96580

PRF29110.1177/0267659113496580PerfusionPark et al.

Original Paper

Total artificial heart in the pediatric patient

with biventricular heart failure

Perfusion
2014, Vol 29(1) 8288
The Author(s) 2013
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0267659113496580
prf.sagepub.com

SS Park,1 DB Sanders,1 BP Smith,1 J Ryan,2 J Plasencia,2

MB Osborn,1 CM Wellnitz,1 RN Southard,1 CN Pierce,3
FA Arabia,3 J Lane,1 D Frakes,2 DA Velez,1 SG Pophal1
and JJ Nigro1
Abstract
Mechanical circulatory support emerged for the pediatric population in the late 1980s as a bridge to cardiac transplantation.
The Total Artificial Heart (TAH-t) (SynCardia Systems Inc., Tuscon, AZ) has been approved for compassionate use by the
Food and Drug Administration for patients with end-stage biventricular heart failure as a bridge to heart transplantation
since 1985 and has had FDA approval since 2004. However, of the 1,061 patients placed on the TAH-t, only 21 (2%)
were under the age 18. SynCardia Systems, Inc. recommends a minimum patient body surface area (BSA) of 1.7 m2, thus,
limiting pediatric application of this device.
This unique case report shares this pediatric institutions first experience with the TAH-t. A 14-year-old male was admitted with dilated cardiomyopathy and severe biventricular heart failure. The patient rapidly decompensated, requiring
extracorporeal life support. An echocardiogram revealed severe biventricular dysfunction and diffuse clot formation in
the left ventricle and outflow tract. The decision was made to transition to biventricular assist device. The biventricular
failure and clot formation helped guide the team to the TAH-t, in spite of a BSA (1.5 m2) below the recommendation of
1.7m2. A computed tomography (CT) scan of the thorax, in conjunction with a novel three-dimensional (3D) modeling
system and team, assisted in determining appropriate fit. Chest CT and 3D modeling following implantation were utilized
to determine all major vascular structures were unobstructed and the bronchi were open. The virtual 3D model confirmed appropriate device fit with no evidence of compression to the left pulmonary veins.The postoperative course was
complicated by a left lung opacification. The left lung anomalies proved to be atelectasis and improved with aggressive
recruitment maneuvers.The patient was supported for 11 days prior to transplantation. Chest CT and 3D modeling were
crucial in assessing whether the device would fit, as well as postoperative complications in this smaller pediatric patient.
Keywords
Total Artificial Heart; ECLS; ECMO; decompensated heart failure; cardiac transplantation; mechanical circulatory
support; ventricular assist device; VAD; 3D modeling

Introduction
Approximately 1 in 100,000 children are diagnosed each
year with symptoms of cardiomyopathy in the USA.1
Dilated cardiomyopathy (DCM) is the primary indication, accounting for 65% of heart transplants in children
11-17 years of age.2 Children diagnosed with DCM and
decompensated heart failure represent a high-risk subgroup, often benefiting from optimal heart failure management and prompt evaluation for a heart transplant.
According to Organ Procurement and Transplantation
Network, in 2011, 375 heart transplants were performed
in pediatric patients.3 Unfortunately, children awaiting a
heart transplant are subject to the highest waiting list
mortality in solid-organ transplant medicine in the
USA.4 Almond et al. state that 17% of children under the
age of 18 years in the USA died while awaiting heart
transplantation in the years 1999-2006.4 It is imperative

1Division

of Cardiothoracic Surgery, Division of Cardiology, Division of

Critical Care Medicine, Childrens Heart Center, Division of Radiology,
Phoenix Childrens Hospital, Phoenix, AZ, USA.
2School of Biological and Health Systems Engineering; School of
Electrical, Computer and Energy Engineering, Arizona State University,
Tempe, AZ, USA.
3Cardiothoracic Surgery Division, Mayo Clinic Scottsdale, Scottsdale, AZ, USA.
Corresponding author:
D Bradford Sanders
Chief Perfusionist
Division of Cardiothoracic Surgery
Childrens Heart Center
Phoenix Childrens Hospital
1919 East Thomas Road
Phoenix, Arizona, 85016, USA.
Email: dsanders3@phoenixchildrens.com
Presented at the 34th annual seminar of the American
Academy of Cardiovascular Perfusion, Los Angeles, California,
24-27 January, 2013.

Park et al.

83

Figure 1A and B. The SynCardia Total Artificial Heart.

Figure 1A depicts the SynCardia Total Artificial Heart. The TAH-t is comprised of two semi-rigid polyurethane ventricles of 70 ml volume with a
flexible, multi-layered, polyurethane diaphragm separating the blood path from the pneumatic chamber. Quick-connect inflow sewing cuffs attach the
artificial ventricles at the 27 mm mechanical valve to the remaining atria. Quick-connect outflow sewing grafts attach the artificial ventricles at the 25
mm mechanical valve to the pulmonary artery and aorta. Percutaneous conduits protrude from the pneumatic chamber, exiting through the upper
right quadrant of the abdominal wall and attaching to 7-foot-long drivelines that connect the ventricles with the external driver and console.
Figure 1B illustrates in situ implantation of the TAH-t complete with connection of the atrial cuffs and outflow graft to and from the ventricles and
their respective anatomic structures, as well as the blood flow path. Images Courtesy: syncardia.com

that children in acute decompensated heart failure be

promptly identified for emergent deployment of cardiac
assist devices to preserve end-organ function and quickly
restore functional status.
Ventricular assist devices (VAD) have been used in
the pediatric population since the late 1980s as a bridge
to transplantation.5 Expanded production of various
types of VADs has allowed the technology to proliferate
in the pediatric population, doubling over the years.3 At
present, nearly twenty percent of adolescent/pediatric
heart transplant recipients are bridged to cardiac transplantation with VADs.2 The Total Artificial Heart (Figure
1A) has been approved by the US Food and Drug
Administration (FDA) as a bridge to transplant in
patients with severe biventricular heart failure.6
The TAH-t eliminates a number of issues encountered
with other VADs, such as valve-related problems, left
ventricular clot formation, right heart failure, arrhythmias and the need for inotropic support.7,9-11 The device
replaces the native ventricles and valves. The prosthetic
ventricles connect directly to the atria and the great vessels, as shown in Figure 1B. It weighs 160 grams and
resides in the pericardial space. The TAH-t provides
biventricular pulsatile flow similar to normal body circu-

lation. With the shortest blood flow path of any assist

device, there is a significantly decreased risk of thromboembolism.1 The TAH-t currently has two 70 cc ventricles
and features adjustable ventricular orientation. Figures
2A and 2B demonstrate the cardiac cycle of the TAH-t in
its systolic and diastolic phases, in detail. The device is
typically set to partially fill at a fixed percentage of systole and a fixed beat rate of 120 to 130 beats per minute,
with a stroke volume of 50-60 ml.10
In selected adolescent and small adult patients with
intractable biventricular heart failure awaiting transplant, the Total Artificial Heart (TAH-t) has been utilized as the assist device of choice.7 The TAH-t has
demonstrated a reduced incidence of stroke and transient ischemic attacks in recent adult studies.6,7
Unfortunately, the deployment of the TAH-t in children
is limited. Of the 1,061 patients placed on the TAH-t,
only 21 (2%) were under the age of 18 years of age.5 For
proper fit and function, the recommended minimum
body surface area (BSA) is 1.7 m2 and the anteroposterior (AP) distance between the sternum and spine at the
level of T10 is 10 cm.7 These criteria represent a significant barrier for TAH-t in the pediatric population. In
addition, there are few certified pediatric centers with

84

Perfusion 29(1)

Figure 2A and B. The Cardiac Cycle of the SynCardia TAH-t.

The cardiac cycle of the SynCardia Total Artificial Heart operates by shuttling pressurized air in and out of the pneumatic chambers of the left and
right ventricles in synchronization through the attached percutaneous drivelines connected to the external driver and console. Figure 2A displays the
systolic phase at the beginning when air is applied from the external driver (1) to the pneumatic chambers of the ventricles. When the sub-diaphragm
ventricular pressures exceed the pulmonary and aortic pressures, the 25 mm outlet valves will open (2) and the diaphragms will displace upward
until they reach maximum distension and all ventricular blood is ejected (3). The build-up of pressure is measured and the active tracing is displayed
on the driver monitor. The accumulated pressurized air inside the ventricles is then released back along the drivelines to the console and exhausted
into the atmosphere at the end of the systolic cycle and the beginning of the diastolic filling cycle. Transition from the systolic to diastolic phase
is shown in Figure 2B; air is exhausted from the pneumatic chambers (1) of the ventricles back to the console to allow room for blood to fill the
ventricles.Ventricular filling begins when the sub-diaphragm ventricular pressures decrease below the right and left atrial pressures and the 27 mm
inlet valves open to allow blood to displace the diaphragms downward (2). The resulting displaced air along the drivelines from ventricular filling is
measured by the console and estimates ventricular stroke volume. The active tracing is displayed on the driver monitor. A small vacuum which draws
up to 15 mmHg during diastole can be applied from the console to help in diaphragm displacement and facilitate ventricular filling. Images Courtesy:
syncardia.com

the experience of implanting a TAH-t in children. The

needs for device modification to accommodate complex
congenital anatomies and the absence of anticoagulation
protocols for children remain challenges to deploying
the TAH-t in pediatrics in this current era.

Case Report
A 14-year-old male was admitted with DCM, severe
biventricular dysfunction, and ventricular arrhythmias. He initially presented to outside facilities with a
3-week history of cough, shortness of breath, upper
respiratory tract symptoms and fever. The patient was
an otherwise healthy adolescent with a past medical
history of asthma. The patient was transferred to this
institution for further evaluation following the identification of severely depressed cardiac failure and
life-threatening ventricular arrhythmias. An echocardiogram revealed biventricular dysfunction with an
ejection fraction of 15-17% and multiple clots were
present in the left ventricle and left ventricular outflow tract (Figure 3). Advanced heart failure medical
optimization was promptly initiated with intravenous

inotropic support. The patient acutely decompensated

and was placed on extracorporeal life support (ECLS)
via a femoral approach on hospital day one. He was
accepted for heart transplantation and placed on the
Status 1-A waiting list. After 8 days of ECLS support
without a donor available, the decision was made to
transition to a ventricular assist device, using an institutional decision algorithm developed by this service.8
Other VADs, such as the Berlin Heart and Thoratec
pneumatic VAD were considered; however, profound
biventricular failure and the identification of multifocal thrombi located in the left ventricle, outflow tract
and proximal aorta deemed these devices relative contraindications.9 Further, an endomyocardial biopsy
demonstrated no evidence of myocarditis, suggesting
no reversible process and little chance of recovery. The
combination of these findings guided the choice to
deploy the TAH-t as a bridge to transplant.
This patient was borderline with respect to the fit criteria recommended for the TAH-t (weight 50 kg, height
152 cm, BSA 1.5 m2 and 10 cm AP distance at T10).
A preoperative chest CT scan assisted in the final determination to proceed with TAH-t implantation.

85

Park et al.

diogram (TEE) showed no impingement of the left and

right atrium. The total bypass time was 185 minutes.
The sternum was closed on postoperative day 1.
During closure, the left pericardium was opened and the
TAH-t was shifted to the left pleural space to reduce pulmonary vein impingement, as evidenced on echocardiography. A study by Leprince et al. suggests this technique
to be useful in small patients with a BSA between 1.5 and
1.6 m2.7 The TEE at the time of chest closure revealed
unobstructed systemic and pulmonary venous return.

Postoperative care phase

Figure 3. Left ventricular thrombus.

Perioperative care phase

The child was transported to the operating room and
transitioned from the ECLS circuit to cardiopulmonary
bypass (CPB). The patient was cooled to a core temperature of 26 Celsius. A median sternotomy was performed
and measurements were recorded. The width of the pericardium and distance from the anterior of the spine to the
posterior table of the sternum were measured and the
dimensions confirmed adequate space for the TAH-t.
Following aortic cross-clamping, the ventricles and all 4
valves were excised, leaving a cuff of myocardium and
annuluses of the tricuspid and mitral valve at the atrioventricular (AV) junction. The ventricular mass was
removed and the TAH-t was implanted without incident.
Upon the excision of the ventricles, organized clots were
confirmed and located throughout the left ventricular
surface. Figure 4 outlines, stepwise, the surgical ventriculectomy and subsequent TAH-t implantation. Following
completion of the implantation, the TAH-t was set to a
low rate and low ejection pressures to completely de-air
the ventricles. Following a thorough inspection for intraventricular air and surgical bleeding, the device rate was
slowly increased to 80-100 beats per minute and ejection
pressures were increased until full ejection was noted in
the pressure waveform. The TAH-t quickly filled to stroke
volumes of 50 ml bilaterally, generating flows of 5-6 L per
minute. The patient was successfully weaned from CPB
and decannulated. The decision was made to leave the
sternum open secondary to an increased risk for subsequent bleeding and urgent re-exploration. The patient
had excellent cardiac output based on cerebral oximetry
(74%), mixed venous saturations (74%) via Swan-Ganz
monitoring and good hemodynamics with a blood pressure (BP) of 113/58, device rate of 100 and a central
venous pressure (CVP) of 10. A transesophageal echocar-

The patient had no major hemodynamic events. The

chest x-ray (CXR) immediately after arrival to the unit
showed clear lung fields, with good position of the
device. On postoperative day 1, he was taken back to the
operating room for re-exploration and chest closure and
was successfully extubated several hours later.
The postoperative course was complicated by an
opacification of the left lung, seen on the chest x-ray
after extubation (Figure 5). The patient was re-intubated
for progressive respiratory insufficiency on postoperative day 2. A bronchoscopy revealed some mucus plugs
found in the left bronchial tree that were suctioned and
a chest CT with 3D modeling after TAH-t implantation
ensured that all venous structures were unobstructed
and the bronchi were open. The virtual 3D model demonstrated no compression of the pulmonary veins by the
device. The left lung was found to be atelectatic and
improved with aggressive recruitment maneuvers. He
was extubated successfully on postoperative day 6.
Following extubation, the patient was able to quickly
advance in his diet and ambulate throughout the hospital. Routine daily exercise and encouragement to eat frequently are key steps, both physically and mentally, for
the patient. Anticoagulation management was started
on the evening of postoperative day 1, following chest
closure, with the initiation of coumadin and aspirin.
The goal of the therapy was to maintain an international
normalized ratio (INR) of 2.5 3.0. The patient was
supported with the TAH-t for 11 days prior to successful
transplantation.

Chest CT with 3D reconstruction and

modeling
The 3D CT reconstruction played an important part in
the diagnosis of the postoperative complication. Postimplantation CXR (Figure 5) revealed left lung opacification despite mechanical ventilation and recruitment
exercises. Contrast-enhanced CT was acquired of the
thoracic cavity and confirmed the findings of the CXR. To
analyze possible obstruction of pulmonary vasculature

86

Perfusion 29(1)

Figure 4A-E. Stages of the TAH-t Surgical Implantation.

Following median sternotomy, all structures are identified and isolated. The drivelines for both ventricles are passed subcutaneously before
heparinization of the patient. The ventricles are positioned lateral to the sternotomy and well covered. Following systemic heparinization, CPB is
instituted and the heart is fibrillated. Dissection around the aorta and pulmonary artery is limited to the proximal portion of the aorta in anticipation
of later transplantation. The ventriculectomy is significantly different from that of transplantation. The annuluses of the tricuspid and mitral valves are
preserved. An incision is made on the ventricular side of the atrioventricular (AV) groove of the right ventricle, extending anteriorly across the right
ventricular outflow tract and just proximal to the pulmonary valve (Figure 6A, B). Posteriorly, the excision extends to the interventricular septum
and across the septum on the left ventricular side of the AV groove, preserving the entirety of the mitral annulus. The anterior and posterior lines
of incision are dissected to the level of the pulmonary bifurcation. The excess muscle is trimmed down to near the AV valves. The great vessels are
then separated (Figure 6C). The coronary sinus is over-sewn. Both atrial cuff complexes are encircled with Teflon felt buttresses to strengthen the
anastomosis and quick connect (Figure 6D). Hemostasis is checked with the plastic tester, which fits within the quick connect. The pulmonary and
aortic anastomoses are completed (Figure 6D). The conduits are cut to the appropriate length, usually 3 to 5 cm, and sewn in place. Following this,
using the aortic tester, they are inserted into both quick connects and checked for leaks. Once hemostasis of all suture lines has been established,
TAH-t insertion begins with placement of the left ventricle (Figure 6E). A careful assessment of the relationship of ventricles to their respective
atrial cuffs and outflow grafts should be considered prior to completing the atrial connection. The left ventricle should be filled with saline through
the aortic valve while the patient is placed in a steep Trendelenburg position and the aorta is vented. The right ventricle is filled by restoring the
caval blood flow to the atrium and air is evacuated via the pulmonary outflow graft. The TAH-t is started on low rate and pressure settings until
satisfactory de-airing of the ventricles is complete.17 Images Courtesy: syncardia.com

Figure 5. Chest radiograph following TAH-t Implantation.

and bronchi, 3D reconstruction was performed. Image

processing via segmentation translated the 2D slice
data into 3D computer images. Cardiovascular, respiratory and skeletal (including costal cartilage) structures
were segmented from the CT, using Mimics (Materialise,

Leuven, Belgium), a medical imaging software suite.

These segmented masks were reconstructed into a 3D
surface mesh model and imported into 3D modeling
software (Figure 6a). As a majority of the TAH-t is
radiographically transparent, only the internal volumes
(filled with air or blood) could be segmented. To obtain
the complete TAH-t (not just the lumen), the surface of
the TAH-t required reconstruction as well. The same
model TAH-t (70 cc) was laser scanned using a
ZScanner 700 (3DSystems, Rock Hill, SC). This process
of scanning projects a laser onto the surface of an
object, which is reflected back to the scanner cameras.
Utilizing photogrammetric processing techniques, the
scanner reconstructs the physical object in a virtual
environment. The computer model of the TAH-t was
imported into the 3D modeling software alongside the
patients anatomical reconstruction. The surface mesh
of the TAH-t was aligned to its segmented lumen from
the CT reconstruction. The combination of patient data
and scanned data provided an inclusive and accurate
view of the TAH-t impact on the anatomy of the patient,
as further described in the preceding section (Figure
6A and B). The 3D reconstruction confirmed that the
size and location of the TAH-t were not constricting
major cardiovascular or respiratory structures. The
postoperative care confirmed the findings of the 3D
reconstruction.

87

Park et al.

Figure 6A and B. (A) 3D Reconstruction of cardiopulmonary structures, skeletal structures and TAH device in situ.
(B) 3D reconstruction with osseous structures subtracted.

Conclusion
Cardiomyopathy in infants and children with heart disease is one of the most common causes for death.1 An
estimated 1,000-5,000 new cases of cardiomyopathy are
diagnosed each year in pediatrics.2 DCM is the leading
cause for heart transplantation among pediatric patients.
Survival of these children depends on prompt accurate
diagnosis and aggressive heart failure therapy. Some children respond to conventional heart failure treatment
while others further decompensate, requiring mechanical circulatory support and heart transplantation.5
Unfortunately, children listed for heart transplantation
in the USA face the highest waiting list mortality irrespective of age.4,5 The emergent deployment of pediatric
cardiac assist technologies has been shown to benefit the
pediatric patient in decompensated heart failure with
multi-organ dysfunction while awaiting a transplant. In
selected pediatric patients, the TAH-t is the assist device
of choice for decompensatory biventricular heart failure.
Commonly encountered VAD issues, such as valverelated problems, left ventricular clot formation, right
heart failure, arrhythmias and the need for inotropes, are
eliminated with the TAH-t.7,9-11 A limiting factor for the
use of the TAH-t in children is the recommended minimum body surface area of 1.7 m2 for proper fit and function. Currently, there are a small number of certified
pediatric centers, with our center being the third transplant center to implant a TAH-t in a child.12 The need for
modification of the device in children with complex congenital anatomies and the absence of current anticoagulation protocols for children make increased pediatric
implants a challenge. Quick transition to ambulation and
robust diets are critical milestones for all TAH-t patients
and can directly impact quality of life and functional

status during device support and subsequently following

transplant.13-15
This case study confirms that the TAH-t can be
successfully deployed in selected pediatric patients
who have a smaller BSA than 1.7 m2. This patients
postoperative course was complicated by left lung atelectasis. A chest CT with 3D reconstruction was crucial in visualizing the pulmonary arteries and veins
and bronchial structures to rule out compression from
the device. Accordingly, this process has shown that
virtual modeling can predict appropriate fit of the
device and assist with trouble-shooting acute complications in pediatric patient populations. The success
of utilizing CT with segmentation and 3D modeling,
as seen in this case study, demonstrates the potential
for patient-specific device fit analysis. A study by
Dowling, et al. acknowledged CT with 3D reconstruction as a powerful tool in predicting appropriate
fit of any potential candidate who was considered
for implantation of the AbioCor Implantable
Replacement Heart System.16 SynCardia has recently
manufactured a new 50 cc TAH-t that has received a
Humanitarian Device Designation for pediatric bridge
to transplant. As development and testing of this new
technology progresses, its future deployment and utilization will broaden pediatric application of TAH-t.
The Freedom Driver has been recently introduced
into clinical trials as a tool to further improve quality
of life and expand the functionality of the TAH-t.15 To
appreciate the advantages of this device for pediatrics,
more experience with placement and further development of novel tools are needed.
Declaration of conflicting interest
The authors declare that there are no conflicts of interest.

88
Funding
This research received no specific grant from any funding
agency in the public, commercial or not-for-profit sectors.

References
1. Pediatric
Cardiomyopathy
Defined.
Childrens
Cardiomyopathy Foundation. 2013. Available at http://
www.childrenscardiomyopathy.org/site/overview.php.
Accessed Jan 22 2013.
2. Pediatric Heart Transplant Statistics. International Society
of Heart and Lung Transplantation. 2012. Available at
http://www.ishlt.org/registries/slides. Accessed Jan 22
2013.
3. OPTN/SRTR 2011 Annual Data Report: Heart. Organ
Procurement and Transplantation Network. 2011.
http://optn.transplant.hrsa.gov/data/annualReport.asp.
Accessed Jan 22 2013.
4. Almond CS, Thiagarajan RR, Piercey GE, et al. Waiting
list mortality among children listed for heart transplantation in the United States. Circulation 2009; 119:
717727.
5. Fuchs A, Netz H. Ventricular assist devices in pediatrics.
Images Paediatr Cardiol 2001; 3: 2454.
6. Copeland JG, Smith RG, Arabia FA, et al: CardioWest
Total Artificial Heart Investigators. Cardiac replacement
with a total artificial heart as a bridge to transplantation.
N Engl J Med 2004; 351: 859867.
7. LePrince P, Bonnet N, Varnous S, et al. Patients with a
body surface area less than 1.7 m2 have a good outcome
with the Cardiowest total artificial heart. J Heart Lung
Transplant 2005; 24: 15011501.

Perfusion 29(1)
8. Sanders DB, Sowell SR, Willis B, et al. The role of extracorporeal life support in acute myocarditis: a bridge to
recovery? JECT 2012; 44: 205209.
9. Smith MC, Arabia FA, Tsau PH, et al. Cardiowest total
artificial heart in a moribund adolescent with left ventricular thrombi. Ann Thorac Surg 2005; 80: 14901492.
10. Copeland JG, Smith RG, Arabia FA, et al. Total artificial heart bridge to transplantation: a 9-year experience with 62 patients. J Heart Lung Transplant 2004; 23:
823831.
11. Copeland JG, Smith RG, Arabia FA, et al. Comparison of the
Cardiowest total artificial heart, the Novacor left ventricular
assist system and the Thoratec ventricular assist system in
bridge to transplantation. Ann Thorac Surg 2001; 71: S9297.
12. Patient Demographic Data. Syncardia patient Registry.
Accessed Jan 2013.
13. Nicholson C, Paz JC. Total artificial heart and physical
therapy management. Cardiopulm Phys Ther J 2012; 21:
1321.
14. Friedline K, Hassinger P. Total artificial heart freedom
driver in a patient with end-stage biventricular heart failure. AANA J 2012; 80: 104112.
15. Jaroszewski DE, Anderson EM, Pierce CN, Arabia FA.
The SynCardia freedom driver: a portable driver for discharge home with the total artificial heart. J Heart Lung
Transplant 2011; 30: 844845.
16. Dowling RD, Gray LA, Etoch SW, et al. Initial experience
with the Abiocor implantable replacement heart system.
Ann Thorac Surg 2004; 127: 131141.
17. Arabia FA, Copeland JG, Pavie A, Smith RG. Implantation
technique for the CardioWest total artificial heart. Ann
Thorac Surg 1999; 68: 698704.

Copyright of Perfusion is the property of Sage Publications, Ltd. and its content may not be
copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.