthley, S. G., Sanders, P., & Abbott, D. (2006). Theoretical modeling of micro-scale biological phenomena in human
coronary arteries. Medical and Biological Engineering and Computing, 44, 971–982.
World Health Organisation, The top 10 causes of death, in, (2018).
Wu, F., Li, X. Y., Chen, W. Q., Kang, G. Z., & Müller, R. (2018a). Indentation on a transversely isotropic half-space of multiferroic composite medium with a
circular contact region. International Journal of Engineering Science, 123, 236–289.
Wu, W.-T., Aubry, N., Massoudi, M., & Antaki, J. F. (2017). Transport of platelets induced by red blood cells based on mixture theory. International Journal of
Engineering Science, 118, 16–27.
Wu, X., von Birgelen, C., Li, Z., Zhang, S., Huang, J., Liang, F., Li, Y., Wijns, W., & Tu, S. (2018b). Assessment of superficial coronary vessel wall deformation
and stress: Validation of in silico models and human coronary arteries in vivo. The International Journal of Cardiovascular Imaging, 34, 849–861.
Yang, C., Bach, R. G., Zheng, J., Naqa, I. E., Woodard, P. K., Teng, Z., Billiar, K., & Tang, D. (2009). In Vivo IVUS-based 3-D fluid–structure interaction models
with cyclic bending and anisotropic vessel properties for human atherosclerotic coronary plaque mechanical analysis. IEEE Transactions on Biomedical
Engineering, 56, 2420–2428.
Zeng, D., Ding, Z., Friedman, M. H., & Ethier, C. R. (2003). Effects of cardiac motion on right coronary artery hemodynamics. Annals of Biomedical Engineering,
31, 420–429.
Zhang, J.-B., & Kuang, Z.-B. (20 0 0). Study on blood constitutive parameters in different blood constitutive equations. Journal of Biomechanics, 33, 355–360.
Zhao, J., Jesper Andreasen, J., Yang, J., Steen Rasmussen, B., Liao, D., & Gregersen, H. (2007). Manual pressure distension of the human saphenous vein
changes its biomechanical properties—implication for coronary artery bypass grafting. Journal of Biomechanics, 40, 2268–2276.
Zheng, J., El Naqa, I., Rowold, F. E., Pilgram, T. K., Woodard, P. K., Saffitz, J. E., & Tang, D. (2005). Quantitative assessment of coronary artery plaque vul-
nerability by high-resolution magnetic resonance imaging and computational biomechanics: A pilot study ex vivo. Magnetic Resonance in Medicine, 54,
1360–1368.
Zhu, H., Ding, Z., Piana, R. N., Gehrig, T. R., & Friedman, M. H. (2009). Cataloguing the geometry of the human coronary arteries: A potential tool for
predicting risk of coronary artery disease. International Journal of Cardiology, 135, 43–52.
Zou, W. N., & He, Q. C. (2018). Revisiting the problem of a 2D infinite elastic isotropic medium with a rigid inclusion or a cavity. International Journal of
Engineering Science, 126, 68–96.

You might also like