RESEARCH ARTICLE

Numerical and Analytical Study of Two-

Layered Unsteady Blood Flow through
Catheterized Artery
Akbar Zaman1*, Nasir Ali1, M. Sajid1, Tasawar Hayat2,3
1 Department of Mathematics and Statistics, Faculty of Basic and Applied Sciences, International Islamic
University, Islamabad, 44000, Pakistan, 2 Department of Mathematics, Quaid-I-Azam University, Islamabad,
4400, Pakistan, 3 Nonlinear Analysis and applied Mathematics (NAAM) Research Group, Department of
Mathematics, Faculty of Science, King Abdulaziz University, Jeddah, 21589, Saudi Arabia

* [email protected]

Abstract
a11111
The pulsatile flow of blood in a catheterized blood vessel is analyzed. The flow of blood in
vessel is modeled as the flow of two immiscible fluids. The fluid in the core region is charac-
terized as a non-Newtonian viscoelastic fluid satisfying the constitutive equation of an Old-
royd-B fluid. The fluid in the peripheral region is treated as a Newtonian fluid. The catheter
inside the vessel is modeled as a rigid tube of very small radius. The resulting differential
system for velocity in each region is computed numerically by finite-difference scheme and
OPEN ACCESS analytically by Laplace transform. A comparison of numerical solution with Laplace trans-
Citation: Zaman A, Ali N, Sajid M, Hayat T (2016) form solution is carried out. Various physical quantities of interest through the computed
Numerical and Analytical Study of Two-Layered
velocity are also analyzed.
Unsteady Blood Flow through Catheterized Artery.
PLoS ONE 11(8): e0161377. doi:10.1371/journal.
pone.0161377

Editor: Keng-Hwee Chiam, Bioinformatics Institute,

SINGAPORE
1. Introduction
Medical catheters are used in medical centers to insert fluids or gases to patients or to deplete
Received: February 19, 2016
bodily fluids such as urine. Apart from that they are utilized for the diagnosis and treatment of
Accepted: August 4, 2016
various arterial diseases. Moreover, the monitoring of velocity and pressure gradient during
Published: August 22, 2016 any treatment/diagnostic procedure is also achieved by using catheters. Despite their advan-
Copyright: © 2016 Zaman et al. This is an open tages, there are certain consequences of clinical significance attached with their usage. It is
access article distributed under the terms of the observed that the insertion of catheter in an artery affects the flow field, disturb the pressure
Creative Commons Attribution License, which permits distribution and enhance the resistance to flow. Therefore it is necessary to study such increase
unrestricted use, distribution, and reproduction in any in flow variables due to catheterization.
medium, provided the original author and source are
Very few attempts have been made in the past to study the blood flow through a catheter-
credited.
ized artery. Back [1] estimated the resistance to the flow by assuming that the blood as a New-
Data Availability Statement: All the relevant data tonian behavior. He concluded that even a very small size of angioplasty guide-wire leads to
are given in the manuscript.
sizable increase in flow resistance. In another study Back and Denton [2] provided the esti-
Funding: The authors received no specific funding mates of wall shear stress in coronary angioplasty and discussed its clinical implications. Mac-
for this work. Donald [3] investigated the characteristics of blood flow in catheterized arteries using finite
Competing Interests: The authors have declared difference technique. Some possible effects of catheter on arterial wall are studied by Karahalios
that no competing interests exist. [4]. He observed that such effects become more significant when the annular gap between

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 1 / 19

 Numerical and Analytical Study of Two-Layered Unsteady Blood Flow through Catheterized Artery

catheter and artery becomes narrow. The influence of catheterization on blood flow in a curved
artery was discussed by Jayaraman and Tiwari [5].
The simultaneous effects of non-Newtonian nature of blood and catheter on flow character-
istics are also reported by some researchers. For instance Dash et al. [6] utilized constitutive
equation of Casson model to estimate the increase in flow resistance due to catheter in a nar-
row artery for both steady and pulsatile flow cases. The analysis of Dash et al. [6] was extended
by Sankar and Hemalatha [7, 8] for Herschel-Bulkley model.
On the other hand the unsteady blood flow through stenosed arteries is also attended by
several investigators. For instance, Mekheimer and El Kot [9] analyzed the effects of Hall cur-
rent on blood flow through an artery under stenotic conditions. Mustapha et al. [10] examined
the characteristics of arterial constrictions and body acceleration in unsteady blood flow. In
another attempt, Mustapha et al. [11] explored the blood flow in irregular multi-stenosed arter-
ies. Here emphasis is given to the effect of applied magnetic field. A mathematical model for
generalized Newtonian blood flow with irregular arterial stenosis is also developed and simu-
lated by Mustapha et al. [12]. Abdullah et al. [13] addressed the influence of magnetic field on
blood flow with irregular stenosis. A mathematical model for axisymmetric artery via cosine
shaped stenosis is computed by Shahed et al. [14]. Apart from above mentioned studies, some
investigators have also explored the effects of catheter on Newtonain and Non-Newtonian
flows through stenosed arteries. Mentioned may be made to the studies carried out by Sarkar
and Jayaraman [15], Dash et al. [16], Reddy et al. [17], Srivastava and Rastogi [18] and Muthu
et al. [19].
It is noted from the available literature that blood flowing through catheterized arteries is
taken as a single-phase Newtonian or non-Newtonian fluid. However, due to accumulation
of red blood cells in the center of the larger arteries, a cell-depleted layer exists near the vessel
wall. To account for such a situation, many researchers have treated the blood as a two-
phase fluid where the core region is modeled as non-Newtonian fluids while the cell-depleted
layer is assumed as a Newtonian liquid. Additionally there are experimental evidences that
blood exhibits viscoelastic properties under certain conditions [20, 21]. Thurston [22] was
the first to identify the viscoelastic nature of blood. He developed an extended Maxwell
model, which is applicable to one-dimensional flow [23]. Yeleswarapu et al. [24] and Yeles-
warapu [25] proposed an Oldroyd-B fluid model with three parameters for studying the
blood flow. It is generally accepted that blood is slightly viscoelastic, and its effect was
ignored in most of the computational fluid dynamics studies. At low shear rates, aggregate of
RBCs behave like solid bodies and has abilities to store elastic energy. However, at high shear
rates due to fluid-like behavior of RBCs, viscoelastic effects are also less prominent. There-
fore, viscoelastic models are more adequate for blood flow at low shear stress and in oscil-
latory flow conditions [26].
Motivated by above facts, we are interested to investigate the flow characteristics of blood in
a narrow catheterized artery by considering blood as a two-phase fluid. The aggregate of red
blood cells in the core region is modeled by Oldroyd-B fluid while the periphery region is
assumed to behave as a Newtonian fluid. The compliant nature of the artery is not incorporated
in the present study. A priori estimate about how the compliant nature of the artery wall may
affect the flow rate, stability etc may be difficult to obtain. However, the interested reader are
referred to reference [27] for further details. The layout of the paper is as follows: Fundamental
laws and geometry of pulsatile flow problem are described in section 2. The governing equa-
tions of the problem and appropriate boundary conditions are derived in section 3. Solution
obtained via Laplace transform is presented in section 4. Numerical solution through finite dif-
ference method is obtained in section 5. Section 6 comprises results and discussion for various
values of parameters of interest. Finally, some conclusions are drawn in section 7.

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 2 / 19

 Numerical and Analytical Study of Two-Layered Unsteady Blood Flow through Catheterized Artery

2. Flow Equations
The continuity and momentum equations governing the flow of an incompressible fluid are
r  u ¼ 0; ð1Þ

du
r ¼ r  T þ rb; ð2Þ
dt
where u is the fluid velocity, ρ is the density, T is the Cauchy stress, b is the body acceleration
and d/dt is the material time derivative given by
dðÞ @ðÞ
¼ þ ðu  rÞ: ð3Þ
dt @t

The Cauchy stress tensor of an Oldroyd-B fluid is [26]

T ¼ p I þ S; ð4Þ

in which p is the pressure, I is the identity tensor and the extra stress tensor S satisfies the fol-
lowing expression


DS DA1
S þ l1 ¼ m A1 þ l2 ; ð5Þ
Dt Dt

where λ1, λ2 are the relaxation and retardation times respectively, μ is the co-efficient of viscos-
ity, A1 is the first Rivilin-Ericksen tensor and D/Dt is the contravariant convective derivative.
The expression for D/Dt and A1 are defined by
DðÞ dðÞ
¼  LðÞ  ðÞLT ; ð6Þ
Dt dt

Where A1 ¼ L þ LT ; ð7Þ

L ¼ ru: ð8Þ

3. Mathematical Formulation
Consider an axially symmetric, unsteady, uni-directional, incompressible and two-phase flow
of blood through an artery of radius R in which a catheter is inserted. The radius of catheter is
kR (0 < k < 1). The blood in core and peripheral regions is modeled by Oldroyd-B fluid and
Newtonian fluid, respectively. We use cylindrical coordinates (r,θ,z) to formulate the flow
under consideration. Following [24], it is assumed that the flow is subject to periodic accelera-
tion in the z-direction. The schematic diagram of the catheterized artery is presented in Fig 1.
Since the flow in the rigid catheterized tube is assumed to be unsteady and uni-directional,
therefore we write velocity field in core and peripheral region as follows:
ui ¼ ½0; 0; uðr; tÞ; i ¼ 1; 2: ð9Þ

The momentum equation in this case simplifies to

@u @p 1@
r ¼  þ rGðtÞ þ ðrS Þ; ð10Þ
@t @z r @r rz
where G(t) is periodic acceleration in z-direction. We divide our domain kR  r  R as kR  r
 R1 (a core region) and R1  r  R (a periphery region). For the present problem, the fluid in

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 3 / 19

 Numerical and Analytical Study of Two-Layered Unsteady Blood Flow through Catheterized Artery

Fig 1. Geometry of the catheterized artery.

doi:10.1371/journal.pone.0161377.g001

the core is an Oldroyd-B fluid while in the periphery region it is a Newtonian fluid. Thus the
constitutive equation for the shear stress in periphery region is

@u2
Srz ¼ m2 ; R1  r  R; ð11Þ
@r

where μ2 is the viscosity of the Newtonian fluid. The general form of the constitutive equation
for shear stress in the core region is [26]
   
@ @ @u1
1 þ l1 S ¼ m1 1 þ l2 ; kR  r  R1 : ð12Þ
@t rz @t @r

The pressure gradient can be put in the form [28]:

@p
 ¼ A0 þ A1 cosop t: ð13Þ
@z

Here A0 is the systolic and A1 is the amplitude (diastolic) components of the pressure gradi-
ent, ωp = 2πfp is the circular frequency and fp is the pulse rate frequency. Following [29], we can
write.

GðtÞ ¼ Ag cosðob t þ Þ; ð14Þ

in which Ag is the amplitude, fb is the frequency (ωb = 2πfb) and ϕ is the lead angle of G(t) with

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 4 / 19

 Numerical and Analytical Study of Two-Layered Unsteady Blood Flow through Catheterized Artery

respect to the heart action. Eliminating Srz between (10) and (12), we get in the core region
   
@u1 @2u A A
r1 þ l1 21 ¼ A0 1 þ 1 cosðop tÞ  l1 1 op sinðop tÞ þ r1 Ag ðcosðob t þ Þ  l1 ob sinðob tÞÞ
@t @t A0 A0
 2    ð15Þ
@ u1 @ 3 u1 m1 @u1 @ 2 u1
þ m1 þ l 2 þ þ l 2 ;
@r 2 @r 2 t r @r @rt

and for the periphery region.

 
@u2 A m @u @u
r2 ¼ A0 1 þ 1 cosðop tÞ þ r2 Ag cosðob t þ Þ þ 2 2 þ m2 22 ; ð16Þ
@t A0 r @r @r

where subscript denotes differentiation with respect to the indicated variable.

The boundary and initial conditions for the present flow situation are [26]:

u1 jr¼kR ¼ 0; u2 jr¼1 ¼ 0; u1 jr¼R1 ¼ u2 jr¼R1 ;

   
@ @u2 @ @u1
m2 1 þ l1 ¼ m1 1 þ l2 ; at interface r ¼ R1 ;
@t @r @t @r

u1 ¼ u2 ¼ 0; at t ¼ 0; ð17Þ

The volumetric Flow rate and shear stress at the wall of the tube are respectively given by

ZR1 ZR
Q ¼ 2p u1 r dr þ u2 r dr;
0 R1

 
@u2
ts ¼ m2 : ð18Þ
@r r¼R

The resistance to flow or impedance experienced by flowing blood at any cross-section is [28]:

ð@p=@zÞ
L¼ : ð19Þ
Q

3.1. Dimensionless formulation

We are interested in numerical solution of Eqs 15 and 16 subject to conditions 17, for this we
first normalize these equations by defining [29]
r u op
r ¼ ; u ¼ ; t ¼ t: ð20Þ
R U0 2p

In terms of new variable the momentum equations after dropping the bars takes the follow-
ing form
 
@u1 @2u
a þ l1 21 ¼ B1 ð1 þ e cosð2ptÞÞ  ð2l1 eB1 pÞsinð2ptÞ þ B2 ðcosð2por t þ ÞÞ
@t @t
 2    ð21Þ
@ u1 @ 3 u1 1 @u1 @ 2 u1
 ð2por B2 l1 Þsinð2por tÞ þ þ l2 2 þ þ l2 ;
@r2 @r t r @r @rt

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 5 / 19

 Numerical and Analytical Study of Two-Layered Unsteady Blood Flow through Catheterized Artery

@u2 1 @u2 @u2

g ¼ B1 ð1 þ e cosð2ptÞÞ þ B2 ðcosð2por t þ ÞÞ þ þ 2: ð22Þ
@t r @r @r

The dimensionless boundary and initial conditions become

u1 jr¼k ¼ 0; u2 jr¼1 ¼ 0; u1 jr¼b ¼ u2 jr¼b ; u1 jðr;0Þ ¼ u2 jðr;0Þ ¼ 0
   
@ @u1  @ @u2 ð23Þ
1 þ l2 ¼ m 1 þ l1 ; at interface r ¼ b;
@t @r @t @r

Similarly the volume flow rate and the shear stress in dimensionless form are
0 1
Zb Z1
Q ¼ 2pR2 B@ u1 rdr þ u2 rdr A;
C ð24Þ
k b

 
1 @u2
ts ¼ : ð25Þ
R @r r¼1

The various parameter appearing in Eqs 21–25 are defined as follows [28]:
 
A0 R 2  l1 op l2 op A o
U0 ¼ ; l1 ¼ ; l2 ¼ ; e ¼ 1 ; or ¼ b ;
m1 2p 2p A0 op

r2  m2 A R2 R2 r1 Ag R
r ¼ ; m ¼ ; B1 ¼ 0 ; B2 ¼ r1 Ag ¼ B1 ; b ¼ 1 ;
r1 m1 m 1 U0 m1 U0 A0 R

r 1 op R2 r2 op R2 r1 op R2 r2 r t
a¼ ; g¼ ¼ ¼ a ; t s ¼ s ;
2p m1 2p m2 2p m1 m r1
 m m2 U0

2 r A R2 r A R2 2
r2 r
^ 1 ¼ A 0 R ¼ B1 ; B
B ^ 2 ¼ 2 g ¼ 1 g ¼ r 1 As R ¼ B 2 : ð26Þ
m1 U0 m m m1 U0 m m1 U0 m m1 U0 r1 m m

4. Analytical Solution
Eqs 21 and 22 subject to the initial and boundary condition given in Eq 23 can be solved analyt-
ically by employing Laplace transform. To this end, let us denote U1(r,s) and U2(r,s), respec-
tively, the Laplace transform of u1(r,t) and u2(r,t). Then Eqs 21 and 22 reduce to following
ordinary differential equations in transformed domain
U1r ðs þ l1 s2 Þ
U1rr þ a U ¼ OðsÞ; ð27Þ
r ð1 þ l2 sÞ 1

U2r
U2rr þ  gsU2 ¼ bðsÞ; ð28Þ
r

where
 
1 C B1 B
OðsÞ ¼   ; ð29Þ
ð1 þ sl2 Þ s2 þ 4p2 s s2 þ 4p2 o2r

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 6 / 19

 Numerical and Analytical Study of Two-Layered Unsteady Blood Flow through Catheterized Artery

B ¼ 2pB2 or sin þ 4p2 B2 l1 o2r  B2 cos s; ð30Þ

C ¼ 4p2 B1 l1 e  B1 e s: ð31Þ
 
B1 eB1 s 1
b1 ðsÞ ¼   2 þ B2 ðcoss  sin 2p or Þ 2 : ð32Þ
s s þ 4p 2 s þ 4p2 o2r

The boundary and initial conditions in the transform domain read

U1 ðkÞ ¼ 0; U2 ð1Þ ¼ 0; ð33Þ

U1 ðb1 Þ ¼ U2 ðb1 Þ; ð34Þ

 
1 þ sl2
U2r ðb1 Þ ¼ m U ðb Þ: ð35Þ
1 þ sl1 1r 1

The condition (35) is established by taking Laplace transform of both sides of interface con-
dition in Eq 17 and using the u1t = u2t at t = 0. The differential Eqs 27 and 28 can be readily
solved in terms of Bessel’s functions to get the expressions of U1(r,s) and U2(r,s) as
OðsÞð1 þ l2 sÞ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
U1 ðr; sÞ ¼  þ c1 I0 ðr a sð1 þ l1 sÞ=ð1 þ l2 sÞÞ þ c2 K0 ðr a sð1 þ l1 sÞ=ð1 þ l2 sÞÞ; ð36Þ
asð1 þ l1 sÞ

b1 ðsÞ pffiffiffiffiffi pffiffiffiffi

U2 ðr; sÞ ¼  þ c3 I0 ðr g sÞ þ c4 K0 ðr gsÞ: ð37Þ
gs

The values of arbitrary constants c1–4 can be easily calculated by implementing the bound-
ary conditions (33)–(35). Due to the complicated nature of the solution in transformed
domain, analytical inversion is difficult. Therefore, we have relied on the numerical inversion
using Mathematica Package NumericalInversion.m. The command “Stehfest” is used to get the
numerical values of the inverted solution.

5. Numerical Solution
Eqs 21 and 22 subject to the boundary condition given in Eq 23 are solved numerically using the
finite difference method [30, 31]. The uniformly distributed discrete points in radial direction are
defined as ri = (i−1)Δr, (i = 1,2,. . ..,Nc + 1) such that rðNc þ1Þ ¼ b and ri = (i−(Nc+1))Δr, i = (Nc + 1,
Nc + 2,. . .,N + 1) such that rN+1 = 1, where Δr is the increment in the radial direction. Similarly we
define tj = (j−1)Δt, (j = 1,2,. . ..) as discrete time points with Δt indicating the small time increment.
For problem under consideration we have chosen Δr = 0.025 and Δt = 0.0001. This choice of Δr and
Δt yields results convergent up to order 10−7. Let us denote the discretized value of uk (k = 1,2) at
(ri,tj) by ujk i . In this notation, finite difference formulas for first and second order derivatives read
@uk ujk iþ1  ujk i1
ffi ¼ uk r ;
@r 2 Dr

@ 2 uk ujk iþ1  2ujk i þ ujk i1

ffi 2 ¼ uk r 2 : ð38Þ
@r 2 ðDrÞ

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 7 / 19

 Numerical and Analytical Study of Two-Layered Unsteady Blood Flow through Catheterized Artery

For the time derivative, we define the approximation:

@uk ujþ1  ujk i
ffi ki ¼ uk t ;
@t Dt

@ 2 uk i ukjþ1i  2ujk i þ uj1

¼ 2
ki
¼ uk t 2 ;
@t 2 ðDtÞ

 
@ @uk ujþ1  ujk iþ1  ukjþ1i1 þ ujk i1
ffi k iþ1 ¼ uk tr ;
@t @r 2 Dr Dt

 
@ @ 2 uk k iþ1  2uk i þ uk i1  uk iþ1 þ 2uk i  uk i1
ujþ1 jþ1 jþ1 j j j
ffi 2 ¼ uk tr2 : ð39Þ
@t @r 2 ðDrÞ Dt

Utilizing (38) and (39), Eqs 21 and 22 may be transformed to the following difference equa-
tions:

    "
1 2l1 l1 j1 Dt
u jþ1
1i ¼  u 1þ k
1i  u1 i þ B1 ð1 þ ecosð2pt j ÞÞ  ð2l1 eB1 pÞsinð2pt j Þ
l1 Dt Dt a
1þ
Dt ð40Þ
n o 1 #
l
þ B2 ðcosð2por t k þ ÞÞ  ð2por B2 l1 Þsinð2por t j Þ þ u1 2 þ l2 u1 2 þ ðu Þ þ 2 u1tr ;
r tr r 1r r

 

Dt ^ ^ 2 ðcosð2por t j þ ÞÞ þ 1 ½u2r  þ u2 :
u2 jþ1 ¼ u2 ji þ B 1 ð1 þ ecosð2pt j ÞÞ þ B ð41Þ
i
g r r2

The boundary and initial conditions in the discretize form can be written as follows:
uj11 ¼ uj21 ; uj2Nþ1 ¼ 0; uj1Nc þ1 ¼ uj2Nc þ1 ; u11i ¼ u12i ¼ 0;
ð42Þ
u2kr þ l1 u2ktr ¼ m ðu1kr þ l1 u1ktr Þ; at Nc þ 1;

The partial differential equations governing the flow are solved both numerically and ana-
lytically. The details of the stability of numerical scheme used by us are given in the book by
Hoffman, in which it is mentioned that the stability of explicit finite scheme depends on time
and spatial step sizes. The numerical solution obtained by us fulfill all these requirements
on spatial and temporal step sizes. Following Hoffman [32], we have chosen Δx = 0.025,
Δt = 0.0001. This choice of Δx and Δt yield results convergent up to order 10−7. However it is
observed during the computations that the numerical results are much sensitive to the choice
of involved parameters. Thus for the present problem the stability of the numerical solution
cannot be guaranteed only by full filling the requirements of the temporal and spatial step
sizes. In fact with the present choice of spatial and temporal step sizes a stable numerical solu-
tion is possible only if material parameters λ1 and λ2 are such that 0.1 < λ1 < 0.5 and 0.1 < λ2
< 0.5.

6. Results and Discussion

In this section, we are interested to analyze the graphical results of velocity, flow rate, wall
shear stress and resistance to flow results for different values of non-dimensional variables
namely the catheter radius ratio (k), amplitude of pulsatile pressure gradient (e), relaxation and
retardation time (λ1 and λ2) and generalized Womersley frequency parameter (α). In the

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 8 / 19

 Numerical and Analytical Study of Two-Layered Unsteady Blood Flow through Catheterized Artery

Table 1. Plausible values of involved parameters for different blood vessels [29, 40, 41].
Parameters Arterioles Coronary artery Femoral artery Capillaries
R(mm) 0.685 1.5 5 0.24
B1 1.41 1.41 6.6 6.6
B2 2.07 21.67 4.64 3.4
fp 1.2 1.2 1.2 2.4
fg 1.2 1.2 1.3 1.2
A0 (Pa/mm) 7 698.5 32 20
ρ (Kg/m3) 1050 1050 1050 1050
doi:10.1371/journal.pone.0161377.t001

present analysis the height of interface is assumed independent of the involved parameters and
therefore it is not determined as a part of the problem. Instead we have taken it as a constant.
The primary motivation for this assumption is based the evidence that RBC is migrate away
from the wall leaving a cell-depleted layer near the wall. In such situation one can model blood
flow in two stages (i) a peripheral layer modeled as a Newtonian liquid and (ii) a core region
modeled as a non-Newtonian fluid. The thickness of both layer is assumed constant. Based on
this fact several authors for instance Maji and Nair [33], Massoudi and Phouc [29], Ikbal et al.
[31], Bugliarello and sevilla [34], Sulkla et al. [35], Akay an Kaye [36], Pralhad and Schultz
[37], Sajid et al. [38] and Srivastava and Sexena [39] have used two-layered fluid model with
constant peripheral thickness to study blood flow through arteries. We have followed these
studies in our analysis. However, we have also shown the results for different interface posi-
tions in Figures (4, 10 and 11). The physiologically relevant values of the above mentioned
parameters is listed in Tables 1 and 2.
The radius ratio k, characterizing the radius of the catheter is assumed to be in the range
0.1–0.5. The length of separation point between peripheral layer and the core region is repre-
sented by β. In this way, we get the thickness of the peripheral layer as 1−β. In the present anal-
ysis, β are taken as 0.9.
Fig 2 illustrates the profiles of velocity at different instants of time within a single cardiac
cycle calculated through both numerical and analytical solutions. A pleasing agreement
between both solutions is observed. This agreement also demonstrates the validity of our
numerical scheme. Fig 2 further indicates an increase in velocity when t increases from 0.1 to
0.2 in systolic phase. However in diastolic phase i.e. at t = 0.3 a decreasing trend in velocity is
observed. The increasing behavior of velocity in systolic phase while opposite trend in diastolic
phase is clearly due to the pulsatile pressure gradient produced by the pumping action of heart.
Fig 3 shows the effects of the catheterization at t = 0.3 on the velocity distribution in the
two-fluid model of blood flow. It is noticed that at a given instant of time t = 0.3 with β = 0.9
and the increasing values of the catheter radius ratio k, the velocity decreases considerably. It
means that due to the insertion of catheter, the magnitude of the velocity reduces significantly.
The velocity distribution for different values of the interface position β is shown in Fig 4 for
k = 0.1. It is found that for a given value of k, the velocity increases significantly with decreasing
β, whereas the behavior is opposite when the width of the peripheral layer decreases (i.e., when
the value of β decreases).

Table 2. Values of Womersley number for different blood vessels [42].

Blood vessel Diameter(mm) α
Femoral artery 12.9 3.5
Arterioles 1.37 0.04
Capillaries 0.48 0.05
doi:10.1371/journal.pone.0161377.t002

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 9 / 19

 Numerical and Analytical Study of Two-Layered Unsteady Blood Flow through Catheterized Artery

Fig 2. Variation of velocity with radial distance for (α = 0.5, β = 0.9, B1 = 2, B2 = 1, k = 0.1, e = 0.3).
doi:10.1371/journal.pone.0161377.g002

We have shown the effects of relaxation and retardation times on the velocity profile in Fig
5. It is readily observed that the role of λ1 here is to increase the magnitude of velocity while λ2
counters the effects of λ1 and it decreases the magnitude of velocity.
Fig 6 shows the variation of dimensionless velocity for different values of Womersley num-
ber α when β = 0.9, t = 0.3 and e = 0.5. It is seen that at a given time instant the magnitude of
velocity increases as the value of Womersley number decreases. Fig 7 shows the velocity profile
of blood for different values of body acceleration. It is evident from this figure that the magni-
tude of the velocity increases with the increase of amplitude of body acceleration.
Fig 8 depicts the variation of flow rate for different values of λ1 and λ2 when k = 0.1, e = 0.5
and β = 0.9. Generally it is clear that the flow rate increases as t increases from 0 to 2.3 and then
achieves its steady state condition. It means that flow rate fluctuates around its mean value
with constant frequency and amplitude after achieving steady state condition. It is further
observed from this figure that the amplitude of oscillations in flow rate increases (decreases) by
increasing λ1(λ2). Similarly as expected that amplitude of oscillations in flow rate increases by
increasing the amplitude of pulsatile pressure gradient (Fig 9). Moreover it also evident from
Fig 9 that oscillation in flow rate diminishes by increasing Womersley number.
Similarly, the variation of dimensionless flow rate for different values of catheter radius
ratio k and interface position β is shown in Fig 10. It is noticed that the flow rate decreases by
increasing both catheter radius and peripheral layer thickness.
In pulsatile flow, the non-dimensional wall shear stress τw can be calculated from Eq 25.
The variation of wall shear stress in a cycle of oscillation for different values of amplitude e and
peripheral thickness β with k = 0.5 is shown in Fig 11. The wall shear stress fluctuates around

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 10 / 19

 Numerical and Analytical Study of Two-Layered Unsteady Blood Flow through Catheterized Artery

Fig 3. Variation of velocity with radial distance for different values of catheter radius (t = 0.3, λ1 = 0.2, λ2 = 0.1, α = 0.5, β = 0.9,
B1 = 2, B2 = 1, e = 0.3).
doi:10.1371/journal.pone.0161377.g003

its mean value as time increases. It is also noticed that amplitude of the wall shear is propor-
tional to the magnitude of e. Its value increases with the increase in the magnitude of e while it
shows opposite trend with the increase of peripheral radius length β.
Fig 12 shows the variation of wall shear stress in a cycle of oscillation for different values of
λ1, λ2 and k with β = 0.9. This figure indicates that the magnitude of wall shear stress decreases
by increasing λ1. while it is suppressed by increasing λ2. Moreover it is also observed through
Fig 12 that the magnitude of the wall shear increases with an increase in the radius of catheter.
The longitudinal impedance (Λ) of the artery is calculated using Eq 19 and its variation dur-
ing a flow cycle for different values of amplitude k with e = 0.5 and β = 0.9 is illustrated in Fig
13. It is observed that these profiles follow an opposite trend as compare to the flow rate pro-
files which is expected from the formula of resistance to flow given by expression (19). It is con-
cluded from these profiles that the magnitude of resistance to flow increases with the increase
of the size of catheter.

7. Conclusion
Two-layer mathematical model is developed for pulsatile flow of blood with the effects of body
acceleration through a catheterized artery using Oldroyd-B fluid model in the core region and
Newtonian model in the peripheral region. The analysis is based on the solution of a linear par-
tial differential equation corresponding to each region. Assuming the continuity of velocity

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 11 / 19

 Numerical and Analytical Study of Two-Layered Unsteady Blood Flow through Catheterized Artery

Fig 4. Variation of velocity with radial distance with different peripheral layer position (t = 1, e = 0.3, λ1 = 0.2, λ2 = 0.1, α = 0.5,
B1 = 2, B1 = 1).
doi:10.1371/journal.pone.0161377.g004

and no-slip conditions, the governing equation are solved in each region by employing finite
difference technique. The validity of numerical solution is checked by comparing it with the
analytical solution obtained through Laplace transform. An excellent agreement between both
solutions is observed. After validation, numerical solution is further utilized to obtain various
flow quantities. It is found that the resistance to flow or impendence, velocity, flow rate and
wall shear stress are greatly affected by blood rheology, catheter radius and amplitude of pulsa-
tile pressure gradient. More precisely a small increment in the catheter radius significantly
decreases the magnitude of blood velocity and as a result of that a decrement in the flow rate is
observed. As a consequence of decrease in velocity the wall shear stress and resistance to flow
also increase considerably. Another important observation is that such quantitative measure-
ment varies in magnitude by changing the rheological parameters of blood. In fact for a given
catheter radius the flow rate of blood increases by increasing dimensionless relaxation time and
as a result the resistance to flow decreases. Such observation may have certain implications for
instance, the rheology of blood can be controlled to maintain the same flow in an artery with
catheter as that in an artery without catheter. The results obtained through present computa-
tions may be related to hypertension. In normal circumstances/conditions, the amplitude of
pressure gradient maintain the sufficient flow of blood in arteries for normal function of the
organs. However, in situation where the person/human is subject to periodic body acceleration
the results are quite different. In such situation the superimposed periodic oscillation due to
body acceleration increase the amplitude of flow rate produced by pulsatile pressure gradient.
This increase in the amplitude of the flow rate may results in hypertension, nausea, loss of
vision and headache. On the contrary, the magnitude of the flow rate decreases with increasing

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 12 / 19

 Numerical and Analytical Study of Two-Layered Unsteady Blood Flow through Catheterized Artery

Fig 5. Variation of velocity with radial distance with (t = 0.3, B2 = 2, e = 0.5, α = 0.5, B1 = 2 β = 0.9).
doi:10.1371/journal.pone.0161377.g005

Fig 6. Variation of velocity with radial distance with the variation of α (t = 0.3, B1 = 2, λ1 = 0.2, λ2 = 0.2, e = 0.5, B2 = 2, β = 0.9).
doi:10.1371/journal.pone.0161377.g006

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 13 / 19

 Numerical and Analytical Study of Two-Layered Unsteady Blood Flow through Catheterized Artery

Fig 7. Variation of velocity with radial distance with the variation of B2 (t = 0.3, B1 = 2, λ1 = 0.2, λ2 = 0.2, e = 0.5, β = 0.9, α = 0.5).
doi:10.1371/journal.pone.0161377.g007

Fig 8. Variation of flow rate in a cycle of oscillation for different values of λ1 and λ2 with (β = 0.9, B1 = 2, e = 0.5, B2 = 0.0, α = 0.5).
doi:10.1371/journal.pone.0161377.g008

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 14 / 19

 Numerical and Analytical Study of Two-Layered Unsteady Blood Flow through Catheterized Artery

Fig 9. Variation of flow rate in a cycle of oscillation for different values of e and α with (e = 0.5, λ1 = 0.2, λ2 = 0.2, β = 0.9, k = 0.3,
B2 = 2, B2 = 1).
doi:10.1371/journal.pone.0161377.g009

Fig 10. Variation of flow rate in a cycle of oscillation for different values of k and β with (e = 0.5, λ1 = 0.2, λ2 = 0.1, α = 0.3, B1 = 2,
B1 = 1).
doi:10.1371/journal.pone.0161377.g010

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 15 / 19

 Numerical and Analytical Study of Two-Layered Unsteady Blood Flow through Catheterized Artery

Fig 11. Variation of wall shear stress in a cycle of oscillation for different values of e and β with (e = 0.5, λ1 = 0.2, λ2 = 0.1, α = 0.3,
k = 0.3, B2 = 1, B1 = 2).
doi:10.1371/journal.pone.0161377.g011

Fig 12. Variation of wall shear stress in a cycle of oscillation for different values of k, λ2 and λ1 (β = 0.9, e = 0.5, B1 = 2, B2 = 1.0,
α = 0.5).
doi:10.1371/journal.pone.0161377.g012

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 16 / 19

 Numerical and Analytical Study of Two-Layered Unsteady Blood Flow through Catheterized Artery

Fig 13. Variation of longitudinal impedance in a cycle of oscillation for different values of k with (β = 0.9,
e = 0.1, B2 = 0, λ1 = 0.2, λ2 = 0.1, α = 0.3).
doi:10.1371/journal.pone.0161377.g013

the catheter radius. In such situation the impedance/ resistance to flow rate increases resulting
in hypotension. Although the significance of the present study is limited but still this study
throws some light on the flow behavior of blood. In fact this study presents the effects interface
position, non-Newtonian nature and an inserted catheter on flow features. Analytical as well a
numerical solution is also presented in this study. The analytical solution may be used as a
benchmarking for more sophisticated 3D CFD analysis accounting pointing by the worthy
reviewer.

Author Contributions
Conceptualization: AZ NA MS TH.
Data curation: AZ NA MS TH.
Formal analysis: AZ NA MS TH.
Funding acquisition: AZ NA MS TH.
Investigation: AZ NA.
Methodology: AZ NA.
Project administration: AZ NA TH.
Resources: AZ NA MS TH.
Software: AZ NA.
Supervision: AZ NA.

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 17 / 19

 Numerical and Analytical Study of Two-Layered Unsteady Blood Flow through Catheterized Artery

Validation: AZ NA MS TH.
Visualization: AZ NA MS TH.
Writing - original draft: AZ NA.
Writing - review & editing: AZ NA.

References
1. Back L. H., “Estimated mean flow resistance increase during coronary artery catheterization”, Journal
of Biomechanics, 27, pp.169–175 (1994). PMID: 8132684
2. Back L. H., Denton T. A., “Some arterial wall shear stress estimates in coronary angioplasty”, Advanced
Biological Engineering, 22, pp. 337–340 (1992).
3. MacDonald D. A., “Pulsatile flow in a catheterized artery”, Journal of Biomechanics, 19, pp. 239 (1986).
PMID: 3700436
4. Karahalios G. T., “Some possible effects of a catheter on the arterial wall", Medical Physics, 17, pp.
922 (1990). PMID: 2233580
5. Jayaraman G., Tiwari K., “Flow in a catheterized curved artery”, Medical & Biological Engineering &
Computing, 33 (1995).
6. Dash R. K., Jayaraman G., Metha K. N., “Estimation of increased flow resistance in a narrow catheter-
ized artery–A theoretical model”, Journal of Biomechanics, 29, pp. 917–930 (1996). PMID: 8809622
7. Sankar D. S., Hemalatha K., “Pulsatile flow of Herschel–Bulkley fluid through catheterized arteries–A
mathematical model”, Applied Mathematical Modeling, 31, pp.1497–1517 (2007).
8. Sankar D. S., Hemalatha K., “A non-Newtonian fluid flow model for blood flow through a catheterized
artery—Steady flow”, Applied Mathematical Modeling, 31, pp.1847–1864 (2007).
9. Mekheimer K. S., El Kot M. A., “Magnetic field and Hall currents influences on Blood flow through a ste-
notic artery”, Applied Mathematical Mechanics, English-Edition, 29, pp. 1093–1104 (2008).
10. Mustapha N., Chakravarty S., Mandal P. K., Amin N., “Unsteady response of blood flow through a cou-
ple of irregular arterial constrictions to body acceleration”, Journal of Mechanics in Medicine and Biol-
ogy, 8, pp. 395–420 (2008).
11. Mustapha N., Amin N., Chakravarty S., Mandal P. K., “Unsteady magnetohydrodynamic blood flow
through irregular multi-stenosed arteries”, Computers in Biology Medicine, 39, pp. 896–906 (2009).
doi: 10.1016/j.compbiomed.2009.07.004 PMID: 19665698
12. Mustapha N., Amin N., Mandal P. K., Abdullah I., Hayat T., “Numerical simulation of generalized Newto-
nian blood flow past a multiple of irregular arterial stenosis”, Numerical Methods for Partial Differential
Equations, 27, pp. 960–981 (2009).
13. Abdullah I., Amin N., Hayat T., “Magnetohydrodynamic effects on blood flow through an irregular steno-
sis”, International Journal for Numerical Methods in Fluids, 67, pp. 1624–1636 (2011).
14. Shahed S. R., Ali M., Saha S., Akhter M. N., “Numerical Study on unsteady flow field of arterial steno-
sis”, Procedia Engineering., 90, pp. 339–345 (2014).
15. Sarkar A., Jayaraman G., “Correction to flow rate–pressure drop relation in coronary angioplasty:
steady streaming effect”, Journal of Biomechanics, 31, pp. 781(1998). PMID: 9802778
16. Dash R. K., Jayaraman G., Mehta K. N., “Flow in a catheterized curved artery with stenosis”, Journal of
Biomechanics, 32, pp. 49–61 (1999). PMID: 10050951
17. Reddy J. V. R., Srikanth D., Murthy S. V. S. S. N. V. G. K., “Mathematical modeling of pulsatile flow of
blood through catheterized un-symmetric stenosed artery—Effects of tapering angle and slip velocity”,
Eur. J. of Mech. B/Fluids, 48 (2014) 236–244.
18. Srivastava V. P., Rastogi R., “Blood flow through a stenosed catheterized artery: Effects of hematocrit
and stenosis shape”, Computers & Mathematics with Applications, 59, pp.1377–1385 (2010).
19. Muthu P., kumar B. V. R., and Chandra P., “A study of Micropolar fluid in an annular tube with applica-
tion to blood flow”, Journal of Mechanics in Medicine and Biology, 4, pp. 561–576 (2008).
20. Thurston G. B., Henderson N. M., “The kinetics of viscoelastic changes due to blood clot formation”,
Theoretical and Applied Rheology, 112, pp. 17–21 (1992).
21. Chien S., King R. G., Skalak R., Usami S., Copley A. L., “Viscoelastic properties of human blood and
red cell suspension”, Biorheology, 12, pp. 341–346 (1975). PMID: 1212514
22. Thurston G. B., “Viscoelasticity of human blood”, Biophysics Journal, 12, pp. 1205 (1972).

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 18 / 19

 Numerical and Analytical Study of Two-Layered Unsteady Blood Flow through Catheterized Artery

23. Thurston G. B., “Rheological parameters for the viscosity, viscoelasticity and Thixotropy of blood”, Bior-
heology, 16, pp. 149 (1979). PMID: 508925
24. Yeleswarapu K. K., Kamaneva M. V., Rajagopal K. R., Antaki J. F., “The flow of blood in tubes: Theory
and experiment”, Mechanics Research Communications, 25, pp. 257–262 (1998).
25. Yeleswarapu, K. K., “Evaluation of continuum models for characterizing the constitutive behavior of
blood”, Ph.D. thesis, University of Pittsburgh, (1996).
26. Yilmaz F., Gundogdu M. Y., “A critical review on blood flow in large arteries; relevance to blood rheol-
ogy, viscosity models, and physiologic conditions”, Korea-Australia Rheology Journal, 20, pp. 197–
211 (2008).
27. Canic S., Tambaca J., Guidoboni G., Mikelic A., Hartley C. J., and Rosenstrauch D., Modeling visco-
elastic behavior of arterial walls and their interaction with pulsatile blood flow, Society for Industrial and
Applied Mathematics, 67, pp. 164–193 (2006).
28. Burton A. C., “Physiology and biophysics of the circulation” Introductory Text, Year Book Chicago:
Medical Publisher, (1966).
29. Massoudi M., Phuoc T. X., “Pulsatile Flow of a blood using second grade fluid model”, Computers and
Mathematics with Applications, 56, pp. 199–211 (2007).
30. Mandal P. K., “An unsteady analysis of non-Newtonian blood flow through tapered arteries with a steno-
sis”, International Journal of Non-Linear Mechanics, 40, pp. 151–164 (2005).
31. Ikbal M. A., Chakravarty S., Mandal P. K., “Two-layered micropolar fluid flow through stenosed artery:
Effect of peripheral layer thickness”, Computers and Mathematics with Applications, 230, pp. 243–259
(2009).
32. K. A. Hoffmann, S. T. Chiang, Computational Fluid Dynamics, A Publication of Engineering Edition Sys-
tem, Wichita, Kansas USA, 1, pp. 67208–1078 (2000).
33. Majhi S. N., Nair V. R., Pulsatile flow of third grade fluids under body acceleration modeling blood flow,
Int. J. Eng. Sci. 32, pp. 839–846 (1994).
34. Bugliarello G., Sevilla J., Velocity distribution and other characteristics of steady and pulsatile blood
flow in fine glass tubes, Biorheol. 7, pp. 85–107 (1970).
35. Shukla J. B., Parihar R. S., Rao B. R. P., Effects of peripheral layer viscosity on blood flow through the
artery with mild stenosis, Bull. Math. Biol. 42, pp. 797–805 (1980). PMID: 7459493
36. Akay G., Kaye A., Numerical solution of time dependent stratified two-phase flow of micropolar fluids
and its application to flow of blood through fine capillaries, Int. J. Eng. Sci., 23, pp.265–276 (1985).
37. Pralhad R. N., Schultz D. H., Two-layered blood flow through stenosed tubes for different diseases,
Biorheol. 25, pp. 715–726 (1988).
38. Sajid M., Zaman A., Ali N., and Siddiqui A. M., Pulsatile Flow of Blood in a Vessel Using an Oldroyd-B
fluid, I. J. Non. Sci. Num. Simu., 16, pp. 197–206 (2015).
39. Srivastava V. P., Saxena M., Two-layered model of Casson fluid flow through stenotic blood vessels:
Application to the cardiovascular system, J. Biomech. 27, pp. 921–928 (1994). PMID: 8063842
40. Chaturani P., Palanisamy V., Pulsatile flow of power-law fluid model for blood flow under periodic body
acceleration, Bio-rheo. 27, pp. 747–758(1990).
41. Chaturani P., Palanisamy V., Pulsatile flow of blood with periodic body acceleration, Int. J. Eng. Sci.
29 pp. 113–121 (1991).
42. Fung Y. C., Biomechanics Circulation, 2nd ed., Springer-Verlag, New York, USA, pp. 571 (1997).

PLOS ONE | DOI:10.1371/journal.pone.0161377 August 22, 2016 19 / 19