Percutaneous Pulmonary Valve Replacement: 3-

Month Evaluation of Self-Expanding Valved Stents

Tim Attmann, MD, Ren Quaden, MD, Thomas Jahnke, MD, PhD,
Stefan Muller-Hulsbeck, MD, PhD, Andreas Boening, MD, PhD,
Jochen Cremer, MD, PhD, and Georg Lutter, MD, PhD
Departments of Cardiovascular Surgery and Radiology, Christian-Albrechts-University of Kiel, School of Medicine, Kiel, Germany
NEW TECHNOLOGY

Purpose. In a recent study our group established an acute animal model of percutaneous
pulmonary valve replacement using self-expanding nitinol stents. The present study was
performed to evaluate these valved stents over a 3-month period.
Description. Bovine jugular xenografts were sutured into nitinol stents. Transfemoral
implantation in the pulmonary position using a modified commercially available applica-
tion device (with a 22-French outer diameter) was evaluated in 9 sheep.
Evaluation. Two sheep died shortly after successful valved stent implantation due to
internal venous hemorrhage. Another 1 sheep died 2.5 months after the procedure due to
vegetations on the neovalve leading to subtotal stenosis. All other animals survived the
3-month study time (n 6). An orthotopic pulmonary valved stent position was achieved
in 4 animals and a supravalvular position in 1. During the deployment procedure, rhythm
disturbances occurred in all animals, and mean arterial blood pressure dropped from 83.9
26.0 mm Hg to 68.3 22.3 mm Hg (p 0.006) (n 5). The peak-to-peak transvalvular
gradient was 5.1 4.0 mm Hg initially (n 5), and 3.6 1.6 mm Hg at follow-up (n 5).
Three-month angiographic and echocardiographic follow-up confirmed competent neo-
valves without paravalvular leakages.
Conclusions. After 3 months of implantation, percutaneously implanted memory nitinol
valved stents demonstrated good function in the sheep.
(Ann Thorac Surg 2006;82:708 14)
2006 by The Society of Thoracic Surgeons

P ercutaneous pulmonary and aortic valve implanta-

tions have been clinically introduced by Bonhoeffer
and colleagues [1] in 2000 and Cribier and colleagues [2]
in 2002. This appealing new approach to valve disease
has made remarkable progress. Recently, Bonhoeffer and
colleagues [1] emphasized the potential impact of percu-
taneous pulmonary valve stent implantation on right
ventricular outflow tract (RVOT) reintervention. This
method proved to be a promising additional and compli-
mentary approach to a successful surgical program [3].
Nevertheless the anatomic spectrum of pulmonary re-
gurgitation after surgery for congenital RVOT disease is
broad. With currently available valved stents and devices
it is not possible to treat all the concerned patients. So far,
wide or severely calcified and kinked RVOTs are not
suitable for percutaneous therapy. To overcome these
problems new stent designs and operative and interven-
tional hybrid approaches are under investigation [4, 5].
Accepted for publication Jan 26, 2006.
Address correspondence to Dr Lutter, Department of Cardiovascular
Surgery, Christian-Albrechts-University of Kiel, School of Medicine,
Arnold-Heller-Str. 7, Kiel, 24105 Germany; e-mail: lutter@kielheart.uni-
kiel.de.
Our group demonstrated the feasibility of totally per-
cutaneous transfemoral pulmonary valve implantation in
an ovine model using self-expanding nitinol stents [6].
The continuously exerted radial force of self-expanding
stents and their high flexibility assure a geometric adap-
tation to anatomical and tissue-property changes. There-
fore the use of this kind of stent may be beneficial for the
percutaneous treatment of patients with the previously
mentioned anatomical anomalies.
The aim of this study was to evaluate the function of
percutaneously implanted self-expanding stents carrying
a biological valve into the pulmonary position of juvenile
sheep during a 3-month period using angiographic and
echocardiographic, hemodynamic, and macroscopic
measurements.
Technology
Valved segments of bovine jugular veins were sutured
into nitinol stents with a length of 28 mm and a diameter
of 24 mm as described previously [6]. The maximal radial
force when expanded to 20 mm at 37C is approximately
0.1 Newton.

2006 by The Society of Thoracic Surgeons 0003-4975/06/$32.00

Published by Elsevier Inc doi:10.1016/j.athoracsur.2006.01.096
Ann Thorac Surg
2006;82:708 14
Fig 1. Pre-procedural loading of the application device with valved
nitinol stent.
Technique
Animal Model and Implantation Procedure
Nine studies in sheep weighing 25 to 32 kg are reported.
Animals received humane care in accordance with the
Principles of Laboratory Animal Care and the Guide
for the Care and Use of Laboratory Animals (National
Institutes of Health, Publication 85-23, revised 1985). A
committee on animal research at Kiel University ap-
proved the protocol.
The anesthesia and implanting procedure were carried
out as previously described [6]. Seven French sheaths
were introduced into the left femoral artery and vein for
hemodynamic measurements. Through a 24-French
sheath positioned in the right groin, the 22-French appli-
cation device (Fig 1) was inserted and the valved stents
were deployed directly over the native pulmonary valve
under fluoroscopic control. The 24-French sheath was
removed and after manual compression for 20 minutes,
the skin in the right groin was closed with a single stitch.
The animals were studied by analyzing various measure-
ments described as follows.
Afterward the sheep returned to the controlled animal
facility where their general health was checked daily.
Animals were not treated for anticoagulation or inhibi-
tion of platelet aggregation. After the occurrence of an
endocarditis in 1 animal, prophylactic i.m. antibiotic
therapy (i.e., penicillin and streptomycin) was adminis-
tered weekly.
Three months later the animals were reanalyzed (de-
scribed as follows) and were sacrificed.
Angiography
An angiography unit (Multistar Top, Siemens, Erlangen,
Germany) was used for fluoroscopic assessment of the
position and function of the neovalves.
Echocardiography
Transthoracic echocardiography (TTE) was performed at
the 3-month follow-up with the Ultrasound System Five
(Vingmed Sound, Horten, Norway) using a multiplane
NEW TECHNOLOGY ATTMANN ET AL 709
PERCUTANEOUS PULMONARY VALVE REPLACEMENT
2.5-MHz transducer (n 4). The echocardiograpic probe
was applied at the right hemithorax between the fourth
and sixth intercostal spaces. The measurements were
recorded and stored on a magneto-optical disk and
analyzed offline by an experienced investigator (RQ).
Hemodynamic Measurements
Arterial pressure, right ventricular pressure, left ventric-
ular pressure, and pulmonary artery pressure (Micro-Tip
Millar Catheter, Millar Instruments, Inc, Houston, TX)
were recorded using Haemodyn-Software (Hugo Sachs
Electronics, Hugstetten, Germany). The software calcu-
NEW TECHNOLOGY
lates the maximum of the first derivative of the ventric-
ular pressures (dp/dtmax) as an indicator for the contrac-
tility and its minimum (dp/dt min ) indicating the
relaxation behavior.
Macroscopic Examination
Valved stents were grossly inspected and photographs
were taken. Special attention was given to the retraction
of the cusps or any deformed or indurated parts of the
valve. The atrial and ventricular chambers and the pul-
monary artery were exposed to look for catheter-induced
damage and to look for penetration of stent struts.
Roentgenogram Assessment
Roentgenogram examination of the explanted valved
stents (n 4) was performed under mammography
conditions to demonstrate and localize macroscopic
calcification.
Statistics
Values are presented as mean standard deviation. Data
were analyzed based on the Wilcoxon test to compare
related data for non-normally distributed data using the
SPSS 10.1 software (SPSS, Inc, Chicago, IL). The p values
less than 0.05 were considered statistically significant.
Clinical Experience
Results
Six of the total 9 sheep from the study survived the entire
3-month observation period. Two sheep died 2 and 4
hours after successful valved stent implantation due to
internal venous hemorrhage from the right common iliac
vein and the caudal vena cava. The vessels were dam-
aged by the 24-French sheath during its insertion. An-
other animal died 2.5 months after the percutaneous
procedure due to subtotal pulmonary stenosis caused by
endocarditic vegetations.
The mean diameter of the pulmonary annuli was 20.0
1.4 mm (range, 18.0 21.8 mm) as revealed by angiog-
raphy. Valved stents with a maximal outer diameter from
20.6 to 23.7 mm were used.
Mean duration of the procedure from insertion of the
application device through the sheath to deployment was
52.9 seconds (range, 9.0 to 124.5 seconds). The fluoros-
copy time of the entire procedure ranged from 7.0 to 26.7
minutes (mean, 11.9 minutes).
 710 NEW TECHNOLOGY ATTMANN ET AL Ann Thorac Surg
PERCUTANEOUS PULMONARY VALVE REPLACEMENT 2006;82:708 14

Fig 2. Representative angiography at the

3-month follow-up demonstrating a compe-
tent valved stent in the pulmonary position.
Chronological sequence from a to d. (a)
Early phase; (b) complete contrast in the
pulmonary arteries; (c) still complete con-
trast in the pulmonary arteries without re-
gurgitation; (d) late phase.
NEW TECHNOLOGY

Angiography and 3.6 1.6 mm Hg at the 3-month follow-up (n 6)

Angiography revealed no insufficiencies after 3 months (Table 1).
and showed competent valved stents in all 6 surviving
animals (Fig 2). Macroscopic Examination
Postmortem examination after 3 months confirmed the
Echocardiography correct position of the valved stents in 5 sheep. In 1
The TTE demonstrated a laminar blood flow across the
valved stents in all four analyzed cases. One mild central
insufficiency and no paravalvular leakage were detected.
Left ventricular and right ventricular functions and di-
mensions seemed normal.
The mean systolic internal diameter of the valved
nitinol stents was measured at 15.6 1.1 mm, and the
mean internal diastolic diameter was measured at 20.1
0.9 mm (Fig 3).

Hemodynamic Measurements
During the deployment procedure, rhythm disturbances
occurred in all animals and the mean arterial blood
pressure dropped from 83.9 26.0 mm Hg to 68.3 22.3
mm Hg (p 0.006). Five minutes after deployment, the
mean arterial blood pressure recovered to 83.5 23.7 mm
Hg (p 0.002) (n 5). There were no significant changes
in the contractility or relaxation behavior of the right and
left ventricle. Fig 3. Representative transthoracic echocardiographic image after 3
The peak-to-peak gradient across the valved stents was months of implantation. Note marked difference in (1) systolic
5.1 4.0 mm Hg at 5 minutes after the implantation valved stent diameter and (2) diastolic valved stent diameter.
Ann Thorac Surg NEW TECHNOLOGY ATTMANN ET AL 711
2006;82:708 14 PERCUTANEOUS PULMONARY VALVE REPLACEMENT

Table 1. Hemodynamic Data

Valved Stent Implantation
3-Month
Before During 5 Minutes After Follow-up

HR (min1) Mean 120.2 139.1 131.5 119.5

SD 28.1 33.7 26.3 12.1
MAP (mm Hg) Mean 83.9 68.3a 83.5b 108.9
SD 26.0 22.3 23.7 5.3
LV dp/dtmax (mm Hg/s) Mean 2064.9 2271.5 2295.5 2068.4
SD 752.4 410.6 584.8 457.0
LV dp/dtmin (mm Hg/s) Mean 2293.4 1897.6 2348.7 2510.5
SD 697.5 529.8 531.9 175.2

NEW TECHNOLOGY
c c
RV dp/dtmax (mm Hg/s) Mean 523.9 396.3
SD 86.5 51.0
RV dp/dtmin (mm Hg/s) Mean c c
354.9 374.4
SD 120.5 138.0
c c
RVP (mm Hg) Mean 28.1 25.8
SD 5.6 5.5
c c
PAP (mm Hg) Mean 23.1 21.1
SD 6.0 5.5
c c
RV to PA systolic gradient (mm Hg) Mean 5.1 3.6
SD 4.0 1.6
a b
Significant MAP drop compared to value before valved stent implantation; Significant increase in MAP compared to value during valved stent
c
implantation; Before and during the implantation procedure no data could be obtained on the RVOT.

HR heart rate; LV left ventricle; MAP mean arterial pressure; PA pulmonary artery; PAP pulmonary artery pressure; RV
right ventricle; RVP right ventricular pressure.

animal the valved stent was found in the pulmonary
artery at approximately 1 cm distal to the pulmonary
valve.
No paravalvular defects were visible. The valved stents
were pliable and the leaflets were thin without indura-
tions. In one valve there was an evident lack of coaptation
with one retracted leaflet. Slight fibrous overgrowth was
seen at the inflow portions of two valved stents (Fig 4).
Cardiac structures were unscathed. No macroscopic
damage of the pulmonary artery was noted. In particular
there was no penetration of stent struts.
Roentgenogram Assessment
The wall of explanted stent valves appeared mildly
calcified at roentgenogram examination. No calcification
of valve leaflets could be detected (Fig 5).
Comment
The surgical reconstruction of the RVOT in children with
congenital heart defects including pulmonary valve re-
placement has a low morbidity and mortality [7]. The
implantation of a pulmonary homograft is considered to
be the gold standard. However graft deterioration is very
common and therefore reoperations are often unavoid-
able in adolescents [8]. Percutaneous pulmonary valve
implantation is emerging as an alternative or additional
option for a successful surgical scheme, recently even
being introduced into clinical practice [1, 9]. Bonhoeffer
and colleagues [1] group successfully performed 35 pro-
cedures with no early deaths and only three procedural
complications. Early and late outcome was comparable
with that of 94 surgically treated patients in the control
group [3]. The advantages of minimally invasive proce-
dures are obvious (i.e., smaller incisions, less postopera-
tive pain, shorter hospital stay, and lower costs).
This is one of the first reports of evaluation of percu-
taneously implanted self-expanding valved nitinol stents
in the pulmonary position. So far, balloon-expandable
valved stents are used in clinical practice [3]. The poten-
tial advantages of nitinol stents concerning the preserva-
tion of the valve and the deployment maneuver were
previously discussed [6]. Interestingly in the present
study, the echocardiographic examinations nicely dem-
onstrated adaptation in the diameter of the valved stent
to the radial motion of the RVOT during the cardiac
cycle. The soft nitinol stents ability to adapt to the
surrounding structures could be beneficial for a physio-
logic blood flow in the proximal great arteries. However
the radial force of this stent will not be high enough to
overcome calcified valves. To reach a sufficient valve
area, a preceding balloon dilation or resection procedure
would be required [6, 9].
For the first time meticulous hemodynamic monitor-
ing is reported for percutaneous heart valve replace-
ment. The contractility of both right and left ventricle
and their relaxation properties were unchanged at the
end of the study. Thus the neovalves had no negative
effect on the cardiac function. During the implantation,
rhythm disturbances and a significant decrease of
mean arterial pressure was observed without affecting
 712 NEW TECHNOLOGY ATTMANN ET AL
PERCUTANEOUS PULMONARY VALVE REPLACEMENT
NEW TECHNOLOGY
Fig 4. Representative macroscopic image of an ovine heart at post-
mortem. Right ventricle and its outflow tract are opened. Valved
stent in pulmonary position with thin and transparent leaflets. The
visible central defect of the valve disappeared during water rinse
testing.
the contractility as indicated by the left ventricular
dP/dtmax.
The Millar catheter measurements revealed a low
transvalvular pressure gradient shortly after the implan-
tation as well as after 3 months. Macroscopy and roent-
genogram assessment revealed no calcification of the
leaflets, which would affect the valves function.
Limitations
The authors wish to address several limitations of this
study. First, the large diameter of the application
device has to be mentioned. Two animals died shortly
after the percutaneous procedure due to hemorrhage
Fig 5. Representative roentgenogram images
of a valved stent after 3 months of pulmo-
nary implantation: (a) side view and (b)
view from above. Note only minimal calcifi-
cation of the wall of the bovine jugular vein
and absent calcification of the leaflets.
Ann Thorac Surg
2006;82:708 14
from venous structures after injury by the large intro-
ducer sheath. This problem was not observed in our
acute experiments, possibly because of the short fol-
low-up period [6]. Only a further reduction of our
catheter systems size will allow the clinical application
in children without an inguinal cut down. Recently,
Ruiz and coworkers [10] published a remarkable long-
term study in pigs of catheter-placed low-profile bio-
degradable pulmonary valves made of small intestinal
submucosa [10]. Their square stent-based valve re-
quired only an 8-French delivery system. Such a low-
profile valve combined with the small flexible delivery
system is ideal for use in pediatric cardiology.
Second, tissue deterioration and ensuing loss of func-
tion is a major problem with biological heart valve
prostheses. In the present study, one of six valves was
slightly affected. The 3-month observation period may be
too short to provide valid information on the durability of
the biological valve. For that reason the development of
catheter-based techniques for the implantation of me-
chanical heart valves that are not affected by this problem
should not be overlooked. An interesting feasibility study
of percutaneous aortic disc valve prosthesis was reported
by Sochman and coworkers [11]. Further studies in this
field should be undertaken.
Third, bovine jugular veins have a limited maximal
diameter of approximately 24 mm. Therefore they are not
suitable for an enlarged RVOT or pulmonary trunk.
Boudjemline and colleagues [5] are working intensively
on this problem. In an ovine study they successfully
tested a two-stage surgical and percutaneous procedure
to overcome the previously mentioned problem [5]. Oth-
ers chose a transthoracic beating heart approach to
introduce a self-expanding valved stent of large size in
the pulmonary position of pigs [4]. In contrast, our
transfemorally inserted cylindrical valved stent can not
provide a solution on this matter.
Nonetheless, percutaneously implanted memory niti-
nol valved stents are evaluated during a 3-month period
Ann Thorac Surg
2006;82:708 14
in an ovine model in the present study showing a good
structural and functional outcome.
Disclosures and Freedom of Investigation
Dr Lutters project of percutaneous valve replacement is
supported by the German Research Foundation, Bonn,
Germany (Grants LU 663/4-1, LU 663/4-2). The authors of
this article had full control of the design of the study,
methods used, outcome measurements, analysis of data,
and production of the written report. The authors had no
financial relationship with any companies.
The authors are indebted to Beata Hoffmann, Marion Frahm,
Christian Knig, Florian Alten, Andrea Freistedt, Kristin Rum-
berg, and Andreas Bohlen for their technical and operative
assistance. Fritz Schaefer, MD, contributed to the roentgeno-
gram assessment.
References
1. Bonhoeffer P, Boudjemline Y, Saliba Z, et al. Percutaneous
replacement of pulmonary valve in a right-ventricle to
pulmonary-artery prosthetic conduit with valve dysfunction.
Lancet. 2000;356:14035.
2. Cribier A, Eltchaninoff H, Bash A, et al. Percutaneous trans-
catheter implantation of an aortic valve prosthesis for calcific
aortic stenosis: first human case description. Circulation
2002;106:3006 8.
3. Coats L, Tsang V, Khambadkone S, et al. The potential
impact of percutaneous pulmonary valve stent implantation
on right ventricular outflow tract re-intervention. Eur J Car-
diothorac Surg 2005;27:536 43.
INVITED COMMENTARY
The article by Attmann and coworkers [1] presents the
results of a chronic animal feasibility study of using a
self-expanding nitinol stent supporting a bovine jugular
vein valve for percutaneous pulmonary valve replace-
ment, and represents continuing work previously re-
ported by the same group [2]. In this experiment, 9 sheep
received the nitinol stented bovine valve percutaneously
with 2 animals experiencing lethal bleeding complica-
tions attributed to the procedure and 1 animal succumb-
ing to endocarditis at 2.5 months. The remaining 6
animals, surviving to 3 months, demonstrated good valve
function with one mild central insufficiency and no para-
valvular leaks. This is early pioneering work, and while
the vascular complications of insertion are concerning,
the investigators and readers should not lose sight of the
true significance of this work (ie, the limitations of
balloon-expandable stent technology to treat calcified or
severely angulated right ventricular outflow tracts). Self-
expanding stents may be an answer to this problem,
2006 by The Society of Thoracic Surgeons
Published by Elsevier Inc
NEW TECHNOLOGY ATTMANN ET AL 713
PERCUTANEOUS PULMONARY VALVE REPLACEMENT
4. Zhou JQ, Corno AF, Huber CH, Tozzi P, von Segesser LK.
Self-expandable valved stent of large size: off-bypass im-
plantation in pulmonary position. Eur J Cardiothorac Surg
2003;24:212 6.
5. Boudjemline Y, Schievano S, Bonnet C, et al. Off-pump
replacement of the pulmonary valve in large right ventricu-
lar outflow tracts: a hybrid approach. J Thorac Cardiovasc
Surg 2005;129:8317.
6. Attmann T, Jahnke T, Quaden R, et al. Advances in experi-
mental percutaneous pulmonary valve replacement. Ann
Thorac Surg 2005;80:969 75.
7. Kanter KR, Budde JM, Parks WJ, et al. One hundred pulmo-
nary valve replacements in children after relief of right
ventricular outflow tract obstruction. Ann Thorac Surg 2002;
73:1801 6.
NEW TECHNOLOGY
8. Wells WJ, Arroyo H Jr, Bremner RM, Wood J, Starnes VA.
Homograft conduit failure in infants is not due to somatic
outgrowth. J Thorac Cardiovasc Surg 2002;124;88 96.
9. Lutter G, Ardehali R, Cremer J, Bonhoeffer P. Percutaneous
valve replacement current state and future prospects. Ann
Thorac Surg 2004;78:2199 206.
10. Ruiz CE, Iemura M, Medie S, et al. Transcatheter placement
of a low-profile biodegradable pulmonary valve made of
small intestinal submucosa: a long-term study in a swine
model. J Thorac Cardiovasc Surg 2005;130:477 84.
11. Sochman J, Peregrin JH, Pavcnik D, et al. Percutaneous
transcatheter aortic disc valve prosthesis implantation: a
feasibility study. Cardiovasc Intervent Radiol 2000;23:384 8.
Disclaimer
The Society of Thoracic Surgeons, the Southern Thoracic
Surgical Association, and The Annals of Thoracic Surgery
neither endorse nor discourage use of the new technol-
ogy described in this article.
although nitinol-based stents, as used in this animal
study, may not possess enough radial force (0.1 Newton)
to overcome even the most minimal stricture or calcifi-
cation. As an example, most of the current work in
percutaneous aortic valve replacement has been with
stainless steel stents. Regardless, self-expanding stents
possess some properties that may make them superior to
their balloon-expanding stent counterparts. By virtue of
their continuously exerted outward radial force, self-
expanding stents are more likely to adapt to ongoing
tissue property changes at the implantation site and
therefore less prone to migration. Furthermore they
possess greater flexibility than the stiffer balloon-
expanded stents that would provide an advantage when
one expects the stent to conform to a more complicated
anatomic geometry.
Although there may be less anatomic hurdles with
percutaneous pulmonary valve replacement than with
percutaneous aortic valve replacement, many impor-
0003-4975/06/$32.00
doi:10.1016/j.athoracsur.2006.02.068