J Artif Organs (2010) 13:189–196
DOI 10.1007/s10047-010-0518-8
ORIGINAL ARTICLE
A new pulse duplicator with a passive fill ventricle
for analysis of cardiac dynamics
Yoshimasa Yokoyama • Osamu Kawaguchi •
Tadahiko Shinshi • Ulrich Steinseifer •
Setsuo Takatani
Received: 15 January 2010 / Accepted: 31 August 2010 / Published online: 28 September 2010
Ó The Japanese Society for Artificial Organs 2010
Abstract A new pulse duplicator was designed for
evaluation of the performance of ventricular assist devices
through pressure–volume (P–V) diagrams of the native
heart. A linear drive system in combination with a pusher-
plate mechanism was designed as a drive system to
implement the passive fill mechanism during diastole of the
mock ventricle. The compliances of the native heart during
both diastole and systole were simulated by placing a
ventricle sack made of soft latex rubber in a sealed
chamber and by varying the air-to-fluid volume ratio inside
the chamber. The ratio of the capacities of the systemic
venous and pulmonary circuits was adjusted to properly
reflect the effects of volume shift between them. As the air-
to-fluid volume ratio was varied from 1:12.3 to 1:1.58, the
contractility of the ventricle expressed by Emax varied from
1.75 to 0.56 mmHg/ml with the mean V0 of 4.58 ml closely
mimicking those of native hearts (p \ 0.05). Because the
Emax value of the normal human heart ranges from 1.3 to
1.6, with a value below 1.0 indicating heart failure, the
Y. Yokoyama  S. Takatani (&)
Department of Artificial Organs,
Institute of Biomaterials and Bioengineering,
Tokyo Medical and Dental University, 2-3-10 Surugadai,
Kanda, Chiyoda-ku, Tokyo 101-0062, Japan
e-mail: takatani.ao@tmd.ac.jp
O. Kawaguchi
Department of Cardiovascular Surgery,
Toyoda Kosei Hospital, Toyota, Aichi, Japan
T. Shinshi
Institute of Precision Engineering,
Tokyo Institute of Technology, Yokohama, Japan
U. Steinseifer
AME Helmholtz Institute of Biomedical Engineering,
RWTH Aachen, Aachen, Germany
mock ventricle is applicable in simulating the dynamics of
the normal heart and the sick heart. The P–V diagram
changes seen with rotary blood pump assistance revealed
changes similar to those reported by other workers. The
effects of the ventricular assist device, either pulsatile or
continuous flow, on cardiac dynamics can be easily simu-
lated with this system to derive design criteria for clinical
circulatory assist devices.
Keywords Pulse duplicator  P–V diagram  Heart
function  Arterial compliance  Ventricular assist device
(VAD)  Emax
Introduction
The pulse duplicator can be used to design and evaluate
ventricular assist devices or heart valve prostheses before
animal studies and clinical trials. The pulse duplicator must
thus represent physiological characteristics of the native
heart as well as the circulatory system. One pioneering
system is the mock loop of the Helmholtz Institute Aachen
(HIA) developed by Reul et al. in 1975 [1], which com-
prises a hydraulic pump, a chamber, a Windkessel com-
pliance, and a resistance element. Because the pulsatile
flow generated by a hydraulic pump is affected by the
compliance, resistance, and inertia elements of the system
external to the heart, the coupling between the ventricular
and arterial systems in the HIA pulse duplicator was
accurately designed to evaluate heart valve prostheses [2].
In simulation of cardiac dynamics, in 1998 Ferrari et al.
showed how to vary ventricular pressure–volume
(P–V) diagrams and Emax (the index to indicate the cardiac
elasticity [3]) by controlling the end-systolic and end-dia-
stolic volumes of the mock ventricle and effective arterial
123
190
elastance (Ea). The model was used to predict changes in
cardiac chamber volume when ventricular assist devices
were used to support the failing ventricle [4]. For dynamic
evaluation of left heart assist devices, the system developed
by Pantalos et al. and Koenig et al. in 2004 mimicked the
normal and failing heart characteristics by utilizing pneu-
matically operated ventricles that were controlled via a
feedback mechanism [5–7]. In 2008, Colacino et al. [8].
developed a mock circulation system by use of a mathe-
matical technique, which added resistive and inductive
terms to the classical Suga–Sagawa elastance model [9] for
the whole cardiac cycle.
Although the literature is filled with abundant reports, a
common and major drawback of the mock circulatory
systems reported above, however, is the active fill mech-
anism in the ventricle as a result of the use of pneumati-
cally operated blood pumps. The native heart passively fills
with the preload pressure during the diastolic phase, fol-
lowed by active contraction. Takatani et al. [10] in 1981
reported use of passive-fill ventricles to simulate total cir-
culation, where the elastance of the left and right hearts
was represented by the pusher-plate type blood pumps
vented to the atmosphere. Although excellent sensitivity to
filling pressure during diastole was shown with the vented
system, alteration in the P–V diagrams was difficult to
achieve once the pump design was fixed with the vented
pump system.
To improve the performance of pulse duplicators,
particularly the ventricular performance, the pulse dupli-
cator designed in this study incorporates a servomotor
coupled with a pusher-plate actuation-decoupling mecha-
nism to implement passive filling of the ventricle. A
ventricle sack fabricated with latex rubber was placed
inside a sealed chamber containing a mixture of air and
fluid to achieve physiological levels of cardiac elastance
and to express various levels of contractility, denoted
Emax, by varying the air-to-fluid volume ratio. An appro-
priate combination of the Windkessel compliance of the
systemic, venous, and pulmonary systems, and the inertia
effects in the arterial system as proposed by Reul et al.
[1]. were incorporated into the system to obtain physio-
logical arterial pressure waveforms. Furthermore, to
properly couple the mock ventricle with the external cir-
cuit, a left–right heart output balance mechanism was
implemented using an appropriate ratio of the systemic to
pulmonary venous compliance.
This study validated the performance of a newly
developed pulse duplicator with a passive fill ventricle
having adjustable P–V relationships, its performance to
control cardiac contractility Emax by simply varying the
air-to-fluid volume ratio inside the LV chamber, and
P–V diagram changes during left heart bypass using a rotary
blood pump.
123
J Artif Organs (2010) 13:189–196
Materials and methods
Mock circulation system
The configuration of the newly designed mock circulation
system are shown in Fig. 1. The system comprises a ven-
tricular model denoted , an arterial system model `, a
venous system reservoir ´, a centrifugal pump representing
the right heart ˆ, and a pulmonary venous compliance ˜; a
venous system reservoir and pulmonary venous compliance
are opened to atmosphere in order to remove the effect of
arterial pulsatility. The system is instrumented with ultra-
sonic flow probes at the inlet and outlet ports (Transonic
Systems, NY, USA) of the left ventricle and pressure
transducers (Edwards Lifesciences, CA, USA) with mea-
surement points depicted in Fig. 1. Pressure and flow rate are
recorded from which P–V diagrams are constructed on a
digital oscilloscope (Yokogawa Denshikiki, Tokyo, Japan).
For simulating left ventricular assistance, a continuous flow
pump (Gyrop Pump C1E3, Kyocera, Kyoto, Japan) is con-
nected from the apex of the ventricular model to the simu-
lated aorta (Þ in Fig. 1); it is clamped at A in Fig. 1 while
the performance of the ventricular model is evaluated.
Passive fill ventricle
The passive-fill ventricular model is shown in Fig. 2. A
ventricular sack (` in Fig. 2) made with latex rubber and
with an unstretched volume of 100 ml was placed inside a
sealed chamber containing a mixture of air and fluid ( in
Fig. 2); it has a u28 mono-leaflet tilting disk Bjork Shiley
valve (Shiley, Irvine, CA, USA) in its inlet port and a u25
bi-leaflet SJM regent valve (St Jude Medical, NJ, USA) in
the outlet port. The air-to-fluid volume ratio was adjusted
to control ventricular elastance during systole and diastole.
Fig. 1 Configuration of a mock circulation system comprising a left
ventricular model , an arterial system model `, a venous system
reservoir ´ (diameter of cross section u10), a centrifugal pump
simulating the right heart ˆ, pulmonary circuit reservoir ˜ (diameter
of cross section is u14), and gyro pump (C1E3) Þ. The figure
indicates the measurement point of pressure and flow rate
J Artif Organs (2010) 13:189–196 191

uncompressed air volume VAir Unload was established

with the left atrial pressure PLA of 10 mmHg. Since the air-
to-fluid volume ratio inside the chamber determined the
E(t)Air, it changed as the elastance of the E(t)LV varied.

Coupling between the mock ventricle, aortic system,

and left atrial system

The pressure change of the mock ventricle, LVP, is

expressed as a function of the compliance of the sack CSack,
the air inside the chamber, CAir, and the ventricle volume,
LVV, as follows:
1 LVV  LVVUnload
LVP ¼ ðLVV  LVVUnload Þ ¼ ; ð4Þ
CV CAir þ CSack
where LVVUnload is the unstressed sack volume, and CV is
the compliance of the ventricle, expressed as the sum of
Fig. 2 Schematic diagram of the left ventricular model. It comprises CAir and CSack.
a compression chamber containing air and tap water  (diameter Equation 5 expresses the flow dynamics during systolic
of cross section u140), a left ventricular sack `, a pusher plate ´, phase, with PVentricular_wall representing the pressure
a linear guide servo motor ˆ, and mechanical valves ˜
applied to the sack by the actuator. RdirAo and RCV are the
A servomotor coupled with a linear actuator (RCS2; IAI, resistances to flow through the aortic valve and inside the
Shizuoka, Japan) (ˆ in Fig. 2) was used to compress the air mock ventricle, respectively, LV is the inertia effect, and
and water mixture via a polyethylene pusher-plate (´ in fAo_Outlet is the flow through the aortic valve.
Fig. 2). The actuator contacts the pusher-plate during sys- dfAo Outlet
tole to compress the ventricle sack by compression of the PVentricular wall  LVP ¼ Lv þ RdirAo fAo Outlet
dt
air inside the chamber, and during diastole the actuator was þ RCV fAo Outlet ; ð5Þ
rapidly retracted to decouple from the pusher-plate,
enabling the ventricle to relax and to fill passively with the where
preload pressure; the stroke displacement of the linear PVentricular wall
FT
¼ PServo Motor ¼ SChamber ¼ ðmWater þmTest Fluid ÞaServo Motor
SChamber
servo actuator was set at 24 mm; the displacement and air-
In Eq. 5, PServo Motor is the pressure generated by the
to-water ratio were experimentally determined on cardiac
servo motor, FT the force transmitted by the servo motor to
output as 5.0 l/min.
the chamber with a cross sectional area of SChamber, mWater
Generally, the ventricular elastance E(t)LV is known to
and mTest Fluid the mass of the fluid inside the chamber and
vary during the cardiac cycle and the maximum of E(t)LV
test fluid inside the duplicator, respectively, and aServo Motor
named as Emax is independent of the preload and afterload as
the acceleration of the linear actuator. By substituting LVP
reported by Suga et al. [9]. E(t)LV is expressed in terms of the
from Eq. 4 into Eq. 5, we obtain the relationship between
left ventricular pressure, LVP, and volume, LVV, by the
the linear actuator compression force, FT, mock ventricle
equation:
volume changes, the inertia effect of the ventricle, air
EðtÞLV ¼ LVP=ðLVV  V0 Þ; ð1Þ compliance, sack compliance, aortic flow resistance, and
flow into the aortic system:
where V0 is the intersection of the ESPVR (end systolic
 
pressure–volume relationship) with the volume axis. LVVSystole  LVVUnload
Likewise, the time-varying elastance of the air inside the PServo Motor 
ðCAir þ CSack Þ
LV chamber ( in Fig. 2), E(t)Air, can be expressed by the dfAo Outlet
equations: ¼ LV þ RdirAo fAo Oulet þ RCV fAo Outlet ð6Þ
dt
PAir Unload The ventricular volume change during the systolic phase,
VAir ¼ VAir Unload ð2Þ
PAir LVVSystole, is expressed by Eq. 7, where LVVEnd Diastole is
EðtÞAir ¼ PAir =VAir ; ð3Þ the end-diastolic ventricular volume.
EndZSystole
where VAir and PAir are the time-dependent volume
and pressure of the air inside the LV chamber. To calculate LVVSystole ¼ LVVEnd Diastole  fAo Outlet dt ð7Þ
VAir by use of Eq. 2, PAir Unload was measured when each End Diastole

123
192
The relationship among the mock ventricular pressure,
its volume, and the preload changes during the diastolic
phase is expressed by Eq. 8, where RdirLA represents
the forward resistance of the mitral valve against the flow,
PAir Diastole the air pressure inside the chamber, and fLA_Inlet
the flow through the inlet mitral valve.
dfLA Inlet
PAir Diastole  LVP ¼ Lv þ RdirLA fLA Inlet
dt
þ RCV fLA Inlet
The ventricular volume change during diastole,
LVVDiastole, is expressed by Eq. 9
EndZDiastole
LVVDiastole ¼ LVVEnd Ssystle þ fLA Inlet dt;
End Systole
where LVVEnd Systole is the end-systolic ventricular volume.
An important factor for simulation of the total circu-
lation is appropriate coupling between the systemic and
pulmonary circuits in controlling the left and right heart
output and preload changes that occur secondary to
changes in the output of each heart. The arterial system
comprises an aortic system and a downstream arterial
system in which each component is represented by
inertia elements and Windkessel compliance with an
adjustable resistance. The centrifugal pump (BP-X80;
Medtronic, MN, USA) was used to represent the pump
function of the right heart. The rotational speed was
experimentally set 1,500 rpm in order that the average
water head of the venous system reservoir (´ in Fig. 1)
and pulmonary compliance (˜ in Fig. 1) did not vary
when the air-to-water volume ratio or the arterial resis-
tance were not changed. For fixing the rotational speed,
the output of the BP-X80 responds to changes in the
head pressure, i.e. the pulmonary pressure minus the
right atrial pressure. Furusato et al. [11] reported a
technique incorporating a centrifugal blood pump
between two tanks that was adapted to another mock
circulatory system in order to reproduce the inflow
restriction of heart failure. The systemic to pulmonary
venous system capacity ratio was fixed to 2:1 by the
cross sectional area of the reservoirs [12].
Validation of the mock circulation system
To validate the performance of the pulse duplicator, we
evaluated changes in the P–V diagram and in heart con-
tractility, expressed in terms of Emax, as the air-to-fluid
volume ratio in the chamber was varied. Finally, a rotary
blood pump was placed from the apex of the mock ven-
tricle to the aorta to assess the feasibility of the mock
circulatory system for evaluation of the ventricular assist
pumps.
123
J Artif Organs (2010) 13:189–196
Variation of Emax as a function of air-to-fluid volume ratio
The elastance E(t)LV defined by Eq. 1 usually takes the
maximum value Emax at the end systole and is modified as
LVPEnd Systole/(LVVEnd Systole – V0).
This study examines the hypothesis that the air-to-fluid
volume ratio in the LV chamber referring to Fig. 2 alters
the contractility of the mock ventricle expressed by Emax.
ð8Þ Experimentally, the compliance effect was simulated as the
air-to-fluid volume ratio in the chamber was varied from
300:3700 (1:12.3), to 600:3400 (1:5.67), 750:3350 (1:4.47),
900:3200 (1:3.56), 1050:2050 (1:1.95), 1200:1900 (1:1.58),
and 1350:1750 (1:1.30) ml. From the P–V diagrams, the
ð9Þ variation in the index Emax and the value of V0 were
derived while the mean AoP was varied (60, 80 and
100 mmHg). Also, the elastance changes of the air inside
the chamber, expressed as E(t)Air max, were obtained from
Eqs. 2 and 3. The Emax derived from the P–V diagrams was
correlated against the E(t)Air max values.
Effects on P–V diagrams of rotary pump assistance
Finally, the clamp on the outflow port of the Gyro Pump C1E3
was released to evaluate its effects on the ventricular
dynamics. The bypass flow rate was varied from 3.0 to
4.0 l/min to obtain effects on the P–V diagrams of the mock
ventricle; the bypass conditions that set 3.0 and 4.0 l/min
mean support rate as 60 and 80%, respectively, but the
total flow for each bypass condition were approximately
5.0 l/min.
Statistical analysis
The results were expressed as mean ± SD (standard
deviation) and statistical significance was tested by the
standard Student’s t test with p value less than 0.05.
Results
P–V diagram, Emax, and ELV(t) versus air-to-fluid
volume ratio of the LV chamber
Figure 3a–c shows the hydrodynamics of the pulse dupli-
cator showing waveforms of mock ventricular pressure
LVP, aortic pressure AoP, preload pressure, mock ven-
tricular volume LVV, and cardiac output for mean AoP of
60, 80, and 100 mmHg. The air-to-fluid volume ratio in the
LV chamber was 1:5.67. Figure 4 shows the variations in
the P–V diagrams with the changes in the air-to-fluid vol-
ume ratio. The mean ± SD of Emax decreased from
1.75 ± 0.015 to 1.42 ± 0.019, 1.06 ± 0.025, 0.89 ±
0.005, 0.77 ± 0.002, 0.56 ± 0.003, and to 0.44 ± 0.015 as
J Artif Organs (2010) 13:189–196
Fig. 3 Wave forms obtained for LVP, AoP, LVV, preload, and cardiac output at Emax = 1.42 and mean AoP of a 60, b 80 and c100 mmHg
the air-to-fluid volume ratio was varied from 1:13.3, to
1:6.66, 1:5.55, 1:4.55, 1:3.0, 1:2.56, and to 1:2.27.
The elastance E(t)LV, as derived by use of Eq. 1, for the
data of Fig. 4 is shown in Fig. 5. The statistical signifi-
cance of differences among Emax of E(t)LV was evaluated
as 0.001. The difference among them was significantly
small with p \ 0.05.
Therefore, the relationship between E(t)Air max and
Emax shows that Emax can be set with the control of E(t)Air
(Fig. 6). As the air volume decreased, E(t)Air max and Emax
both increased, from 0.072 to 0.713 and from 0.521 to
1.921, respectively, with the regression line between
them expressed by the quadratic equation y = 2.33x2 ?
3.86x ? 0.35 (r2 = 0.9846).
Effects of the rotary blood pump assistance
on P–V diagrams
Figure 7 shows the alterations in the P–V diagrams when a
rotary blood pump assisted the failing heart with the Emax
193
of 0.6 and with the bypass flow rate varying from 3.0 to
4.0 l/min.
Discussion
A new pulse duplicator with a passive fill ventricle was
designed to implement cardiac dynamics as evaluated by
use of P–V diagrams. We were successful in obtaining
physiological P–V diagrams of the heart. Successful
duplication of the physiological P–V diagrams relies on
three key elements:
1 a passive fill mechanism of the mock ventricle;
2 compliant characteristics of the mock ventricle during
systole and diastole; and
3 proper adjustment of the circuit elements in both
systemic and pulmonary systems.
In order to mimic the passive fill mechanism of the
native heart during diastole, first of all a linear drive system
123
194
Fig. 4 Variation of pressure–volume diagram and Emax with air-
to-fluid volume ratio. For these air-to-fluid volume ratios, the values
of Emax were: 1.75, `1.52, ´1.06, ˆ0.89, ˜0.77, Þ0.56, and
þ0.44. The ESPVR lines connecting the end-systolic points of the
pressure–volume diagrams showed decrease in the slope of Emax as
the air-to-fluid volume ratio was increased
Fig. 5 Variation of E(t)LV calculated from the P–V diagram in Fig. 4.
Emax are 1.75, `1.52, ´1.06, ˆ0.89, ˜0.77, and Þ0.56
Fig. 6 E(t)Air max versus Emax of the mock ventricle could be
expressed by the quadratic equation y = 2.33x2 ? 3.86x ? 0.35; the
correlation coefficient was enough strong at r2 = 0.9846
in combination with a pusher-plate actuation-decoupling
mechanism was incorporated in the cardiac chamber.
After completion of each ejection as determined from a
123
J Artif Organs (2010) 13:189–196
Fig. 7 Alterations in the P–V diagrams when the rotary blood pump
was connected; the three diagrams are heart failure mode
(Emax = 0.6 mmHg/ml) with clamped bypass, and heart failure mode
with the bypass flow of 3.0 and 4.0 l/min
pusher-plate position using a laser light displacement sen-
sor (LB-60; Keyence, Osaka, Japan), a linear actuator was
rapidly retracted decoupling from the pusher-plate.
A large-orifice 28 mm mono-leaflet valve placed in the
inlet of the left ventricular model minimized resistance to
filling. A similar mechanism was reported by Takatani
et al. in 1981 [10] where a Hall-effect pusher-plate position
signal was used for feedback control of both left and right
hearts in the full-fill and full-eject modes.
Second, the compliant characteristics of the ventricle
during both diastole and systole were implemented by
placing a 100 ml sack made of soft latex rubber in the
chamber that contained a variable volume percentage of
sealed air and water. The air inside the chamber was
compressed during ejection before transmitting the applied
force by the linear actuator to the mock ventricle, while
pressure and volume changes of the air during diastole
assisted mimicking of the filling characteristics of the
native heart. The timing of the ventricular ejection, and
hence the time of the volume change of the ventricle sack,
was detected from the flow waveform monitored in the
aorta. The P–V diagrams for the air-to-fluid volume ratio
1:5.67 yielded the Emax of 1.56 with the V0 of 4.5 ml,
irrespective of the aortic pressure change, confirming
similar characteristics as those of the native hearts.
The third key factor in duplicating physiological
P–V relationships was utilization of properly designed
circuit elements. A 28 mm inflow valve and a 25 mm
outflow valve were used to couple the mock ventricle with
the atrial reservoir and aortic system with low resistance to
fill and eject. The compliance in each element and fluid
inertia force were properly adjusted to ensure physiological
pressure and flow wave forms. The ratio of the capacities of
the systemic venous and pulmonary circuits was adjusted
to 2:1 so as to appropriately reflect the effects of volume
shift between the systemic and pulmonary circuits. In
particular, the effects of the preload changes, which reflect
J Artif Organs (2010) 13:189–196
changes in the filling characteristics and pump output of the
mock ventricle were expressed using a constant-flow right
heart represented by a centrifugal blood pump. These fill-
ing and ejection characteristics controlled through feed-
back mechanism of the pulse duplicator were instrumental
in duplicating physiological pressure and flow waveforms
from which physiological P–V diagrams were derived.
Variations in the air-to-fluid volume ratio inside the
ventricle chamber from 1:12.3, to 1:5.67, 1:4.47, 1:3.56,
1:1.95 and to 1:1.58 resulted in appropriate Emax changes
from 1.75 to 0.56. The increase in the air volume inside the
chamber reduced the compliant characteristics of the
ventricle exhibiting a lower contractile strength. The larger
the air volume, the lower was the Emax or the less the
contractile strength of the ventricle. The maximum elas-
tic property of the air inside the chamber, expressed as
E(t)Air max, correlated with the Emax derived from the
P–V diagrams with correlation coefficient of r2 = 0.9846.
Although the value of V0 remained fairly constant until the
Emax decreased to 0.5, further reduction in the Emax sec-
ondary to increase in the V0 value, deviated from the native
heart characteristics. Because the Emax value of the normal
human heart ranges from 1.3 to 1.6, with a value below 1.0
indicating heart failure [9, 12, 13], the newly developed
pulse duplicator may be applicable to simulation of the
dynamics of the normal heart as well as the sick heart.
Because the characteristics of P–V diagram of this dupli-
cator are similar to those of other systems [4–6, 8] and
ESPVRs in Fig. 4 showed Emax varied when V0 does not
change [4–6, 8, 14, 15], this pulse duplicator can simulate
the mechanical character of the ventricle similar to the
Suga-Sagawa elastance model [9].
The P–V diagram changes seen with rotary blood pump
assistance revealed similar changes in the P–V diagrams
reported by other workers [16]. When the clamp on the
cannula connected from the apex of the ventricle to the
assist device was released, the end diastolic volume
decreased, peak LVP increased, and the ejection time
period revealed by the P–V diagrams (Fig. 7) shortened
and the area inside the P–V diagram was reduced in
comparison with the non-assisted P–V diagram. This is
because the rotary blood pump continually drained fluid
from the ventricular model during the period of isovolumic
contraction and relaxation; 146 ml end diastolic LVV was
shifted to 131 (3.0 l/min) and 114 (4.0 l/min) ml, and
69 mmHg peak LVP to 104 (3.0 l/min) and 80 (4.0 l/min)
mmHg. Thus, the pulse duplicator can represent not only
recovery of LVV and LVP, but the ventricular pumping
mechanism and the change of cardiac external work
dependent on drainage by ventricular assist devices. Also,
although negative pressure was found in Fig. 4, it is not
considered a large problem to evaluate the ventricular
assist device. The effects of the ventricular assist devices,
195
whether pulsatile or continuous flow, can be easily simu-
lated with this system from which design criteria for
efficient circulatory assist devices may be derived before
in-vivo verification in animals. More interesting studies
from the point of understanding the interaction between the
cardiac system and the mechanical circulatory support
devices can be promoted by utilizing the newly designed
pulse duplicator system.
Conclusion
A newly designed pulse duplicator system yielded physi-
ological P–V diagrams with Emax values ranging from 0.5
through 1.75, simulate those of the normal heart and the
sick heart. The newly designed pulse duplicator is appli-
cable to studying the interaction between the native heart
and continuous or pulsatile ventricular assist devices before
animal studies and clinical trial.
Acknowledgments The authors would like to thank the Tokyo
Institute of Technology, Center for Precise Fabrication Technology,
for constructing the hydraulic system essential for the conduct of this
investigation. This study was partially supported by a grant-in-aid
from Japan Society for Promotion of Science Basic Science Research
B #18300149 (PI: Professor Setsuo Takatani).
References
1. Reul H, Minamitani H, Runge J. A hydraulic analog of the sys-
temic and pulmonary circulation for testing artificial hearts.
Proceeding ESAO2 (1975);120–27.
2. Medeart D, Schmitz C, Rau G, Reul H. Design and in vitro
performance of a novel bileaflet mechanical heart valve pros-
thesis. Int J Artif Organs. 2005;28:256–63.
3. Maughan WL, Burkhoff D, Sunagawa K. Left ventricular inter-
action with arterial load studied in isolated canine heart. Am J
Physiol. 1983;245:H773–80.
4. Ferrari G, De Lazzari C, Mimmo R, Tosti G, Ambrosi D, Gor-
czynska K. A computer controlled mock circulatory system for
mono- and biventricular assist device testing. Int J Artif Organs.
1998;21:26–36.
5. Pantalos GM, Koenig SC, Gillars KJ, Giridharan GA, Ewert DL.
Characterization of an adult mock circulation for testing for
cardiac support devices. ASAIO J. 2004;50(1):37–46.
6. Koenig SC, Pantalos GM, Gillars KJ, Ewert DL, Litwak KN,
Etoch SW. Hemodynamic and pressure–volume responses to
continuous and pulsatile ventricular assist in adult mock circu-
lation. ASAIO J. 2004;50:15–24.
7. Schroeder MJ, Perrault B, Ewert DL, Koenig SC. HEART: an
automated beat-to-beat cardiovascular analysis package using
Matlab. Comput Biol Med. 2004;34(5):371–88.
8. Colacino FM, Moscato F, Piedimonte F, Danieli G, Nicosia S,
Arabia M. A modified elastance model to control mock ventricles
in real time: numerical and experimental validation. ASAIO J.
2008;54:563–73.
9. Suga H, Sagawa K. Instantaneous pressure–volume relationships
and their ratio in the excised supported canine left ventricle. Circ
Res. 1974;35:117–26.
123
196
10. Takatani S, Harasaki H, Suwa S, Murabaryashi S, Sukalac R,
Jacobs G, Kiraly R, Nose Y. Pusher-Plate Type TAH system
operated in the left and right free-running variable rate mode.
Artif Organs. 1981;5(2):132–42.
11. Furusato M, Shima T, Kokuzawa Y, Ito K, Tanaka T, Iwasaki K,
Qian Y, Umezu M, Yan Z, Zhu L. A reproduction of inflow
restriction in the mock circulatory system to evaluate a hydro-
dynamic performance of a ventricular assist device in practical
conditions. Life Syst Model Simul. 2007;4689:553–8.
12. Guyton AC, Hall JE. Textbook of medical physiology, 11th edn.
Pennsylvania: Elsevier Saunders; 2006. p. 171–2.
123
J Artif Organs (2010) 13:189–196
13. Guyton AC, Hall JE. Textbook of medical physiology, 11th edn.
Pennsylvania: Elsevier Saunders; 2006. p. 111–3.
14. Maughan WL, Sunagawa K, Burkhoff D, Sagawa K. Effect of
arterial impedance changes on the end-systolic pressure–volume
relation. Circ Res. 1984;54:595–602.
15. Baloa LA, Boston JR, Antaki F. Elastance-based control of a
mock circulatory system. Ann Biomed Eng. 2001;29:244–51.
16. Moscato F, Vollkron M, Bergmeister H, Wieselthaler G, Leonard
E, Schima H. Left ventricular pressure–volume loop analysis
during continuous cardiac assist in acute animal trials. Artif
Organs. 2007;31:369–76.