Development of Totally Implantable Pulsatile Biventricular Assist Device

Article (PDF Available)inArtificial Organs 27(1):119-23 · February 2003with32 Reads
DOI: 10.1046/j.1525-1594.2003.07177.x · Source: PubMed
Approximately 10% to 15% of all patients implanted with left ventricular assist devices (LVADs) have required right heart support with another device. The necessity of aggressive biventricular support has already been proposed. Therefore, the totally implantable biventricular assist device (BVAD) was developed. The width of the BVAD main body was 87 mm, the thickness 67 mm, and the height 106 mm, while the weight was 785 g. The automatic control algorithm was developed to prevent lung edema and atrial rupture.
Recently, the implantable LVAD has been con-
sidered a useful therapeutic modality for bridge-
to-transplantation or bridge-to-recovery patients
with congestive heart failure (1–3). However,
approximately 10% to 15% of all patients implanted
with wearable LVADs have required right heart
support with another device (1). It has also been real-
ized that more diseased natural hearts would recover
if the diseased heart were given rest for at least one
year through the utilization of blood pumps (4).
Therefore, several researchers have already insisted
on the necessity of an implantable biventricular assist
device (5–8). But there are only extracorporeal and
paracorporeal pulsatile BVAD systems among
the clinically available assist devices, while a few
implantable centrifugal pump systems and pulsatile
pumps are under development for BVAD appli-
cation (5–7).
The one-piece BVAD was developed using the
modified moving actuator mechanism. In this article,
the authors report only the engineering aspects of
totally implantable pulsatile BVAD.
BVAD system design
The BVAD is composed of one moving actua-
tor and two smooth polyurethane blood sacs (Pel-
lethane2363, Dow Chemical, Midland, MI, U.S.A.)
with the actuator placed between the blood sacs. The
volume of both blood sacs is about 100 cc, while the
whole volume of the BVAD without all connectors
is about 500 cc. The moving actuator mechanism that
was devised for the small size pulsatile total artificial
heart (TAH) was utilized as a pumping principle to
develop a one-piece implantable device (9–10). The
moving actuator rotates bidirectionally around the
base shaft from left to right along the guide gear to
alternately press the blood sacs and discharge blood
from them (Fig. 1). The shaft and the gear are fixed
in the metal housing that is packed with the actuator
and the blood sacs in the rigid polyurethane chamber
(Isoplast301, Dow Chemical). Fluor oil (Perfluo-
ropolyether oil, 50 cc, Ausimont, Milan, Italy) is used
Development of Totally Implantable Pulsatile
Biventricular Assist Device
*Chan Young Park, ‡Jun Woo Park, ‡Jung Joo Lee, ‡Wook Eun Kim, *Chang Mo Hwang,
*Kyong-Sik Om, §Jaesoon Choi, §Jongwon Kim, §§Eun Bo Shim, **Young Ho Jo, and
*†Byoung Goo Min
**Institute of Medical and Biological Engineering, Medical Research Center; †Department of Biomedical Engineering,
College of Medicine; and ‡Interdisciplinary Program in Medical and Biological Engineering Major, Graduate School,
Seoul National University, Seoul; §Biomedlab Company, Seoul; §§Department of Mechanical Engineering, Kumoh
National University of Technology, Kumi; and **Biomedical Engineering Branch, Division of Basic Sciences,
National Cancer Center, Koyang, Korea
Artificial Organs
27(1):119–123, Blackwell Publishing, Inc.
© 2003 International Society for Artificial Organs
Received September 2002.
Presented in part at the 3
Japan–Australia Cardiovascular
Bioengineering Symposium, held November 9–10, 2001, in Sendai,
Address correspondence and reprint requests to Dr. Byoung
Goo Min, Department of Biomedical Engineering, College of
Medicine, Seoul National University, 28 Yungun-dong, Chongno-
gu, Seoul 110-744, Korea. E-mail:
Abstract: Approximately 10% to 15% of all patients
implanted with left ventricular assist devices (LVADs)
have required right heart support with another device. The
necessity of aggressive biventricular support has already
been proposed. Therefore, the totally implantable biven-
tricular assist device (BVAD) was developed. The width of
the BVAD main body was 87 mm, the thickness 67 mm, and
the height 106 mm, while the weight was 785 g. The auto-
matic control algorithm was developed to prevent lung
edema and atrial rupture. Key Words: Biventricular
assist device—Moving-actuator—Pulsatile—Implantable
—Circulatory assist device.
for the purpose of lubricating gear trains and the con-
tacting surface between the actuator and blood sacs,
and for heat dissipation of energy converter. The dif-
ference between left and right stroke volumes is com-
pensated by the flip-flop motion of the flexible
polyurethane membrane of the compliance window
and by the volume change of the compressible air in
the variable interventricular space.
It was defined that the distance from the left side
to the right side of the body of the device was the
width, while the anterior to posterior distance and
the superior to inferior distance were thickness and
height, respectively. The A-P thickness was defined as
the thickness between the anterior surface and the
posterior surface of the main pump body. The A-P
thickness was thought to be important in the anatom-
ical point of view because the shorter A-P thickness
makes it easier to implant a device into the abdomen
such as the preperitoneal space. The locations of all
ports were moved to the sidewall of the main body,
and the two of the ports were made L-shaped to face
in the same direction, toward the patient’s own heart.
The cannulae for left heart assistance were designed
to connect from the left atrium to the descending
aorta through the BVAD, while the cannulae for
right heart assist were devised to bypass flow from
the right atrium to the main pulmonary artery. All
cannulae were custom-made in house and had 3/8”
internal diameter. The polyurethane based polymer
valves developed by the authors’ group were
adopted, but mechanical valves could also be used
with a special fitting. All the surfaces contacting
blood or tissues except the cannulae were made
of polyurethane based polymer or coated with
polyurethane solution. The energy and data trans-
mitting systems were also developed. The trans-
cutaneous energy transmission system (TETS) and
infrared telemetry system were constructed to elim-
inate the transcutaneous drive line and power line.
The energy provided by the external battery pack
was transmitted through TETS into the motor, and
the data from the motor was sent through telemetry
to the external monitoring system (Fig. 2).
Control system
The automatic control system using motor current
wave forms was adopted, which can change the
pumping condition automatically to accommodate
the status of the filling pressure. As the moving actu-
ator starts to rotate leftward, after a short period of
low motor current, motor current rises rapidly at the
time when the outlet valve is opened and the blood
in the blood sac starts to move out. During this short
period, the moving actuator does not contact with the
blood sac but just moves toward it. If the filling pres-
sure is enough to fill the blood sac fully, the time
before the actuator contacts the blood sac will be
very short. But if the venous return is less and the
blood sac is partially filled, the time will become
longer. Therefore, this time before contact can be
120 B.G. MIN ET AL.
Artif Organs, Vol. 27, No. 1, 2003
Blood Filling
Blood Ejecting
Left Blood Sac
(Systolic Phase)
Right Blood Sac
(Diastotic Phase)
Energy Converter
FIG. 1. This is a schematic diagram of the moving actuating
FIG. 2. A totally implantable pulsatile BVAD system is shown.
used as an indicator of the filling status of the blood
sac or that of the atrial pressure. Thus, a normalized
parameter, percent-time before contact (PTBC), was
introduced (11). PTBC is defined as follows, and can
be measured directly from the motor current wave-
form (Fig. 3).
The higher the left atrial pressure is, the faster the
left blood sac fills, and the smaller the PTBC value is,
and vice versa. Through the extensive bench tests, it
was found that the PTBC value was correlated with
the inlet pressure or preload. Therefore, it can be
used as the criterion to adjust operating condition to
prevent the left atrial pressure rise that will cause
lung edema. If the PTBC value is low, the preload is
estimated to be high and the pump rate of BVAD is
increased to pump more blood out. If the PTBC
value is high, the pump rate is decreased to prevent
arterial collapse. In automatic control mode, the
normal operating condition of the pump is set to be
3.5 L/min. If the PTBC value rises above the upper
criterion, the pump rate is reduced to eject only
2.5 L/min a the set time and returned to normal oper-
ating condition automatically. If the PTBC value falls
below the lower criterion, the pump rate is increased
to pump out as much as 4.5 L/min for the set time and
returned to normal condition. The lower and upper
criteria and set time were determined empirically.
Time Before Contact
Stroke Time
100 %
In vitro tests
In vitro tests were done using a Donovan type
mock circulation system made in our laboratory (12).
The system efficiency defined by the following
formula, was also calculated using the mock circula-
tion system. Only the main pump and motor con-
troller without the wireless transmission system were
included in this test. To simplify calculations, the
mean values of all the variables were used.
The width of the BVAD main body was 87 mm,
the thickness 67 mm, the height 106 mm, and the
weight was 785 g (Fig. 4). The A-P thickness of the
main body was 67 mm. This value may be too large
for smaller patients like women, but could be accep-
table for adult patients. (Cadaver fitting studies
were performed to verify the fitting capability of the
BVAD into adult patients, but that work will not
be included in this article and will be reported in
another article.) The BVAD could pump a maximum
Hydraulic power output
Electrical power input
Aortic Pressure Left Atrium Pressure
Pulmonary Pressure Right Atrium Pressure
Pump Flow Rate
Input Current Input Voltage
Artif Organs, Vol. 27, No. 1, 2003
FIG. 3. A percent-time before
contact (PTBC) conceptual
diagram is shown. Current
waveform according to inflow
condition is depicted in good
filling condition (a) and bad filling
condition (b). L1 = Time Before
Contact, and L = Stroke Time.
of 7.5 L/min at a maximum of 170 beats per minute
(BPM), and normally pumps 4 L/min at 90 BPM. The
efficiency was 8% at the 4 L/min flow condition and
5.5% at the 6 L/min output condition. The maximum
output of the BVAD was not as much as the
maximum cardiac output of other total artificial
heart (TAH) systems, but the BVAD can pump up to
7.5 L/min and can thus provide 30–100% of the
patient’s cardiac output, with an exception of certain
cases, like exercise. The ability to pump up to 100%
of the cardiac output sets it apart from other ven-
tricular assist devices and allows it to be more than
just a bridge to heart transplantation. A new model
that optimizes the blood sac geometry is currently
being developed and will be able to totally replace
the pumping function of both ventricles.
122 B.G. MIN ET AL.
Artif Organs, Vol. 27, No. 1, 2003
FIG. 4. The photograph shows the pulsatile BVAD pump body;
the transparent membrane on the top surface of the body is the
compliance window.
FIG. 5. The graphs show an in
vitro test of the automatic
control algorithm: mean Left
Atrial Pressure (LAP) measured
from mock circulation system
and left PTBC (a), and Pump
rate (PR) (b).
Because of a dynamically variable interventricular
space between the ventricular sacs and because of
the residual function of the patient’s native heart, any
difference between the ejection volumes of the two
artificial ventricles can be accommodated, thus pre-
venting pulmonary edema (13,14).
Figure 5 shows the result of an in vitro experiment
when the suggested automatic control algorithm was
applied. It can be seen that as the left atrial pressure
drops, the PTBC value is increased, and the pump
rate is then reduced from 97 beats per minute (BPM)
to 75 BPM, lasts for 84 seconds, and returns back
to 97 BPM. It was confirmed that the controller
changed the pump rate automatically in response to
the changes of preload levels.
The newly developed BVAD does not require the
resection of the patient’s own heart. So it can be
called a “heart-saving” artificial heart. The first
advantage of the “heart-saving” artificial heart deals
with the ability to balance better the cardiac outputs
of the two ventricles as mentioned above. Secondly,
in case of malfunction of the mechanical heart, the
native heart may provide the critical survival period
to replace the damaged mechanical heart. Thirdly,
the native heart may recover to some extent in
response to new treatment modalities that are being
developed or that will be developed in the future.
Such modalities may include gene therapy and cell
therapy (14).
The totally implantable pulsatile BVAD was
developed. The A-P thickness of the pump was 67
mm and the pump was designed to be implanted into
adult patients. The automatic control algorithm was
developed to prevent lung edema and atrial rupture.
To make this pump a real alternative to transplanta-
tion, the future improvements in efficiency, the incre-
ment of pump output, and the reduction of size
should be achieved together with biocompatibility.
Acknowledgments: This study was Supported by
a grant (#HMP-98-G-2-040) of the Good Health
R&D Project; Ministry of Health & Welfare, R.O.K.
1. Stevenson LW, Kormos RL, Bourge RC, Gelijns A, Griffith BP,
Hershberger RE, Hunt S, Kirklin J, Miller LW, Pae Jr WE,
Pantalos G, Pennington DG, Rose EA, Watson JT, Willerson
JT, Young JB, Barr ML, Costanzo MR, Desvigne-Nickens P,
Feldman AM, Frazier OH, Friedman L, Hill JD, Konstam MA,
McCarthy PM, Michler RE, Oz MC, Rosengard BR, Sapirstein
W, Shanker R, Smith CR, Starling RC, Taylor DO, Wichman
A. Mechanical cardiac support 2000: Current applications and
future trial design. J Am College Cardiol 2001;37:340–70.
2. Willman VL. Expert panel review of the NHLBI total
artificial heart program. June 1998–November 1999. Availabte
cessed November 21, 2002.
3. LVAS Study Group. Evaluation of an implantable ventricular
assist system for humans with chronic refractory heart failure.
ASAIO J 1995;41:11–15.
4. Hetzer R, Muller J, Weng Y, Wallukat G, Spiegelsberger S,
Loebe M. Cardiac recovery in dilated cardiomyopathy by
unloading with a left ventricular assist device. Ann Thorac
Surg 1999;68:742–9.
5. Nosé Y, Ohtsubo S, Tayama E. Therapeutic and physiological
artificial heart: Future prospects. Artif Organs 1997;21:592–6.
6. Nosé Y, Nakata K, Yoshikawa M, Letsou GV, Fujisawa A,
Wolner E, Schima H. Development of a totally implantable
biventricular bypass centrifugal blood pump system. Ann
Thorac Surg 1999;68:775–9.
7. Andrade A, Nicolosi D, Lucchi J, Biscegli J,Arruda AC, Ohashi
Y, Mueller J, Tayama E, Glueck J, Nosé Y. Auxiliary total
artificial heart: A compact electromechanical artificial heart
working simultaneously with the natural heart. Artif Organs
8. Nosé Y,Yoshikawa M, Murabayashi S,Takano T. Development
of rotary pump technology: Past, present and future. Artif
Organs 2000;24:412–20.
9. Min BG, Kim HC, Lee SH, Chang JK, Choi JW, Kim JW, Seo
KP, Rho JR, Ahn H, Kim SW. A tether-free, moving actuator
total artificial heart. ASAIO Trans 1990;36:M249–51.
10. Min BG, Kim HC, Lee SH, Kim IY, Kim JW, Choi JW, Kim
JT, Jung DY. Development of a new moving-actuator type
electromechanical total artificial heart. Artif Organs 1991;15:
11. Om KS. A study on the pump output control of the moving-
actuator type implantable biventricular assist device [disser-
tation]. Dept. of Biomedical Engineering, Seoul National
University, Seoul, Korea, 2000.
12. Donovan Jr FM. Design of a hydraulic analog of the circula-
tory system for evaluating artificial hearts. Artif Organs
13. Jo YH, Choi JS, Kim WE, Park JW, Choi WW, Kim HC, Kim
WG, Ahn H, Rho JR, Min BG. Analysis of the interventricu-
lar pressure waveform in the moving actuator total artificial
heart. ASAIO J 2000;46:749–55.
14. Park KW. “The Dawn of Artificial Hearts?, SCA Newsletter,
October 2001. Available at:
letter/2001 oct/park.shtml. Accessed November 21, 2002.
Artif Organs, Vol. 27, No. 1, 2003
  • [Show abstract] [Hide abstract] ABSTRACT: Heart disease is the developed world's biggest killer, and the shortage of donor hearts has accelerated the development of mechanical alternatives. Scientists, engineers and clinicians have attempted to replicate the human heart with a mechanical device for over 50 years. Although a number of pulsating devices have been developed, and in some cases worked briefly, they have invariably failed to match the success of heart transplantation. In an attempt to produce a suitable alternative, current research is focused on devices that do not replace the heart; but rather work along side it to assist its function. Many of these devices help the failing left ventricle; however some patients require the additional implantation of a second device to assist a failing right ventricle. This increases implantation time and associated risk, and because of the size of the current devices, reduces the access of smaller patients to this vital technology. The overall thesis objective focuses on the progressive design, development and preliminary evaluation of two novel centrifugal type ventricular assist devices, a bi-left ventricular device (Bi-LVAD) and a single bi-ventricular assist device (Bi-VAD). The devices have the respective capability to assist either the left ventricle, or both ventricles of a failing heart. The current concept for each VAD employs both magnetic and hydrodynamic suspension techniques to float a rotating double impeller, a technique that aims to reduce blood damage and component wear, two of the major problems encountered with current generation devices. Each VAD design was developed by conducting experimentation and drawing conclusions from a variety of engineering research fields, such as flow visualization, rotary pump design and testing, fluid dynamics, hemodynamics and heart failure, and magnetic motor bearing design. In order to evaluate pump prototype designs, it was necessary to design and develop a novel pulsatile systemic and pulmonary mock circulation loop capable of reproducing the hemodynamics of heart failure in the systemic and pulmonary circuits. The investigation then specifically examined the static hydraulic forces on the impeller of a centrifugal blood pump during operation in this mock circulation loop. The recorded magnitude and direction of radial and axial thrust then influenced the selection of magnetic and hydrodynamic bearing configurations to minimise impeller touchdown in the intended hemodynamic environment. This research required the development of correctly designed impeller (semi-open/closed) and volute (single, double, circular) components for each ventricular assist application and a unique test facility to isolate impeller hydraulic forces in addition to the mock circulation loop. The proposed Bi-LVAD incorporates symmetrical blade designs on each side of the double sided impeller. The device assists the function of the left ventricle only with symmetrical axial pressure distribution and elimination of stagnant regions beneath the impeller. These features improve axial touchdown capacity and reduce thrombus formation respectively. The proposed Bi-VAD incorporates different blade designs on each side of the double impeller to augment the function of both the left and right cardiac chambers. The design has the additional potential to act as a total artificial heart (TAH). To date there is no Bi-VAD/TAH system available that incorporates an LVAD and RVAD in one rotary pump. Successful development of each innovative VAD will provide an alternative to heart transplantation, potentially saving lives of many terminal heart patients each year. No longer would heart transplant candidates need to wait for the untimely death of a donor to provide a suitable heart. Instead, this new generation device would be available immediately, and be almost universally compatible with all patients. It has the potential to dramatically increase a patient’s expected lifetime, and to deliver them a higher quality of life.
    Article · Jan 2005 · Journal of Artificial Organs
  • [Show abstract] [Hide abstract] ABSTRACT: AnyHeart is a single-piece, implantable biventricular assist device. This electromechanical BVAD has a moving-actuator mechanism. To monitor the status of AnyHeart from anywhere at any time, a portable personal digital assistant (PDA) monitor and web-based remote monitoring system were developed. The PDA local monitoring system has replaced bulky personal computer monitoring systems. The web-based remote monitoring system has several functions such as data collecting, storing, and posting through the internet. Basically, interventricular pressure (IVP) is a parameter indicating the filling level of the blood chambers of AnyHeart. The pump output can be estimated using IVP, which is acquired noninvasively from AnyHeart. With the proposed method, we can estimate the pump output with a small margin of error.
    Article · Jun 2003
  • [Show abstract] [Hide abstract] ABSTRACT: Blood flow in the twin-pulse life-support system (T-PLS) pulsatile blood pump was simulated using a three-dimensional rigid body-fluid-solid interaction model. This model can delineate the blood flow in the T-PLS resulting from operation of a moving actuator. The numerical method used in this study was a commercial finite element package called ADINA. We used a contact and fluid-solid interaction model to compute the blood hemodynamics in the sac. Blood flow is generated by the motion of the actuator, which strongly interacts with the solid material surrounding the blood. To obtain basic bioengineering data on the optimum operation of the T-PLS, we simulated four models in which the actuator moved at different speeds and investigated both the flow pattern and the distribution of shear stress. During the contraction phase, a strong axial flow is observed around the outlet, whereas there is stagnant flow around the inlet. The maximum shear stress in each model depends on the operation mode; however, all four models have similar flow rates. The sinusoidal mode exhibited the lowest maximum shear stress and is thus considered the most efficient of the four operating modes.
    Article · Feb 2004
Show more