Cardiac Catheterization in Congenital Heart Disease: Pediatric and Adult - Part 10 doc

CHAPTER 31 Purposeful vascular perforations
844
and to the valve. Valve perforations with each of these
“instruments”, and from both approaches, occasionally
were successful. Unfortunately, the force and “push” that
were necessary to penetrate the valve, often pushed the
supporting catheter, which contained the stiff needle
and/or the retrograde wire, backwards and away from
the plate-like valve rather than causing the perforating
instrument to puncture the valve. When the guide or sup-
port catheter was pushed away, it prevented the perfora-
tion, or even worse, displaced the sharp instrument away
from the center of the “plate-like” valve and into the
perivalvular area. A puncture into the adjacent areas
resulted in perforation into the pericardium and/or an
adjacent chamber/vessel, often with catastrophic results.
Experimental, and recently, clinical experience demon-
strated the feasibility of “drilling” through tissue, and
more specifically pulmonary valve tissue, with laser or
radio-frequency (RF) energy. Both energy sources have
demonstrated considerable success in penetrating the
valve. When this is followed by a balloon dilation of
the valve, very adequate pulmonary valve openings
are achieved allowing unobstructed prograde blood flow
into the pulmonary artery. The amount of this prograde
flow then is dependent on the size of the tricuspid
valve, the potential volume and the compliance of the
right ventricle.
Laser energy was the first energy source to be used in
clinical trials in Europe
5–7

. The laser energy is delivered
through a fine fiberoptic strand or “wire” to a very specific
area, and proved very successful at perforating the atretic
pulmonary valve in patients with pulmonary valve atre-
sia by essentially vaporizing the tissue in front of the
beam. The laser beam, unfortunately, also easily continues
to perforate any tissues in its path beyond and/or adja-
cent to the valve. The laser energy was successful at
perforating an opening in the atretic pulmonary valve
in approximately 80% of the small number of patients in
whom it was tried initially. The opening allowed the pas-
sage of a guide wire through the valve and subsequent
dilation of the valve.
Although successful in perforating the valve, the laser
system has several disadvantages in addition to the poor
controllability of the depth of penetration. The Excimer
Laser™ generator is large and very expensive. Unless
the particular pediatric cardiac catheterization laboratory
works in conjunction with an adult catheterization labor-
atory and/or performs many laser-assisted pacemaker
lead extractions, the capital expense of the laser generator
“just” for the very few pulmonary atresia patients who
present to even large pediatric cardiac centers, cannot be
justified easily. In addition to the ease of perforating
unwanted structures within the heart with the laser, the
intense laser energy also carries a risk of “stray” laser
beams in the area of the patient, which can create retinal
damage to the operators and other employees in the
catheterization laboratory. As a consequence, all person-
nel in the laboratory are required to wear special, some-

what cumbersome, protective eye wear. The fiberoptic
laser “wires” are expensive and finally, and absolutely the
greatest deterrent to operators in the US until very
recently: there was no laser “wire” that was approved for
use outside of specific protocols in the United States.
As an excellent alternative to laser energy, radio-
frequency energy delivered through a very fine insulated
wire can also be used to perforate atretic tissues, and the
atretic pulmonary valve in particular. The RF energy is
less powerful for perforating structures, but it is consider-
ably more controllable than the laser energy. A radio-
frequency generator for perforation is considerably less
expensive and less complex than any laser generator. In
addition, radio-frequency generators are commonly avail-
able in pediatric and congenital catheterization labor-
atories, where they are used for the ablation of abnormal
intracardiac “electrical” conduction tracts. However, the
current, and now standard RF “ablation” generator, which
is low impedance and uses low-voltage (30–50 volts), high-
power (30–50 watts) sustained (60–90 second) energy
for the “ablation” of tissues without perforation, requires
significant electrical modifications to convert the gener-
ator into a high-impedance, high-voltage (150–180 volts),
low-power (3–5 watts) and short-duration (1–2 second)
energy generator, which is necessary for “perforation”.
The special BMC Radio Frequency Perforation
System™ (Baylis Medical Co. Inc., Montreal, Canada) is a
generator that is designed and dedicated specially for
“perforation”. The generator now is available commer-
cially (even in the US) and is reasonably priced. This

generator has built into it the necessary high-voltage,
low-power and short-duration pulses of energy that are
necessary to generate the high impedance necessary for
perforation. A single use RF perforating “catheter”
matched to the RF generator along with an “injection”
coaxial catheter and a connecting cable are available as
“perforating kits” to be used with the specific RF gener-
ator. The total RF generator and the “catheter kit” are
approved for “intravascular perforation”, even in the
United States.
The disposable “kit” consists of the Nykanen Radio
Frequency Perforation Catheter™ and a special Coaxial
Injectable Catheter™ (Baylis Medical Co. Inc., Montreal,
Canada). The “perforation catheter” is a 0.024″, 265 cm
long, teflon “catheter” tightly bound over a 0.016″ conduct-
ance wire. Only the distal 1.5 mm and several mm of
the proximal bare wire are exposed. The teflon over the
wire provides an insulated coating to the wire and pro-
duces a relatively stiff, “pushable” shaft for the combina-
tion. The distal region of the “catheter” is flexible and can
be bent or pre-formed into a specific curve for easier
CHAPTER 31 Purposeful vascular perforations
845
maneuverability. This allows the perforating “catheter”
to be maneuvered similarly to a fine torque-controlled
guide wire. The proximal end of the teflon “catheter” has
no “hub” but attaches to a removable connecting cable,
which, in turn connects it to the generator. The “coaxial
injectable catheters” are thin walled 0.035″ or 0.038″ dia-
meter, 145 cm long catheters with a distal radio-opaque

marker and a floppy, distal 10 cm tip. The inner diameters
of the two catheters are 0.024″ and 0.027″, respectively.
These catheters also have removable hubs so that the
coaxial catheter and the contained perforation catheter
together can act as a “thick guide wire” over which other
catheters (balloon catheters) can be introduced.
Technique for perforation of the pulmonary
valve in patients with pulmonary atresia and
intact ventricular septum
The diagnosis of pulmonary atresia with or without a
ventricular septal defect usually is made clinically in the
newborn period with confirmation by echocardiographic
evaluation. By the time these infants are seen by a cardi-
ologist, they usually already are receiving prostaglandin
to keep the ductus arteriosus open, and this provide the
infants with some pulmonary flow. Rarely infants with
pulmonary atresia and intact ventricular septum arrive
in the catheterization laboratory at several months of age,
having had a naturally persistent ductus arteriosus and/
or a previously created systemic to pulmonary artery
surgical shunt as palliation. As opposed to patients with
pulmonary atresia and an intact ventricular septum,
patients with pulmonary valve atresia and a ventricular
septal defect often have extensive systemic to pulmonary
collateral flow to the lungs and/or, occasionally, a large
persistent patent ductus arteriosus and, as a consequence,
these patients can survive to an older age with no prior
intervention.
A cardiac catheterization laboratory that has very high-
quality, biplane X-ray imaging and angulation capabilit-

ies for the X-ray tubes, is necessary for these perforation
procedures. Because of the precarious nature of these
infants, the extensive catheter manipulation required and
the potential for inadvertent occlusion of the ductus arter-
iosus during the procedure, these infants are intubated
and ventilated before starting the catheterization. Any
patient in whom a purposeful perforation is considered, is
type and cross-matched for one or two units of fresh
whole blood. If the replacement of blood becomes neces-
sary, the clotting factors as well as the oxygen carrying
capacity of whole blood are desirable. When an RF perfora-
tion is anticipated, a large “grounding plate” is placed
under the back of the patient at the very beginning of
the procedure. The grounding plate is attached to the RF
generator via the conductive cable as the patient is being
positioned on the catheterization table. Percutaneous
access to at least one femoral vein and a femoral artery is
established.
In the catheterization laboratory, the diagnosis is con-
firmed and the details of the right ventricular and pul-
monary artery anatomy are defined with selective biplane
right ventricular (outflow tract!) and aortic angiography.
The angiograms not only define the anatomy of the valve,
but demonstrate any right ventricular to coronary artery
fistulae in patients with pulmonary atresia with an intact
ventricular septum. In these patients, who usually do
have a good pulmonary artery in the presence of coronary
artery to RV fistulae, a right ventricular (RV) dependent
coronary circulation must be excluded before valve perfora-
tion is considered. The laser or RF techniques for valve

perforation also are used in patients with pulmonary atre-
sia and a ventricular septal defect, but only when there
is a well-developed main pulmonary artery and the RF
catheter/wire or laser wire, which is advanced from the
ventricle, can be advanced into the right ventricular
infundibulum and supported exactly at and against the
“valve” area. Whether using laser (which has only
recently become available in the US) or RF energy, the
techniques for pulmonary valve perforation are similar.
Radio-frequency perforation of the pulmonary
valve from the right ventricular approach in
patients with pulmonary atresia and intact
ventricular septum
The technique for pulmonary valve perforation using the
radio-frequency perforating system, which is designed
specifically for RF tissue perforation and is available
around the world (even in the United States), is described
in detail in this chapter. In addition to quality, biplane
angiograms in the right ventricular outflow tract, a
biplane angiogram of the main pulmonary artery is neces-
sary to visualize the valve annulus from the pulmonary
side. It is desirable to position a catheter in the main pul-
monary artery against the pulmonary side of the atretic
pulmonary valve. The catheter in the pulmonary artery is
introduced retrograde and passed into the pulmonary
artery through either the patent ductus arteriosus or a pre-
viously placed shunt. The ductus in the newborn, and par-
ticularly when the infant is on prostaglandins, is very
“mushy”, friable and often tortuous. Force never should
be used in crossing the ductus. If the ductus cannot be

crossed readily and almost inadvertently, a biplane aorto-
gram is performed in the descending aorta with the
injection of contrast immediately adjacent and/or slightly
distal to the aortic end of the ductus. This aortogram will
determine the exact course of the ductus and will define
the pulmonary artery/pulmonary valve more precisely.
Occasionally, even with the course of the ductus clearly
CHAPTER 31 Purposeful vascular perforations
846
defined, it is necessary first to cross the ductus with a
small, soft tipped, torque-controlled wire and then to
advance a multipurpose angiographic catheter over this
wire into the pulmonary artery. The catheter itself in the
pulmonary artery serves as a constant “target” during the
perforation from the right ventricle, and is used for
repeated angiography in the pulmonary artery during
the perforation.
If a catheter cannot be placed in the pulmonary artery,
at the very least there must be some capability of obtain-
ing repeated, good quality, biplane angiographic imaging of
the main pulmonary artery/pulmonary valve. When the
pulmonary artery cannot be entered reasonably, the pul-
monary artery imaging is obtained from the contrast
injected in the aorta and the flow through the ductus,
through a previous shunt or through collaterals. The tip of
the angiographic catheter for these injections in the aorta
is maintained immediately adjacent to or actually in the
origin of the vessel(s) providing the pulmonary flow in
order that the maximum contrast reaches the pulmonary
artery with each injection.

In the very rare instance where there is no demonstrable
systemic to pulmonary artery flow, the biplane imaging of
the pulmonary artery is obtained from a biplane pul-
monary vein wedge angiocardiogram. This technique is
satisfactory only if the main pulmonary artery can be visu-
alized adequately and repeatedly by this technique. The
use of repeated pulmonary vein wedge angiograms to
visualize the pulmonary arteries requires the presence of
an additional venous catheter situated in the vein wedge
position throughout the entire perforation procedure.
Biplane “freeze frame” images from the right ventricu-
lar angiogram, which demonstrate the right ventricular
outflow and the atretic pulmonary valve areas most satis-
factorily, are displayed as “road maps” for the subsequent
catheter positioning. After the pulmonary artery is visual-
ized angiographically and/or the retrograde catheter is
placed in the pulmonary artery adjacent to the pulmonary
valve, a 4- or 5-French, pre-shaped “right coronary” or
“cobra” guiding catheter is advanced into the right vent-
ricle and manipulated very carefully and precisely into the
right ventricular outflow tract (RVOT). This “guiding”
catheter is maneuvered until it is against the center of
the atretic valve in the right ventricular outflow tract. The
specific guiding catheter that is used depends upon the
size and, particularly, the right ventricular anatomy of
each individual patient. The angle at the tip of the parti-
cular guiding catheter, which is positioned in the RVOT,
should point the tip of the catheter directly at the other
catheter in the pulmonary artery and/or at the center of
the atretic valve in both the PA and lateral views as seen on

previous angiograms. Several different shaped guiding
catheters often must be tried in order to position the tip
of the guiding catheter pointing precisely in the exact
direction in both X-ray planes. As much time and effort is
taken as is necessary to achieve this precise positioning
before proceeding with the perforation. Any misalign-
ment in either plane very likely will result in perforation
out of the vascular channels and into the pericardium.
When the right ventricular catheter tip appears to be in the
ideal, proper position, a repeat, small, not too forceful,
hand injected, biplane angiocardiogram is performed
through this catheter. This angiogram demonstrates the
outflow tract and valve even better and illustrates clearly
any distortion to the area created by the catheter itself.
When the tip of the catheter is aligned precisely, these
images are displayed as the new “road map”.
If none of the available pre-shaped guiding catheters
can be positioned precisely in a direct line to, and against,
the valve, the guiding catheter is withdrawn. The tip of
the catheter is softened by immersing it in sterile, boiling
water. Once softened, a different, more appropriate curve
is formed at the tip. After reshaping the guiding catheter,
it is reintroduced and positioned properly against the valve.
Alternatively or in addition, a Mullins™ deflector wire
outside of the body is pre-shaped with very smooth curves
to correspond to the desired course from the right atrium,
to the right ventricle, to the right ventricular infundibu-
lum and finally against the atretic valve. All of the bends
on the wire are formed “tighter” than the existing curves
through the right heart to allow for some straightening of

the wire as it passes within the guiding catheter. When a
Mullins™ wire is used to hold the guiding catheter tip in
place, the guiding catheter must be at least one French size
larger in order to accommodate both the Mullins™ wire
and the perforating catheter side by side within the lumen
of the guiding catheter.
The Mullins™ wire is introduced into the pre-shaped
and previously positioned catheter through a Tuohy™/
side port back-bleed valve. The wire is advanced within
the catheter to a position just within the distal tip of the
catheter. The purpose of the Mullins™ wire is to redirect
and maintain the guiding catheter in its position against,
and pointing directly at, the center of the atretic valve
while the perforating wire/catheter is introduced. Again,
it is even more important that once this considerably
stiffer combination is positioned against the valve a re-
peat small biplane angiogram is performed through the
guiding catheter to demonstrate any further distortion
of the area.
Once the catheters are in place, preferably on both sides
of the valve, the RF perforation catheter, which has been
advanced through the BMC coaxial catheter while they
still are outside of the body, is introduced into the guiding
catheter that is pre-positioned in the right ventricular
outflow tract against the atretic valve. The perforating/
coaxial catheter is introduced through a wire back-bleed
flush valve or a Tuohy™ side port adaptor attached to the
CHAPTER 31 Purposeful vascular perforations
847
hub of the guiding catheter. Otherwise, with the sig-

nificantly larger lumen of the guiding catheter than the
diameter of the perforating/coaxial catheter and with
the high pressure in the right ventricle, there will be
significant bleeding into and externally out of the catheter
around the perforating catheter through the hub of the
guiding catheter. The guiding catheter is cleared of blood
by allowing it to bleed back passively through the
Tuohy™ valve and then placed on a continuous flush.
Any blood that remains in the catheter can clot, and potenti-
ally represents an embolus. The Tuohy™ type side port
adaptor also allows contrast injections through the side
port into the guiding catheter and around the perforat-
ing/coaxial catheters. If a Mullins™ wire is used to sup-
port the guiding catheter, a second Tuohy™ side port
valve is “piggy-backed” onto the angled side port of the
first Tuohy™. The R-F wire is passed through the first
Tuohy™ adaptor, which is attached directly to the guide
catheter. The Mullins™ wire is introduced through the
second Tuohy™ valve, which is attached to the angled
port of the first Tuohy™.
The perforating catheter, which is within the coaxial
catheter, is advanced to the tip of the guiding catheter. An
alternative technique in a small, very tight RVOT is to
introduce only the “perforation catheter” into the guiding
catheter without the covering coaxial catheter. The “perfor-
ating catheter” alone is more flexible and causes less dis-
placement of the precisely positioned guiding catheter. In
this circumstance, the coaxial catheter can be introduced
and advanced over the perforating catheter after the valve
has been perforated. Once the perforation catheter is in

the RVOT at the tip of the pre-positioned guiding catheter,
a repeat biplane angiogram of the RVOT/pulmonary
valve is performed to verify that the catheter in the RVOT
is still pointing exactly at the center of the pulmonary
valve. If the tip of the guiding catheter is displaced at all
away from the center of the atretic valve, the guiding
catheter with the contained perforating catheter is read-
justed by very slight torque and/or to-and-fro motion.
The biplane angiogram is repeated to verify the exact rela-
tionships after any readjustment.
When the tip of the guiding catheter is in position and
pointing in the precise direction, the proximal end of the
wire of the perforation catheter is attached to the genera-
tor with the BMC™ connecting cable. The tip of the RF
perforation catheter is advanced just barely out of the tip
of the guiding catheter and into the tissue of the atretic
valve. The flush on the side port of the catheter is stopped.
The generator is set for a one second duration and 5 watts
power. While continuously observing the tip of the per-
forating catheter on biplane stored fluoroscopy or slow
frame rate biplane cine angiography, a single burst of RF
energy is delivered while simultaneously holding, but not
advancing, the tip of the perforating catheter against the
valve. Usually this is sufficient for the perforating catheter
to pass through the valve into the pulmonary artery. If
not, the positioning of the guide and perforating catheter
tips is rechecked on biplane fluoroscopy. If the positions
are still not ideal, the perforating catheter is withdrawn
within the tip of the guiding catheter while the guiding
catheter repositioned. When the guiding catheter is in the

exact position, the perforating catheter tip is re-advanced
into the valve tissues. When both catheters are in the pre-
cise position, the RF energy is reapplied to the perforating
catheter while again holding the tip of the perforating
catheter against the valve without pushing forcefully. This
process is repeated until the perforating catheter advances
through the valve “plate” and into a “free” position
within the pulmonary artery just beyond the valve while
no energy is being applied. Any advancing of the tip of the
perforating catheter within the pulmonary artery must be
with no energy applied, as any RF energy applied to the
tip will allow the tip to perforate any structure (wall!) in its
vicinity! The exact position of the tip of the perforating
catheter in the pulmonary artery is verified with a biplane
pulmonary artery angiogram before proceeding further.
Once the tip of the perforating catheter has entered
the pulmonary artery freely, the perforating catheter is
advanced with no energy applied as far as possible distally
into the branch pulmonary artery or through the patent
ductus into the descending aorta. When the perforating
catheter is well into the pulmonary artery or the descend-
ing aorta, the coaxial catheter is advanced over the per-
forating catheter to the tip of the perforating catheter.
The subsequent maneuvers depend upon the associated
anatomy and the position of the perforating/coaxial cath-
eters after they are advanced following the perforation.
If the perforating/coaxial catheters pass into the
descending aorta, together they are snared there with a
snare catheter, which is introduced retrograde from the
femoral artery. With traction held on the perforating/

coaxial catheter with the snare catheter, the guide catheter
is removed from the femoral vein over the combined per-
forating/coaxial catheter and replaced with a 2–4 mm
diameter low-profile dilation balloon. When passed over
the combined perforating/coaxial catheter, this requires a
balloon catheter with a catheter lumen that will accommo-
date a 0.035″ guide wire. An alternative technique, when
the perforating/coaxial catheters pass into the descending
aorta, is to withdraw the perforating catheter from the
coaxial catheter and exchange it for a stiff 0.014″ or 0.016″
exchange length “coronary” guide wire. This exchange of
the perforating catheter for wire is performed while the
guiding catheter still is in position in the RVOT and the
snare is around and gripping the coaxial catheter loosely.
If the coaxial catheter begins to withdraw or buckle while
the new, stiffer wire is introduced into it, the distal end of
the coaxial catheter is grasped firmly with the snare in the
CHAPTER 31 Purposeful vascular perforations
848
descending aorta. This supports the passage of the stiffer
exchange guide wire through the relatively flimsy coaxial
catheter as it passes through the tortuous course from the
inferior vena cava through the right heart, pulmonary
artery and ductus to the descending aorta.
Once the stiffer, smaller exchange wire emerges from
the tip of the coaxial catheter in the descending aorta, this
wire alone is grasped securely with the retrograde snare.
The snared distal end is withdrawn into the femoral area
or even out through the femoral artery sheath. Either way,
a very secure “through-and-through” or “rail” wire sys-

tem is created. The coaxial and the guiding catheters are
removed over the fixed wire. The through-and-through
wire allows simultaneous strong traction from the two
ends of the wire which, in turn, allows a very forceful for-
ward push on the balloon dilation catheter without the
catheter and/or the wire buckling in the right ventricle as
the balloon passes through the tight valve. With the trac-
tion applied at both ends of the exchange wire, the very
low profile 2–4 mm diameter coronary balloon is passed
over the wire and advanced through the “plate” of the pul-
monary valve to initiate a sequential dilation of the valve.
After the dilation with the initial, small coronary bal-
loon, the balloon is exchanged over the same “rail” wire
for a larger dilating balloon. Dilation balloons of progres-
sively increasing size are used until a balloon that is
appropriate in diameter for a single balloon pulmonary
valve dilation of the particular valve annulus can be
introduced.
An alternative technique, which can be used when there
is a patent ductus that can be traversed easily, is to posi-
tion a retrograde snare catheter instead of an angiographic
catheter in the pulmonary artery before and during the
actual perforation of the valve. Instead of maneuvering
the original retrograde angiographic catheter into the pul-
monary artery, a 4-French Microvena™ snare catheter is
passed retrograde through the patent ductus into the pul-
monary artery. The standard snare catheter often is easier
and safer to position in the pulmonary artery than an
angiographic catheter. A floppy, soft tipped, torque-
controlled wire is advanced totally atraumatically from

the aorta through the ductus and against the atretic pul-
monary valve. The end-hole snare catheter then advances
easily over this previously positioned wire. A small 5 or
10 mm diameter snare (depending upon the diameter of the
pulmonary annulus) is opened in the annulus of the pul-
monary valve on the main pulmonary artery side of the
valve. The properly sized snare aligns perpendicular to
the long axis of the pulmonary artery, around and outlin-
ing the circumference of the valve annulus. The “circle” of
the snare serves as a very clear “target” for the perforation
with the RF catheter. As soon as the RF catheter with the
coaxial catheter has advanced through the valve tissue,
the perforating catheter also will be through the loop of
the snare in the pulmonary artery! If there is any difficulty
passing the coaxial catheter along with the RF perforating
catheter through the new “puncture”, the RF perforating
catheter alone can be advanced through the valve and
grasped with the snare. Once the RF catheter is snared
securely, traction is placed on the RF catheter and the
coaxial catheter is drawn into the descending aorta with
the snare. Either the coaxial catheter or a balloon dilation
catheter is advanced over this fixed RF catheter and
through the valve as described previously.
If a retrograde angiographic catheter is positioned from
the aorta, through the ductus and into the pulmonary
artery, and the perforating catheter is not manipulated on
its own from the pulmonary artery through the ductus
into the descending aorta after the perforation of the
valve, the retrograde angiographic catheter in the pul-
monary artery is replaced with a snare catheter. Again,

because of the frequent tortuosity and “mushy” nature of
the ductus in these patients, the ductus is crossed with a
very soft tipped torque wire and the snare catheter is
advanced through the ductus over this wire. Once the
snare is open in the main pulmonary artery, the perforat-
ing wire/catheter almost automatically will be through the
loop of the snare! The perforating catheter is grasped with
the snare in the pulmonary artery and withdrawn into the
descending aorta through the ductus, as described above.
The worst-case scenario is when there is no patent duc-
tus, or when present, the ductus cannot be crossed from
either direction. In that case, the perforating catheter,
immediately after perforating the valve, is manipulated as
far as possible into a distal right or left branch pulmonary
artery. With the guiding catheter still positioned in the
RVOT and forced against the pulmonary valve over the
perforating catheter, the coaxial catheter is advanced over
the perforating catheter, through the valve and to the tip
of the perforating catheter/wire. With the guiding and
coaxial catheters both fixed in these positions, the perforat-
ing catheter/wire is withdrawn carefully and replaced
with a 0.014–0.018″ (depending upon which coronary
dilation balloons are available) stiff, exchange length,
“coronary” guide wire. The guide wire is advanced out of
the tip of the coaxial catheter until the long floppy tip of
the guide wire is “balled up” completely in a distal pul-
monary artery branch. This “wadding up” of the floppy
tip is essential in order to ensure that the stiff portion of the
guide wire will be across the valve and well out into the
branch pulmonary artery.

Once the stiffer exchange guide wire is in this secure
position in the distal pulmonary artery, the guide catheter
and then the coaxial catheter are removed over the guide
wire and replaced with a very low-profile, 2–3 mm dia-
meter, “coronary” dilation balloon. Occasionally, even the
very low-profile balloon will not follow over the wire
through the thick valve tissue. In that circumstance, the
CHAPTER 31 Purposeful vascular perforations
849
balloon is withdrawn over the wire and replaced with a
larger guiding catheter that can accommodate the low-
profile balloon. The guiding catheter is manipulated very
gingerly over the wire through the right ventricle and up
against the pulmonary valve. With the guiding catheter as
an additional support, the low-profile balloon is passed
over the wire, through the guiding catheter and through
the valve.
The sequential dilation of the valve is started over this
wire. Once the “waist” in the initial balloon has been elim-
inated, the balloon is removed over the wire and replaced
with a slightly larger, 3–5 mm balloon. The balloons are
replaced sequentially until a balloon is introduced that is
appropriate in size for a single-balloon valve dilation,
according to the annulus diameter of the pulmonary
valve. Occasionally, the initial smaller wire must be ex-
changed for a larger and stiffer wire to support the larger
balloons, which will not pass through the guiding catheter.
Laser technique for perforation of the
pulmonary valve in pulmonary atresia with
intact ventricular septumCfrom the right

ventricular approach
Excimer Laser™ energy has been used for the perforation
of the atretic pulmonary valve outside of the United States
for over a decade, but the lack of a small laser catheter
approved by the US FDA precluded its use in the US until
recently
6,7
. The Excimer Laser uses ultraviolet light with a
wavelength of 308 nm to ablate tissues in the path of the
laser light. The laser energy is generated with a VCX-300
Excimer Laser System (Spectranetics, Colorado Springs,
CO), which often is available in an interventional catheter-
ization laboratory for laser lead extractions. The recent
approval by the US FDA of the Point 9™ Extreme Excimer
Laser catheter (Spectranetics, Colorado Springs, CO) for
the treatment of total occlusions of peripheral and coron-
ary arteries in humans has made this very small laser
catheter available for use for selected congenital lesions in
the US. The Point 9™ Laser catheter is a 0.9 mm diameter,
fairly flexible catheter, consisting of multiple layers of
optical fiber strands, which run the length of the catheter
and through which the laser energy is delivered. The bun-
dled fibers surround an open lumen, which accepts a
0.014″ wire.
The Point 9™ Laser catheter is advanced to the atretic
pulmonary valve through a pre-positioned guiding
catheter that has a lumen of at least 1 mm (3-French) dia-
meter. The specific guide catheter that is optimal for the
particular patient varies with the size and anatomy of each
individual patient. The guide catheter should have pre-

formed curves at the distal end that correspond to the
course from the inflow to the outflow of the particular
right ventricle. The guide catheter is maneuvered into the
narrow outflow tract of the right ventricle to a position as
close to the center of the “plate” of the atretic valve as pos-
sible. The Point 9™ Laser catheter is delivered over a stiff
0.014″ guide wire with a fine floppy tip as well as through
the guiding catheter. The floppy tip of the guide wire
extends beyond the tip of the laser catheter and often
loops back on itself in the right ventricular outflow tract as
the Point 9™ catheter is maneuvered to the atretic valve.
As the tip of the Point 9™ catheter approaches the valve,
the wire is withdrawn into the laser catheter.
Once positioned against the “plate” of the valve, 45
microJoules, at 15 kV and 25 Hz, of laser energy are deliv-
ered through the catheter in one second bursts. The tip of
the catheter usually passes through the atretic tissue with
1 or 2 bursts of energy with each burst penetrating
approximately 100 microns. If there is any forward push
applied to the catheter during the delivery of the energy,
the laser catheter will continue through any tissue in front
of it including out of the vascular space! Once the tip of the
laser catheter has advanced through the atretic tissue into
the main pulmonary artery, the guide wire is advanced
out of the catheter and preferably into the descending
aorta through the ductus. Once the wire is successfully
through the “valve” the remainder of the procedure is
identical to the procedure using RF energy.
Laser energy has the disadvantages of requiring a large
and expensive generator, which may not be available in

all congenital heart catheterization laboratories, and the
potential of retinal injury to surrounding personnel from
the “scatter” of the high-intensity ultraviolet light. Until
more experience shows a distinct advantage of laser over
RF energy, the RF systems now appear preferable for pul-
monary valve perforation in congenital heart lesions.
Technique for perforation of the pulmonary
valve in pulmonary atresia with intact
ventricular septum retrograde through the
patent ductus from the pulmonary arterial
approach
When there is significant difficulty or even the absolute
impossibility of positioning the guiding catheter properly
in the right ventricular outflow tract (RVOT) and/or there
also is an easily crossed patent ductus, the perforation of
the pulmonary valve in patients with pulmonary atresia
with intact ventricular septum can be performed from the
pulmonary artery side of the valve using a retrograde
approach through the ductus
9
. Before the availability of
the current and safer “burning” techniques to perform the
perforation, stiff wires in association with the use of
“brute force” had been used to push through the atretic
pulmonary valves into the right ventricular outflow tract
from the pulmonary artery approach. Because the “tar-
get” area of the right ventricular outflow tract is small and
CHAPTER 31 Purposeful vascular perforations
850
very narrow, the strong force (push), which necessarily

had to be applied to the retrograde catheter, could easily
displace the direction of the catheter tip with the result
that this technique had a very high likelihood of perfora-
tion into the pericardium instead of into the right vent-
ricle. The retrograde perforation of the pulmonary valve is
far more reasonable with the availability of RF wires and
RF energy for the perforation.
When the guiding catheter cannot be positioned in the
RVOT with the tip of the catheter directed precisely at the
valve in both X-ray planes, the retrograde approach for
perforating the valve should be considered. The retro-
grade perforation still requires that a catheter is posi-
tioned in the RVOT for the purpose of performing
selective biplane angiography even if the catheter cannot
be directed precisely at the valve. The RVOT must be visu-
alized very clearly, precisely and repeatedly with biplane
imaging during a retrograde perforation. The right ven-
tricular outflow tract usually tapers to a very fine tip or
point just below the atretic valve. As a consequence, the
RVOT presents a much smaller “target” when perforating
from the pulmonary artery toward the RVOT.
With a prograde venous catheter positioned in the
RVOT, a Swan™ floating balloon catheter (Arrow
International Inc., Reading, PA) is introduced retrograde
into the femoral artery and advanced retrograde through
the patent ductus and into the pulmonary artery with or
without a pre-positioned floppy tipped wire through the
ductus. With some retrograde “push” applied to the
Swan™ catheter, the balloon is inflated in the main pul-
monary artery directly in the pulmonary valve annulus

and against the atretic valve. The inflated balloon in con-
junction with the usual course through the ductus usually
orients the lumen of the Swan™ catheter parallel to the
long axis of the pulmonary artery and also often points the
end hole of the Swan™ catheter directly toward the blind
RVOT. The precise direction of the tip of the catheter that
is seated in the atretic pulmonary valve is changed in
order to point the lumen exactly toward the RVOT by
varying the amount of “push” and/or torque on the
Swan™ catheter. When the tip of the Swan™ catheter
is pointing in the exact direction, the RF perforating
wire/catheter is advanced through the Swan™ catheter to
its tip, until the RF perforating wire is positioned against
the atretic pulmonary valve. The relative relationships of
the tip of the Swan™ and the RF catheter to the RVOT are
verified with a small selective biplane angiocardiogram in
the RVOT. The angle of the Swan™ catheter/perforating
catheter together also can be adjusted slightly by minimal
to-and-fro motion on the shaft of the Swan™ catheter
while the perforating catheter is passing through it. When
the tip of the Swan™ catheter/perforating wire/catheter
is “aimed” exactly at the small, blind RVOT, the RF perfo-
rating wire/catheter is advanced until the tip of the wire
is embedded in the “valve” tissue. The position is
rechecked with a repeat small selective biplane angio-
cardiogram in the RVOT, and adjustments to align the
Swan™ catheter again are made as necessary. Since the
RVOT “target” is so small, perforation from the retro-
grade approach is not attempted unless the perforating
wire/catheter tip and the narrow RVOT are aligned

exactly. When the RF perforating wire/catheter is point-
ing precisely at the RVOT, and only then, is the RF energy
applied while the perforating wire/catheter is held in the
valve adjacent to the small RVOT.
Instead of the retrograde Swan™ catheter, a pre-
formed, end-hole only catheter can be used for the retro-
grade perforation. The tip of a non-Swan™ type catheter
is harder to keep exactly aligned in the center of the pul-
monary valve on the pulmonary side of the atretic valve.
This can be accomplished eventually with patience and
often multiple exchanges of catheters with different curves
at the tip.
Once the perforation wire/catheter is through the
atretic pulmonary valve and free in the RVOT, a 5 or 10
mm Microvena™ snare wire is introduced through the
catheter in the RVOT and the tip of the perforating
wire/catheter snared in that location. Occasionally, in
very small infants, only the perforating wire without
the covering “coaxial catheter” can be passed through the
lumen of a small Swan™ catheter. In that circumstance the
RF perforating wire alone is advanced through the valve
after the perforation and grasped with the snare in the
RVOT. If possible, the snared wire and/or catheter is
withdrawn through the ventricle and tricuspid valve and
exteriorized through the femoral vein but, once the per-
forating wire/catheter is held securely, the Swan™ bal-
loon catheter is withdrawn from the femoral artery over
the proximal end of the perforating wire/catheter and is
replaced over the perforating wire/catheter with either
the coaxial catheter or a 4-French, multipurpose, end-hole

catheter with a tapered tip. This step is unnecessary if the
original retrograde catheter for the perforation was a non-
Swan™ end-hole catheter. With traction held on the tip of
the perforating wire/catheter in the right ventricle with
the snare catheter, the coaxial or tapered end-hole catheter
is advanced over the perforating wire/catheter, retro-
grade through the ductus and then push–pulled across
the pulmonary valve using traction at both ends of the
perforating wire/catheter. Once the end-hole, retrograde
catheter is in the RV, the snare around the wire is loosened
enough to allow the catheter to pass over the perforating
wire/catheter and through the snare. The perforating
wire/catheter is withdrawn out of the femoral artery
catheter and replaced with a floppy tipped, exchange
length wire. If necessary during this exchange, the snare is
tightened over the end-hole catheter that is passing
through the pulmonary valve to hold the catheter in place.
CHAPTER 31 Purposeful vascular perforations
851
Snaring the distal end of the catheter in the RVOT sup-
ports the retrograde passage of the stiffer exchange wire
through the ductus, the pulmonary artery, the perforated
pulmonary valve and into the right ventricle.
Once the tip of the exchange wire is through the valve
and in the right ventricle, the floppy tip of the wire is
grasped in the right ventricle with the snare and the com-
bination wire and catheter is pulled back, and very care-
fully through the tricuspid valve. Extra care is taken not to
catch on the tricuspid valve structures and/or to pull too
vigorously through the tricuspid valve. If the combination

cannot be pulled easily through the valve, the snare is
opened, releasing the grip on the wire. The retrograde
catheter is withdrawn partially off the tip of the wire and
back toward the pulmonary valve. The floppy tip of the
wire alone is re-grasped with the snare, which is still
within the right ventricle, and a repeat attempt is made
to withdraw the snare catheter with the grasped retro-
grade wire back through the tricuspid valve and into the
right atrium.
If the snare/wire still catches on the tricuspid valve, the
wire is released from the snare and the snare loop with-
drawn completely into the snare catheter. The snare
catheter is withdrawn into the right atrium. A very careful
attempt then is made at manipulating the retrograde
catheter/wire, which is passing through the pulmonary
valve, through the tricuspid valve and back into the right
atrium. The catheter may “bind” in the thick, tight pul-
monary valve structure and care must be taken to prevent
it from buckling and pulling out of the recently perforated
valve! If the wire is maneuvered to the right atrium, the tip
of the wire is re-snared in the right atrium.
If the wire/catheter cannot be maneuvered back to the
right atrium, the snare catheter is manipulated through a
different area of the tricuspid valve and back into the right
ventricle, and/or an end-hole Swan™ balloon catheter is
introduced from the femoral vein and used to cross the tri-
cuspid valve into the right ventricle from the right atrium
and then the Swan™ catheter is used as the snare delivery
catheter. When an inflated Swan™ balloon advances
across the small tricuspid valve, there is a better chance

that the balloon will pass through the largest orifice of the
tricuspid valve, and the snare wire can be used through
the Swan™ catheter. The small size of the hypoplastic tri-
cuspid valve and/or the tricuspid valve regurgitation,
however, may prevent a Swan™ balloon catheter from
“floating” into the right ventricle.
Once the floppy tip of the wire is grasped with the snare
in the right ventricle with either catheter that has passed
through a different area of the tricuspid valve, the retro-
grade exchange wire is pulled carefully into the right
atrium. From the right atrium, the exchange wire is
exteriorized through the femoral vein sheath, creating a
through-and-through femoral vein to femoral artery wire.
The remainder of the procedure is the same as when the
perforation of the atretic pulmonary valve was from
the prograde approach. Sequential dilations of the pul-
monary valve are accomplished introducing the dilation
balloons from the femoral vein over the through-and-
through wire, as described previously.
Once the pulmonary valve is open and there is pro-
grade access to the pulmonary artery, the necessity for
further palliation of the patient in the catheterization lab-
oratory during the same procedure is determined in the
laboratory at that time. Following a perforation and dila-
tion of the pulmonary valve in patients with pulmonary
atresia and intact ventricular septum, there almost always
is a question about the adequacy of the right ventricular
volume and the need for a systemic to pulmonary shunt
and/or an atrial septostomy. These patients usually were
dependent upon the patent ductus for most of the pul-

monary flow and all have an existing patent foramen
ovale/atrial septal defect; however, usually one or both of
these sources of blood flow is/are inadequate.
Once the infant stabilizes after the valve perforation/
dilation and there is adequate pulmonary flow, then fur-
ther clinical assessment determines the need for an atrial
septostomy. In the presence of a very small right ventricu-
lar cavity and/or persistent very high right ventricular
end diastolic and/or systolic pressures, the right ventricle
often is not capable of accommodating an adequate dias-
tolic volume from the systemic venous blood return. The
resultant small right ventricular systolic volume then will
be inadequate to provide enough forward blood flow
through the lungs to the left heart to sustain an adequate
systemic output. In the absence of an adequate opening or
“vent” at the atrial level, the systemic venous blood pools
in the systemic venous vascular bed and right atrium, the
right atrium becomes massively dilated with the systemic
venous return, and the cardiac output remains low. When
this occurs acutely in the catheterization laboratory, a bal-
loon or blade and balloon atrial septostomy is performed
during the same catheterization.
At the same time, when the patient does have even a
marginal systemic cardiac output without significant
right atrial/hepatic congestion after the pulmonary valve
perforation, atrial septostomy is not performed during
the initial catheterization. Some elevation of the right
atrial pressure may augment the right ventricular filling
of these small ventricles. When an atrial septostomy is
performed, it lowers the right atrial pressure, and in turn,

may compromise right ventricular filling! By eliminating
this extra filling pressure and volume, the potential
growth of the right ventricle also may be compromised.
The balloon or a blade/balloon atrial septostomy always
can be performed hours, days or weeks later if the sys-
temic output decreases and/or the right atrium and liver
become distended.
CHAPTER 31 Purposeful vascular perforations
852
In addition to the question of adequate return of the sys-
temic venous blood to the systemic output, the adequacy
of the net pulmonary flow is assessed before the infant
leaves the catheterization laboratory. After the valve has
been opened successfully, the adequacy of the forward
flow through the opened valve, the effect of the pul-
monary regurgitation on the net forward flow and how
much of the pulmonary flow still is from the ductus are
determined from the angiograms. If a catheter interven-
tion to increase the pulmonary flow is even considered, a
catheter is advanced either prograde or retrograde across
the ductus, prostaglandin is stopped and the infant
observed in the laboratory for 30 minutes or longer.
When the ductus remains patent after the prostaglandin
has been stopped, no further intervention is considered
at that time. When the ductus patency and flow are
prostaglandin dependent, the 30 minutes usually are
sufficient time for the prostaglandin effect to wear off and
for the ductus to close functionally. When the net pro-
grade pulmonary flow is insufficient after the ductus
closes, the infant will become significantly desaturated

and/or hypoxic. The choices at that time are to restart
prostaglandin and terminate the case with plans for a sub-
sequent surgical shunt or to consider the implant of a stent
in the patent ductus arteriosus as a means of establishing a
more permanent systemic to pulmonary artery “shunt”.
With the newer, pre-mounted, flexible, small stents this is
a much more viable option.
The technical details of the atrial septostomy proced-
ures are discussed in Chapters 13 and 14 and the tech-
nique for stenting the patent ductus are discussed in
Chapter 25. Until far more definitive data are available
about which ventricles grow after the pulmonary valve is
open and which patients have adequate pulmonary flow
with the ductus closed, the decisions for further catheter
intervention are “on-the-spot”, somewhat arbitrary judg-
ment decisions in the catheterization laboratory during
each individual case, but usually some type of an aug-
mented systemic to pulmonary shunt, if not an atrial
septostomy, is required.
Perforation of the pulmonary valve in patients
with pulmonary atresia and an associated
ventricular septal defect
Hausdorf and associates extended the use of radio-
frequency perforation of the plate-like pulmonary valve to
perforation of the “muscular tract” between the right vent-
ricular outflow tract (RVOT) and the main pulmonary
artery for the attempted palliation of ten patients with
pulmonary atresia and a ventricular septal defect
10
. The

distance between the RVOT and the main pulmonary
artery varied between 1.2 and 12 mm. Except for two new-
borns, their ten patients were much larger and older
patients, some even years past the newborn period. All
of the patients except the two newborns had either
significant systemic to pulmonary collaterals or a surg-
ically placed systemic to pulmonary artery shunt as their
source of pulmonary artery blood flow.
Because patients with pulmonary atresia and a vent-
ricular septal defect have either a very tortuous patent duc-
tus arteriosus and/or present at a later age with no ductus
arteriosus, perforation of the atretic pulmonary valve
without the use of a through-and-through wire usually is
necessary in these patients. Before being considered for
valve perforation, patients with pulmonary atresia and a
ventricular septal defect should have an adequate dia-
meter, main pulmonary artery documented angiographi-
cally. This angiographic anatomy is obtained from biplane
angiograms in the aorta adjacent to the “source” vessels
for the pulmonary flow, from selective biplane injections
into aortopulmonary collaterals, or even from biplane
pulmonary vein wedge angiograms. The “indirect” angio-
graphic pictures of the main pulmonary artery are stored
as “road maps”.
After the anatomy of the main pulmonary artery and its
precise location are identified, an end-hole “guiding”
catheter is pre-shaped to conform to the course from the
right atrium to the right ventricular outflow tract (RVOT).
This guiding catheter must be manipulated into the right
ventricular infundibulum with the tip directed exactly in

the direction of the pulmonary artery as visualized on the
biplane “road maps”. It may be possible to advance the tip
of the guiding catheter only to the proximal, or inflow end,
of the infundibulum. The RF perforating catheter is
advanced through and out of the tip of the guiding
catheter. With occasional good fortune or even luck, the
very thin perforating catheter passes through the in-
fundibulum until it is close to, or against, the area of a tiny
atretic valve structure. When this occurs, the course
through the traversed infundibulum tends to align the tip
of the perforating catheter more directly at the center of
the stump of the atretic main pulmonary artery segment.
The position is verified with biplane angiography, inject-
ing through the guiding or a separate venous catheter.
With the tip of the RF wire pushed against the atretic tis-
sues and in the precise direction of the pulmonary artery
as visualized in both PA and LAT X-ray planes, RF energy
is applied for several seconds. The perforating wire is
advanced in short steps toward the pulmonary artery
between bursts of RF energy, and the RF energy re-applied.
Very rarely, the RF perforating catheter is advanced in
very short distances during the application of the energy
while observing the course through the tissue very care-
fully. Once through the atretic tissues and into the pul-
monary artery, the perforating catheter is advanced
without energy applied to it into a distal pulmonary
artery branch and distally as far as is possible. The tract
CHAPTER 31 Purposeful vascular perforations
853
of the RF wire from the RVOT to the pulmonary artery is

examined carefully to verify that the wire is in the center
of the muscular tract in both planes. The procedure then
is similar to a pulmonary atresia with intact ventricular
septum where the ductus arteriosus was not traversed,
although maneuvering through the tract of RVOT tissue
usually is even more difficult.
More often, in patients with pulmonary atresia and a
ventricular septal defect, the perforating catheter/wire
cannot be advanced beyond the tip of the guiding catheter
positioned at the proximal end of the infundibulum. If, on
biplane imaging, the guiding catheter is pointing directly
toward, although still some distance away from, a pul-
monary artery of an adequate diameter, the RF energy is
utilized to perforate through the infundibulum, to the
area of the “valve”, and then through the remaining tis-
sues into the pulmonary artery. This is performed with
very short bursts of RF energy and small advances of the
perforating catheter/wire between the applications of
energy. Between each advance, the position of the perfo-
rating catheter/wire and its orientation toward the valve
is rechecked with small, selective, hand injected, biplane
angiograms in both planes of imaging and injecting
through the guiding catheter or an adjacent venous
catheter in the outflow tract. The attempted perforation is
continued only as long as the perforating catheter/wire
stays on a direct course to the valve and the patient
exhibits no hemodynamic deterioration during the proced-
ure. Once the valve is crossed, the wire positioning and
dilation procedure are the same as previously described
for pulmonary atresia with intact ventricular septum with-

out a patent ductus. After the communication has been
established and the tract partially dilated, usually the tract
is maintained open with an intravascular stent.
Purposeful perforation of other vascular
structures or total vessel obstructions
Pediatric and congenital catheterization laboratory inter-
ventionalists who are very comfortable with transseptal
atrial puncture have extended the needle puncture tech-
nique of the transseptal procedure to the perforation of
multiple other structures, some for diagnostic purposes
and others in order to create permanent openings.
Puncture of surgically created interatrial baffles and/or
patches were the logical extension of the standard atrial
transseptal procedure. The venous approach usually is
used, however, for baffles and some patches, the needle
tip is positioned and directed very differently according to
the orientation of the patch or baffle
11
. Usually, more force
must be applied to the needle/transseptal set to penetrate
the thicker structures of a baffle/patch. The extra force
required increases the risk of the perforating needle
continuing forward beyond or on the “other side” of the
desired perforation into unwanted structures. The use of
an RF perforating catheter through a special transseptal
sheath/dilator set (Baylis Medical Co. Inc., Montreal,
Canada) eliminates the extra force (and in turn, extra
risk) necessary for these “transseptal” perforations. Un-
fortunately, RF perforation is not applicable through
patches and/or baffles that are made of synthetic (non-

tissue) materials.
In addition to the native atrial septum, patches in
the interatrial septum and baffles within the atria, the
transseptal needle and set are used to perforate and, in
turn, re-cannulate totally occluded vascular channels. The
transseptal needle requires a relatively straight line of
access or “straight shot” from the site of catheter introduc-
tion to the site being “re-cannulated” in order to transmit
the forward force from outside of the body, along the
needle and to the needle tip at the puncture site. When
forward force is applied within any significant, “non-
contained” curve in the course of the needle, the force
applied to the proximal needle causes the needle to “bow”
proximally and, in turn, the forward force is dissipated
into the curve rather than being delivered to the tip, and
the direction of the tip is changed significantly.
Even with this limitation, the transseptal needle has
been used successfully to puncture and rebuild long (5–
6 cm!) total obstructions in multiple different vascular chan-
nels (see Chapter 24, “Venous Stents”). These include total
obstructions in the native superior vena cava, the superior
limbs of intracardiac venous baffles, totally disconnected
right pulmonary arteries in postoperative “hemi-Fontan”
patients
12
, aortic coarctations with a discrete membranous
interruption, and all varieties of ilio-femoral/IVC total
venous obstruction
13
. The access for these punctures is

from the femoral, jugular or hepatic vein approach
depending upon the vessel, the location and orientation of
the obstruction and which approach provides the straight-
est route to and through the obstruction.
The availability of the radio-frequency perforating sys-
tems has extended the possible sites of vascular obstruc-
tion that might be perforated and reconstituted. With the
flexibility of the RF perforating catheter (wire), and since
little or no forward “force” is required for the perforation,
RF perforating catheters can traverse a very tortuous
route to the site to be perforated. Perforating RF catheters
(wires) readily pass around relatively acute curves and, in
turn, can approach the area to be punctured from sharp or
acute angles. The use of the RF wire allows the perforation
of the atrial septum
14
, atrial baffles and other vascular
structures from a variety of venous access sites. As long as
contact is maintained against the surface to be punctured
by the RF perforating catheter, the RF energy will pene-
trate native tissues.
The usual interatrial septum is aligned parallel to, and
even away from, a catheter that is introduced from the
CHAPTER 31 Purposeful vascular perforations
854
superior vena cava (SVC). The curved tip of a transseptal
needle, even with an added curve formed on the shaft of
the needle, often cannot engage the interatrial septum
from a superior vena cava approach. Even when a very
curved needle does “catch” on the septum from the SVC

approach, the straight direction necessary to apply for-
ward force to the tip of the needle for perforation is not
possible. Using the RF perforation system, a pre-formed
guiding catheter or special RF transseptal set with nearly a
right angle curve at the tip can be introduced from the
jugular vein/superior vena cava approach and then the
catheter/set is rotated until the pre-formed curve at
the tip of the catheter positions the tip against, and nearly
perpendicular to, the interatrial septum. The RF perfora-
tion wire and coaxial catheter are advanced through the
guiding catheter and just against the septum. The RF
energy is applied as the RF perforation wire and coaxial
catheter are advanced through the septum with virtually
no force. Once the RF perforation wire and catheter are in
the left atrium, they are exchanged through the coaxial
catheter for a stiffer exchange length wire, and then what-
ever sheath/catheter system is desired.
Excimer Laser™ energy also has been used to perforate
other totally occluded structures in addition to atretic pul-
monary valves. In our institution, a very scarred-in dis-
tance of 5–6 mm between the side of the main/right
pulmonary artery and the totally disconnected stump of
the proximal left pulmonary artery was perforated suc-
cessfully with a laser catheter on a compassionate use
basis. The left pulmonary artery had been totally discon-
nected as a consequence of multiple prior surgeries. Both
of the vessels were encased in dense scar tissue. The con-
vex, cephalad and left side of the main/right pulmonary
artery just at the origin of the right pulmonary artery was
parallel to the side of the disconnected proximal LPA. The

perforation was performed through the convex, left side
of the main/right pulmonary artery, through scar tissue
surrounding the walls of the two adjacent vessels and
through the side and into the discontinuous LPA. After
perforation, the tract between the two vessels could be
dilated and an intravascular stent implanted in the chan-
nel to maintain the patency.
Laser energy has the advantage of greater perforating
capability. The RF energy for perforation is more control-
lable and is readily available in more catheterization
laboratories. The RF techniques now are used for almost
all of the applications for which laser perforations previ-
ously were utilized in previous studies even outside of the
United States.
The dedicated RF perforating system with the RF
perforating catheter has extended the type and number
of obstructed structures that can be opened or recom-
municated in the catheterization laboratory. RF perfora-
tion already has been used to reconnect a chronic total
left pulmonary artery obstruction
15
. In that patient, the
distal left pulmonary artery became totally isolated four
years earlier following an even earlier surgical patch
repair of the artery. The intravascular course to the
obstruction was the usual intracardiac tortuous course to
a left pulmonary artery. The distal pulmonary artery
beyond the obstruction was identified by pulmonary vein
wedge angiography. The obstruction could not be crossed
with any type of mechanical wire probing but was crossed

with a 2-French RF catheter using 11 watts of power deliv-
ered for 11 seconds.
When a totally obstructed vascular structure is encoun-
tered, maximum information should be obtained about
the length and course of the obstruction as well as the sta-
tus and size of the chamber or vessel lumen that is open at
both ends or on both sides of the obstruction. This infor-
mation is acquired from magnetic resonance imaging and
with biplane angiography before a perforation through
the obstruction is considered. In branch pulmonary artery
occlusions, information about the distal vessel is obtained
angiographically from the flow into the obstructed vessel
through shunts, systemic to pulmonary collaterals and/or
from pulmonary vein wedge angiograms. The length and
the course of the obstruction are “road mapped” and
placed in the longest axis of the biplane review screens.
A guiding catheter is chosen which, when placed in the
proximal end of the obstruction, aligns the tip of the
catheter parallel to the long axis of the two discontinuous
segments of the obstructed vessel. Changing the guiding
catheter, changing the curve of the guiding catheter
and/or the use of a Mullins™ wire within the guiding
catheter adjacent to the RF perforating catheter are used to
accomplish the precise direction of the guiding catheter.
The RF perforating catheter is passed through the tip of
the pre-positioned guiding catheter and engaged in the
proximal end of the obstruction. The guiding catheter, the
tip of the RF perforating catheter and the entire length of
the obstruction are visualized in both X-ray planes with
biplane fluoroscopy or biplane cine imaging while the RF

energy is delivered to the RF catheter. The tip of the RF
perforating catheter is advanced only between the appli-
cations of RF energy, and must follow a precise “road
mapped” course through the obstructed area and toward
the area of the previously visualized “lumen” at the other
end of the obstruction.
If the RF perforating catheter begins to detour away
from this precise course at all, the RF energy is not applied
and the RF catheter is withdrawn. The “tract” created is
visualized with a biplane angiogram performed with a
very small contrast injection through the guiding catheter.
This angiogram demonstrates any new and/or erroneous
“tract” that has been created with the RF catheter. If the
tract is precisely in line between the two portions of the
vessel, the RF catheter is advanced back into the tract and
CHAPTER 31 Purposeful vascular perforations
855
the energy delivery repeated. If the newly created tract is
angling away from the distal vessel but is not outside of
the vessel, the guiding catheter is repositioned into the
proper direction and the process restarted. If the new tract
extended out of the vessel, extravasation of contrast will
be seen. With a small amount of contained extravasation,
the patient remains stable, particularly the post-operative
patient, who usually will have extensive scarring around
the vessel. In the presence of significant extravasation, the
perforation attempt is abandoned at least temporarily.
The area is observed intermittently on fluoroscopy for at
least 30 minutes as long as the patient remains stable.
If the extravasation continues to increase, occlusion of

the newly created tract with a micro coil is carried out. The
fine delivery catheter for the micro coil is delivered
directly into the tract through the already positioned
guiding catheter. A small, “straight” micro occluding coil
is used for the occlusion, as described in Chapter 26.
If there is no extension of the extravasation and/or once
the extravasation is stable, the guiding catheter is reposi-
tioned to redirect the RF catheter in the more appropriate
direction and the RF perforation restarted. As long as the
patient remains clinically stable and there still is a desire
to open the obstruction, the process is continued until the
distal segment of the vessel is entered. Once the distal ves-
sel is entered, the coaxial catheter is advanced over the RF
perforating catheter and the RF perforating catheter is
withdrawn. The position in the distal vessel is confirmed
angiographically through the coaxial catheter. Once the
position in the distal segment is verified, a stiff exchange
guide wire is advanced through the coaxial catheter and
positioned as far distally in the vessel as possible. The
exchange guide wire must be small enough in diameter
to accommodate a small “coronary” balloon dilation
catheter. The newly created communicating tract is
dilated sequentially as described previously. Once the
perforated tract is enlarged to the diameter of the adjacent
vessel, intravascular stents are implanted to maintain the
new communication.
Recannulation of total venous occlusions
Acute total venous occlusion usually is a result of an acute
thrombus formation in a vein that has sluggish flow
and/or has been damaged by surgical or catheter inter-

vention, including chronic indwelling intravenous
catheters and infusions. Occlusions of small veins usually
go unrecognized until access to the specific vein is desired
at a later time. Acute occlusions of large, central veins
present with the acute onset of venous congestion
“upstream” from the obstruction. The most notable exam-
ples are the appearance of the “superior vena cava syn-
drome” with distention of the head, neck and arm veins
and swelling of the head and face when the superior vena
cava becomes obstructed, or lower extremity venous con-
gestion and edema when the iliofemoral and/or inferior
vena cava become(s) obstructed.
Venous occlusions usually involve very large thrombi
and should be addressed when they occur (or are recog-
nized) and very aggressively. The goal is to remove the
thrombus along with the cause of thrombus if possible.
The fresh venous thrombus is crossed with a standard
guide and/or Terumo™ wire. If an indwelling venous
line is in place, the line is removed. Once the thrombus is
crossed, it is macerated and the debris from the macera-
tion withdrawn with a mechanical thrombectomy device/
catheter, as described in Chapter 12. Once the debris has
been removed, any remaining clot is compressed against
the vessel wall with an angioplasty balloon and/or
intravascular stent. The balloon used should be the size of
the unobstructed vein or the diameter of adjacent non-
involved veins. If there is a discrete narrowing and/or
narrowing of a surgical anastomosis in the vein as the
source of the thrombus, the stenotic lesion is treated with
dilation and the implant of an intravascular stent.

When not recognized and/or not treated acutely,
venous thrombi result in permanent chronic venous
occlusions. Even long, chronic, total occlusions of periph-
eral as well as central veins frequently can be traversed
and recanalized using long needles and/or stiff wires in
conjunction with finely tapered dilators, as described in
Chapter 21. This is accomplished relatively blindly using
“brute force” to penetrate through the obstructions. The
perforation of the obstruction usually is directed by aim-
ing for a catheter or wire that is positioned in the vessel at
the opposite end of the obstruction.
It appears that the directional control of the Safe Steer
Wire™ (Intraluminal Therapeutics Inc., Carlsbad, CA)
combined with a radio frequency source of energy in the
Safe Cross System™ (Intraluminal Therapeutics Inc.,
Carlsbad, CA) adds purposeful directional control to re-
cannulations, while the RF energy produces a more con-
trolled perforation and, as a unit, it reduces the “force”
necessary to recanalize totally occluded vascular tracts
16
.
The Safe Steer Wire™ uses optical coherence reflectome-
try to distinguish between thrombus and the viable tis-
sues of the vascular wall in order to navigate the wire only
within the limits of the vascular walls. The addition of
radio-frequency energy delivered through the same
“wire” provides the penetrating ability of the Safe Cross
System™. The Safe Cross System™ has been used suc-
cessfully for recanalization of totally occluded coronary
arteries, totally occluded intravascular stents and totally

occluded peripheral vessels
17
.
Recently intravascular ultra sound (IVUS) has been
added to the armamentarium to help identify the true
lumen during the recanalization of totally occluded cor-
onary arteries
18
. Both the Safe Cross System™ and the use
CHAPTER 31 Purposeful vascular perforations
856
of IVUS potentially could be used in the treatment of the
vascular obstructions (both arterial and venous) that are en-
countered in pediatric and congenital heart patients and
present exciting challenges for new developments in the
pediatric/congenital cardiac catheterization laboratory.
Recannulation of total acute arterial occlusions
Acute occlusion of a femoral artery during, and/or imme-
diately following, a retrograde arterial catheterization is
not an uncommon complication of cardiac catheterization
in pediatric and congenital heart patients. Arterial occlu-
sions, like most other complications, are best treated by
prevention. The meticulous handling of the arterial access
sites to prevent complications is discussed in Chapter 4.
However, particularly during interventional procedures
that require the introduction of large sheaths and/or
catheters into sometimes relatively small arteries, com-
promise of the artery may be inevitable. Occlusion of an
artery may be manifest with any of a very wide spectrum
of signs and symptoms. A totally obstructed artery can

manifest as only a decreased pulse in the involved extrem-
ity, as decreased (or absent) capillary perfusion in the
extremity, as an absent pulse with a cool and pale extrem-
ity or, in the worst case, as a cold, very pale and painful
extremity. The latter, more extreme signs of obstruction
obviously need immediate attention, while various
degrees of urgency are applied to the treatment of the
“lesser” signs of obstruction. When any degree of arterial
compromise is recognized, it should be treated at that time.
Usually, even decreased peripheral pulses in the involved
extremity indicate an artery that is totally occluded by
thrombus although, because of collateral flow around the
obstruction, the totally occluded artery may not cause
acute symptoms. Any arterial occlusion eventually may
result in claudication, growth retardation of the extremity
and/or loss of a potential access site for a future and
essential intervention.
The initial management of an acute arterial occlusion is
to continue or initiate treatment with intravenous heparin
in therapeutic doses for one to two hours while observing
the extremity closely. With return of the pulse/perfusion,
the heparin is discontinued. If the pulse/perfusion does
not return to normal and/or if there is any deterioration in
the pulse/perfusion of the extremity, further more vigor-
ous intervention is recommended. One alternative, more
aggressive courses of management is to begin intravenous
thrombolytic therapy with streptokinase, urokinase or
rtPA, which is covered in Chapters 2 and 4. When the
pulse and perfusion return with thrombolytic therapy, the
thrombolytic infusion is stopped while continuing intra-

venous heparin for at least another 24 hours. The situation
becomes more complicated when thrombolytic therapy is
initiated but is not effective and/or there is progression of
the signs of obstruction in spite of the thrombolytic. In that
situation, surgical or catheter intervention is required.
However, following the use of a thrombolytic, mechanical
intervention can be very hazardous. This has led some
centers to proceed with catheter recannulation of the
artery immediately when the initial heparin therapy was
not effective.
When mechanical intervention in the catheterization
laboratory is instituted, heparin therapy is continued. In
the infant and small child, the involved femoral artery can
be approached with a catheter/wire introduced from the
venous system, advanced prograde through the left heart,
into the aorta, manipulated to the descending aorta and
eventually, the involved artery. The entry into the left
heart is through a pre-existing PFO/ASD or through a
transseptal atrial puncture. The prograde approach
allows the introduction of multiple and larger catheters
and their use for longer periods of time without compro-
mise of an additional artery. The prograde approach
does hinder the precise control over the tip of the
catheter/wire somewhat and requires the availability of
very long exchange wires, and long diagnostic and bal-
loon dilation catheters.
In larger adolescent and adult patients, the approach
to the involved femoral artery is usually from the con-
tralateral femoral artery or even from a brachial artery.
The catheters and/or balloon dilation catheters used are

much smaller relative to the vessel in the larger patient,
and the turn into the contralateral femoral artery is less
acute, making the manipulations from another introduct-
ory artery less traumatic in the larger patient. Also the
available catheters and/or balloon dilation catheters often
are not long enough to allow a prograde approach in the
larger patient, where the catheter would have to extend
from a femoral vein to the heart, loop through the left
heart and then back to a femoral artery.
Whichever route is used, first an angiogram is obtained
with an injection in the descending aorta proximal to the
obstruction in order to identify the area of obstruction
and to provide a “mirror image” view of the uninvolved,
opposite iliofemoral arterial system as a representation of
the normal vessel anatomy. With the lesion localized, an
end-hole catheter is manipulated selectively into the
involved vessel and a floppy tipped guide wire and/or a
straight Terumo™ wire is used to probe through the fresh
obstruction (clot ± spasm) as the catheter is advanced
along with the wire. Once successfully probed, the
catheter is replaced over the wire with an appropriately
sized, low-profile, balloon angioplasty catheter. In the
infant and small child coronary artery balloons are ideal in
size and have very long catheter shafts. In smaller patients
the initial catheter is replaced over the wire with the ap-
propriate coronary balloon and/or the wire and catheter
are replaced with a fixed wire, coronary angioplasty
CHAPTER 31 Purposeful vascular perforations
857
balloon catheter. The obstructed area is dilated to a dia-

meter equal to the normal, comparative, ipsilateral vessel.
A repeat small angiogram is performed to visualize the
degree of opening and whether further repeat dilations
are necessary. Once the lumen is open and flow estab-
lished, the balloon and wires are removed. The patients
are maintained on intravenous heparin for at least an
additional 24 hours.
Complications of “purposeful
perforations”
The purposeful perforation techniques are designed to
do just thataperforate! The only, very slight, differences
between the desired, purposeful perforation and an in-
advertent, catastrophic, erroneous perforation are the
meticulous attention to the details of the planned per-
foration, total awareness of the anatomy adjacent to the
structure being perforated and the very early recognition
of the beginning of an abnormal course of the perforating
device. A purposeful perforation that goes astray com-
pletely usually results in a perforation of the external
wall of a vascular structure into the pericardium, pleura
and/or peritoneum and, as a consequence, results in a
major adverse event.
Awareness, and, in turn, prevention are the best treat-
ments of inadvertent perforations. Repeated biplane angio-
graphic visualization of the anatomy, before and during the
purposeful perforation, continually verifies the direction
and location of the perforating device. Any deviation in
the desired direction of the perforating device necessitates
a redirection of the device or termination of the procedure.
When either the fine tip of a transseptal needle or the tip

of a 2-French RF wire alone perforates through the wall of
a vascular structure, an extremely small opening is cre-
ated. These small perforations, if recognized and if noth-
ing larger is pushed through the opening, often seal on
their own. When an erroneous perforation occurs in a
patient who has had previous surgery, the area almost
always is encased in dense scar with no “space” for
“extravasation” and, in turn, is “self sealing”. In “native”
(not previously operated) areas and/or when the pressure
in the vascular structure that is perforated is very high
(e.g. the right ventricle in pulmonary atresia with intact
ventricular septum or the left atrium with severe mitral
stenosis), sealing of the perforation usually does not
occur, and certainly the perforation cannot be relied upon
to seal on its own. The abnormal perforations also cannot
seal if the abnormal course is not recognized and a larger
catheter/dilator is advanced through the initial tiny erron-
eous opening.
When an abnormal perforation occurs, the surgical and
anesthesia services are notified immediately of a pending
emergency case. When extravasation continues through
an erroneous perforation, the opening occasionally can be
“tamponaded” by the inflation of a Swan™ or angioplasty
type balloon within the vessel or chamber adjacent
and/or proximal to the site of perforation. This can be
accomplished only in vascular channels that are not the
sole vascular supply to a vital structure. If available, a
“covered stent” can be used to “cover the area” in the wall
of a perforated vessel. When a small perforation is recog-
nized while the perforating wire/needle is still in the

opening, occasionally a micro coil can be used through a
coaxial catheter over the wire/needle to occlude the per-
foration. Rarely one of the vascular occlusion devices (a
coil, umbrella or an Amplatzer™ device) can be placed in
the abnormal opening to seal a larger opening. Each unex-
pected perforation creates its own separate circumstances
and should be anticipated before attempting a desired
perforation.
Even as attempts are made to “tamponade” or occlude
the leaking perforation, a blood infusion with the previ-
ously “typed and crossed” whole blood is started. The area
into which the blood is draining (pericardium, pleura) is
tapped and an adequate sized drain secured in the space.
When significant blood loss continues, the withdrawn
blood is returned to the patient by autotransfusion
through a filtered blood administration set. The patient
is prepared for surgical intervention as expediently as
possible.
Another, somewhat rare complication during pul-
monary valve perforations in newborn patients is the pre-
mature and unexpected closure of the ductus arteriosus
due to the manipulations through the ductus during the
procedure. When the ductus is the sole source of pul-
monary blood flow, the closure of the ductus results in
progressive hypoxemia and acidosis and can lead to
death. The likelihood of spontaneous occlusion of the duc-
tus occurring is reduced by minimizing the manipulations
through the ductus, accurately maintaining and/or
increasing the prostaglandin infusion to the patient, and
maintaining the infant’s fluid volume. In the event of a

spontaneous closure of the ductus, the rate of the
prostaglandin infusion is increased and an attempt is
made to cross the ductus gently with a fine, floppy tipped
guide wire. If the ductus can be crossed expediently with
the wire, the implant of a stent in the ductus to maintain its
patency should be considered. The definitive treatment
once intractable ductus occlusion occurs, however, usu-
ally is an emergency surgical shunt.
Conclusion
As with all interventional therapeutic catheter proced-
ures, the definite risks of each procedure must be weighed
CHAPTER 31 Purposeful vascular perforations
858
against the potential benefits of the procedure. Considera-
tion of the alternative procedure should include the risks
of the alternative procedure. Cardiac lesions undergoing
purposeful catheter perforations are all very complex, and
the alternative surgical procedures for these lesions all
carry significant risks. The purposeful transcatheter per-
foration of most lesions in an experienced catheterization
laboratory are more than justified and are considered as
the first-line of therapy.
References
1. Ross J Jr, Braunwald E, and Morrow AG. Transseptal left
atrial puncture; new technique for the measurement of left
atrial pressure in man. Am J Cardiol 1959; 3(5): 653–655.
2. Park SC et al. A new atrial septostomy technique. Cathet
Cardiovasc Diagn 1975; 1(2): 195–201.
3. Kan JS et al. Percutaneous transluminal balloon valvulo-
plasty for pulmonary valve stenosis. Circulation 1984; 69: 554.

4. Latson LA. Nonsurgical treatment of a neonate with pulmon-
ary atresia and intact ventricular septum by transcatheter
puncture and balloon dilation of the atretic valve membrane.
Am J Cardiol 1991; 68(2): 277–279.
5. Parsons JM, Rees MR, and Gibbs JL. Percutaneous laser
valvotomy with balloon dilatation of the pulmonary valve as
primary treatment for pulmonary atresia. Br Heart J 1991;
66(1): 36–38.
6. Qureshi SA et al. Transcatheter laser-assisted balloon pul-
monary valve dilation in pulmonic valve atresia. Am J Cardiol
1991; 67(5): 428–431.
7. Redington AN, Cullen S, and Rigby ML. Laser or radiofre-
quency pulmonary valvotomy in neonates with pulmonary
atresia and intact ventricular septumadescription of a new
method avoiding arterial catheterization. Cardiol Young 1992;
2: 387–390.
8. Rosenthal E et al. Radiofrequency-assisted balloon dilatation
in patients with pulmonary valve atresia and an intact ven-
tricular septum. Br Heart J 1993; 69(4): 347–351.
9. Coe JY et al. Transaortic balloon valvoplasty of the pul-
monary valve. Am J Cardiol 1996; 78(1): 124–126.
10. Hausdorf G, Schneider M, and Lange P. Catheter creation of
an open outflow tract in previously atretic right ventricular
outflow tract associated with ventricular septal defect. Am J
Cardiol 1993; 72(3): 354–356.
11. El-Said HG et al. 18-year experience with transseptal proced-
ures through baffles, conduits, and other intra-atrial patches.
Catheter Cardiovasc Interv 2000; 50(4): 434–439; discussion
440.
12. McMahon CJ, El-Said HG, and Mullins CE. Transcatheter

creation of an atriopulmonary communication in the Hemi-
Fontan or Glenn circulation. Cardiol Young 2002; 12(2):
196–199.
13. Ing FF et al. Reconstruction of stenotic or occluded
iliofemoral veins and inferior vena cava using intravascular
stents: re-establishing access for future cardiac catheteriza-
tion and cardiac surgery. J Am Coll Cardiol 2001; 37
(1):
251–257.
14. Justino H, Benson LN, and Nykanen DG. Transcatheter cre-
ation of an atrial septal defect using radiofrequency perfora-
tion. Catheter Cardiovasc Interv
2001; 54(1): 83–87.
15. Fink C et al. Transcatheter recanalization of the left main
pulmonary artery after four years of complete occlusion.
Catheter Cardiovasc Interv 2001; 53(1): 81–84.
16. Cordero H et al. Initial experience and safety in the treatment
of chronic total occlusions with fiberoptic guidance technol-
ogy: optical coherent reflectometry. Catheter Cardiovasc Interv
2001; 54(2): 180–187.
17. Lee PY et al. Percutaneous recanalization of chronic subcla-
vian artery occlusion using optical coherence reflectometry-
guided radiofrequency ablation guidewire. Catheter Cardiovasc
Interv 2003; 60(4): 558–561.
18. Matsubara T et al. IVUS-guided wiring technique: promising
approach for the chronic total occlusion. Catheter Cardiovasc
Interv 2004; 61(3): 381–386.
859
Percutaneous pulmonary valve implant
A transcatheter-delivered valve mounted within a stent

and a technique for its delivery were devised and devel-
oped by Dr Philip Bonhoeffer in conjunction with Alan
Tower of the NuMED™ Corporation
1
. The Bonhoeffer™
valve and technique introduce an exciting new era to the
catheter treatment of congenital heart patients. Although
not yet a “routine” procedure, this unique device and
technique for the implant of a prosthetic valve into the
right ventricular outflow tract/pulmonary artery have
been demonstrated to be, not only doable, but an effective
and safe procedure
2
. This technique should become an
integral part of the armamentarium of the interventional
cardiologist for use in congenital heart lesions.
There are a large number of patients with congenital
heart disease who have very significant pulmonary valve
regurgitation as a result of prior surgical and/or interven-
tional procedures on the pulmonary valve and/or right
ventricular outflow tract (RVOT). These prior procedures
include all surgical patches/reconstructions of the RVOT,
surgical ventricular to pulmonary artery conduits (includ-
ing the RVOT in the Ross™ procedure), most, if not
all, surgical pulmonary valvotomies and possibly even
balloon dilation of the pulmonary valvea all regardless
of the original underlying lesion. There is an increasing
number of these patients who, over time, have developed
severe right ventricular dilation and significant right
ventricular failure as a consequence of the pulmonary

valve regurgitation (with or without residual stenosis).
The current standard therapy for these patients who have
pulmonary valve regurgitation and significant right ven-
tricular failure is the surgical implant and/or replacement
of a valved conduit. Most of these patients already have
had at least several prior major surgical procedures, and
all of the valved conduits that are implanted have their
own relatively short (relative to the life span of the
patient) duration of functional competence. These factors
make the prospect of repeat surgery even less palatable
and the idea of replacing the valve at least once percutan-
eously during a cardiac catheterization, a very desirable
alternative.
The Bonhoeffer™ valve is a glutaraldehyde-preserved
valve, which is harvested from a bovine jugular vein
and mounted within a Cheatham-Platinum™ (C-P™)
stent (NuMED Inc., Hopkinton, NY). The stent/valve is
mounted on a specially designed balloon dilation/deliv-
ery catheter (NuMED Inc., Hopkinton, NY) and is deliv-
ered percutaneously from a femoral vein puncture. The
balloon delivery catheter is specially manufactured with
a 16-French shaft and a 20 mm (or desired diameter
less than 20 mm) BIB™ delivery balloon (NuMED Inc.,
Hopkinton, NY), a transparent, thin-walled, plastic “cov-
ering” sheath which extends from the shaft of the catheter
over the entire catheter/balloon/stent/valve, and a spe-
cial 18-French “carrot” dilator tip which is incorporated
onto the tip of the delivery catheter just distal to the
balloon.
The candidates for percutaneous valve replacement are

large adolescent and adult patients with predominantly
pulmonary valve regurgitation and a fixed diameter pul-
monary valve annulus and/or RVOT. At the present time
these valves are suitable only when the valve annulus/
RVOT is no larger than 22 mm in diameter. The stent/
valves are best suited for implant in previous RVOT
conduits, which provide some length as well as the fixed
diameter of the outflow tract. The valves can be implanted
in smaller outflow tracts, however an attempt is made to
dilate the stent/valve up to 18–20 mm in diameter in
order to create the optimal diameter for the function of the
valve within the stent and to accommodate the total
cardiac output of most full grown adult patients. The
presence of associated stenosis and/or calcification in the
original valve/valve annulus do/does not, necessarily,
represent a contraindication to the implant of a percutan-
eous pulmonary valve implant; however, any stenotic
area where the valve is to be implanted always should be
32
Special innovative or new, therapeutic
catheterization procedures and devices
CHAPTER 32 Innovative, new therapeutic procedures
860
pre-dilated to within 2 mm of the proposed implant dia-
meter of the valve in order to ensure that the percutaneous
stent/valve can be expanded to its functional diameter.
Before the patient is even considered for a stent/valve
implant, the area of the pulmonary annulus and the RVOT
where the valve is to be implanted are imaged and
measured very accurately using biplane angiography. The

measurements for the seating of the stent/valve are very
critical and are made using a very accurate and reliable
calibration system for the measurements, as described in
Chapter 11. Often, the angles of one or both of the X-ray
tubes must be changed in order to orient the area of
implant precisely on edge and to obtain the most accurate
measurements of this area. At the same time, it usually is
more “comfortable” and convenient to perform the valve
implant in the straight PA and lateral views. If necessary,
“road map” images are obtained in these views to be
used for the implant in addition to the views used for the
measurements.
Once the angiograms have been obtained, the measure-
ments are verified and even before the stent/valve is
opened from its sterile packaging, a 0.035″ Super Stiff ™
delivery wire with a short floppy tip (Medi-Tech, Boston
Scientific, Natick, MA) is positioned as far distally as
possible into a branch pulmonary artery. The wire must
be positioned so that only the stiff portion of the wire
remains positioned across the RVOT and that the wire is
in a very secure and stable position. This Super Stiff™
wire must be able to support the passage of the large,
stiff, delivery balloon/catheter on which the stent/valve
is mounted through the often curved and/or circuitous
course to the RVOT. Since these patients almost always
have very large dilated right ventricles, marked pul-
monary valve regurgitation and often tricuspid valve
regurgitation, the manipulation of any catheter into a sat-
isfactory distal pulmonary artery location for this posi-
tioning of the wire represents a significant challenge in

each of these procedures. Any extra time spent in achiev-
ing a very secure position of the wire in a very distal branch
pulmonary artery, is time well spent during the catheter-
ization procedure. Once the wire is in place the catheter
that was used for the delivery of the wire is left in place
over the wire and kept on a slow continuous flush through
a wire back-bleed valve to protect the ventricle and valve
structures from the rough surface of the wire, to keep clots
from forming on the wire, and to allow repositioning of
the wire should it work its way backward while the
stent/valve is being prepared. Although not essential, it
simplifies the procedure to introduce a separate angio-
graphic catheter from a separate vein and position it in the
area of the proposed stent/valve implant. Although
angiograms can be performed with injection through the
protective sheath over the balloon/stent once it has been
withdrawn off the balloon/stent/valve, these angiograms
cannot be performed before the sheath is withdrawn, and
the limited angiograms through the sheath and around
the catheter never are as satisfactory as an angiogram
obtained with a power injection through a separate
catheter in the area.
Once the wire is in place, the packaged valve is opened,
inspected and the valve is checked for its competency. The
tissue valve comes mounted within a 40 mm long C-P™
stent. The stent/valve is packaged in the dilated configura-
tion with the stent/valve opened to a 20 mm diameter.
Before the stent/valve is used, it is washed sequentially
four, or more, times, each time in a separate, fresh solution
of normal saline or Ringer’s lactate. During the wash, the

valve function is tested. The stent is pulled longitudinally
through the flush solution so that the solution enters one
of the open ends of the stent. When the fluid enters from
the distal (pulmonary artery) end of the stent, the flow of
the fluid entering the stent closes the valve and should not
allow the fluid to pass through it. When the stent/valve
is filled with fluid and is held upright (vertically) with
the distal end up, the fluid within the stent/valve is held
within the stent/valve like a cup. When the stent is pulled
through the fluid in the opposite direction (from proximal
to distal), the fluid passing into the stent opens the valve
completely and allows the fluid to flow freely through it.
Once the Super Stiff ™ delivery wire is in place and the
valve has been tested, the valve/stent is mounted on the
delivery balloon/catheter. First, a separate, deflated, 8 or
10 mm diameter, 4 cm long, but very smooth, standard
angioplasty balloon is passed gently through the fully
expanded valve/stent from the proximal to the distal end
and the balloon is inflated carefully, but to its designated
diameter and pressure. The stent/valve then is com-
pressed (crimped) circumferentially over the inflated bal-
loon very carefully, in slow, sequential steps moving
circumferentially around the stent until the stent with the
contained valve is compressed fairly tightly over the 8 or
10 mm balloon. This balloon is deflated and very carefully
withdrawn from the valve/stent. The withdrawal must be
very meticulous in order not to catch the valve nor put any
tension on the valve leaflets within the stent since the
surface of the deflated balloon, which was previously
inflated, will be rough and the balloon is being withdrawn

against the opening direction of the valve mechanism. The
large special balloon delivery catheter for the stent/valve
then is introduced into the partially compressed stent/
valve, paying very careful attention to inserting the tip of
the balloon delivery catheter into the proximal (right vent-
ricular) end of the stent/valve. As a “reminder”, the stent
valve has blue sutures around its distal (pulmonary artery)
end to correspond to the “blue carrot” dilator tip of the
introducer balloon/catheter.
The stent/valve is centered exactly over the BIB™
delivery balloon and again circumferentially and firmly
CHAPTER 32 Innovative, new therapeutic procedures
861
but carefully, slowly and sequentially compressed and
crimped over the deflated balloon. Once the stent/valve is
compressed tightly on the balloon, the clear protective
sheath is advanced off the more proximal shaft of the
delivery catheter to cover the balloon/stent/valve. The
initial introduction of the protective sheath over the prox-
imal end of the stent is performed very meticulously with
manual compression of the tip of each of the struts of the
stent in order to be sure that each separate tip passes
inside of the protective sheath as it is advanced onto the
stent. The protective sheath is advanced until it abuts the
proximal end of the “carrot” dilator tip. The valve/stent
then is ready for introduction and delivery. There are two
separate marks on the shaft of the catheter, which corre-
spond to the two separate positions of the proximal end of
the protective sheath when the tip of the sheath is entirely
over, or withdrawn completely off the stent/valve. These

marks are embedded in the walls at the proximal end of
the shaft of the delivery catheter. The protective sheath
is not radio-opaque, so these marks represent the only
way of verifying the position of the protective sheath in
relation to the stent/valve.
The catheter that was utilized for positioning the wire
and the short introductory sheath are removed over the
wire, being very careful to maintain the wire in its secure
distal position in the pulmonary artery. Once the short
sheath and catheter are removed, the skin incision and
subcutaneous tissues over the wire are enlarged enough
to accommodate the circumference of a 22-French system
over the wire. The skin/subcutaneous tract/vein are pre-
dilated with a separate 22-French short dilator. A dilator
that is larger than the delivery catheter is used to ensure
the easy passage of the mounted stent/valve/balloon
through the skin/subcutaneous tissues and the wall of the
vein at the introductory site. The dilator is removed over
the wire and the stent/valve on the special balloon deliv-
ery catheter is introduced over the wire and through the
dilated tract. With careful observation of the tip and the
course of the wire on fluoroscopy, the mounted stent/
valve is advanced over the wire to the area of implant.
This part of the procedure can be very difficult, as even
the 0.035″ Super Stiff™ wire may not support the stent/
valve/balloon delivery catheter adequately. Advancing
the catheter/stent/valve usually pushes the combination
to the “outer circumference” of the curved path to the
position in the outflow tract. Pushing the catheter/stent/
valve a short distance over the wire and then alternately

withdrawing the shaft of the wire an equally short dis-
tance (without allowing the tip of the wire to withdraw!)
usually allows the catheter/stent/valve to be “inched”
forward over the wire and into position.
Once the balloon/valve/stent has reached the area of
implant, a biplane angiogram is performed and compared
with the baseline, “road map” angiogram in the same
views. Often the large and stiff catheter/stent/valve over
the wire will change the relative position of structures
in the area. Any necessary adjustments of the balloon/
valve/stent position compared to the exact location for
implant are made and then the protective sheath is with-
drawn off the valve. The protective sheath is not visible
on fluoroscopy, however its position is checked by the
calibrated marks at the proximal end of the delivery
catheter. After the protective sheath is withdrawn, a
repeat angiogram is performed to ensure that the with-
drawal of the sheath did not change the relative position
of the balloon and stent in relation to the desired position
for implant. When the stent appears to be in the precise,
desired position and the measurements appear satisfac-
tory, the inner balloon of the BIB™ dilation balloon is
inflated. Once the inner balloon is inflated completely, the
position in the annulus area is rechecked with a repeat
angiogram, the position of the stent/valve adjusted, and
the outer balloon inflated to fix the stent/valve in position.
The BIB™ balloon is inflated at least one more time to
ensure full dilation and fixation of the stent in the annulus.
Subsequent maneuvers depend upon the degree of
“fixation” of the stent/valve in the annulus. A repeat angio-

gram is performed through the second catheter (or the
protective sheath) to verify the location and fixation of
the stent/valve. When the stent is fixed in a satisfactory
position, the balloon/delivery catheter is withdrawn very
carefully over the wire and out of the stent/valve. This is a
very critical part of the procedure as the deflated BIB™
delivery balloon has a very rough and irregular surface
and easily catches on the valve and/or the supporting
stent. It frequently must be “teased” out of the stent/
valve. It may be possible to re-advance the protective
sheath at least partially over the balloon as the balloon is
deflating within the stent/valve. Once over the balloon,
the smooth surface of the sheath provides a separation
between the rough balloon and the valve. When the
original pulmonary/RVOT annulus was only a few mm
smaller than the expanded diameter of the stent/valve,
the withdrawal is even more precarious and must be
performed very cautiously.
Once the delivery balloon/catheter has been with-
drawn from the stent, the second angiographic catheter
is manipulated gently through the stent/valve and a
repeat pulmonary artery angiogram is recorded. If a sec-
ond catheter is not utilized, the balloon/delivery catheter
is withdrawn over the wire and replaced with an end-
hole, multipurpose catheter or a Multi-track™ catheter
(B. Braun Medical Inc., Bethlehem, PA), which then is
re-advanced carefully through the valve to the pulmonary
artery distal to the valve. Repeat hemodynamics are meas-
ured and a repeat pulmonary artery angiogram recorded
with the injection of the contrast performed distal to the

valve. If the stent/valve appears at all precarious in its
CHAPTER 32 Innovative, new therapeutic procedures
862
freshly implanted position, the hemodynamic/anatomic
assessment of the valve postimplant is performed by
echo/Doppler interrogation only, without any attempt to
cross the freshly implanted stent/valve with a catheter.
Annulus/outflow tract “reduction” with a
“banded” self-expanding covered stent
In its present configuration, the Bonhoeffer™ stent/valve
is suitable only for patients with a rigid, fixed diameter,
outflow tract of 20 mm or less. To overcome this problem,
a unique catheter-delivered self-expanding covered
stent/internal band has been developed and tested in
animals to percutaneously reduce the diameter of the
widely dilated pulmonary artery/RVOT
3,4
. It has been
tested with a self-contained bovine jugular vein valve and
as an internal stent/band only device. The basis of this
device is a self-expanding, covered, Nitinol™ stent, which
has a central, open tubular portion of a fixed diameter and
with both ends of the tubular device widely flared (AMF,
Groupe Lepine, Lyon, France). The entire stent is covered
with a 0.3 mm polytetrafluoroethylene (PTFE) membrane
(Zeus Inc., Orangeburg, SC) to make the walls impervi-
ous. The central tubular portion creates a lumen with a
fixed diameter of 18 mm while both of the distal ends
“flare” away from the central portion and are capable of
expanding up to 30 mm in diameter. The very wide dia-

meter of the ends allows fixation of the stent/band into
much larger diameter areas in an aneurysmal vessel/
outflow tract. The expanded device has the appearance
of a large “dumbbell” with a large lumen extending
from end to end through its center. This configuration
allows the creation of a circumferential band with a
fixed lumen, which is the diameter of the central portion,
while the expanded ends fix the stent/band into very
large, aneurysmally dilated right ventricular outflow
tracts/pulmonary arteries (RVOT/PA) and occlude flow
from around the central lumen by the PTFE covering. This
covered stent/band can be implanted either as only
the tubular covered stent/band or can have a bovine
jugular vein valve incorporated into the tubular portion
of the central lumen for a “one-stage” outflow tract
reduction–valve implant procedure.
Most patients with significant pulmonary regurgitation
do have some remnant of the original and/or a tissue pul-
monary valve, which may or may not be stenotic. Even in
a markedly aneurysmal pulmonary artery/right ventricu-
lar outflow tract, when there is any residual stenosis of the
outflow tract, the stenosis is pre-dilated before the implant
of the stent/band with or without the incorporated bovine
valve. The minimal diameter of the outflow tract must
allow the full expansion of the valve segment of the stent/
band device and the eventual valve.
When the stent/band is implanted with no incorpor-
ated valve, it creates a fixed, rigid diameter of the RVOT
with a maximum diameter of 18 mm. This tubular “band”
is ideal for the subsequent percutaneous implant of

a Bonhoeffer™ stent/valve in its present configuration.
The compressed stent/band is delivered percutaneously
from the femoral vein, over a pre-positioned Super Stiff™
wire. The band portion of the self-expanding stent is cen-
tered in the area where the valve is to be implanted and
the entire stent/band is extruded from the delivery
sheath. As the stent/band is extruded, the ends flare to
their predetermined wide diameters with the central area
fixed at the diameter of the band. The wide diameters of
the ends of the self-expanding stent fix the stent/band in
place, while the central tubular area serves as an eventual
site to fix a stent/valve. The covering of the flared ends
of the stent funnel the entire flow through the lumen of
the central, “banded” area. When the valve is implanted
separately, the valved stent is delivered with the enclosed
bovine valve and is implanted in this tubular portion
of the stent/band approximately two months after the
“stent/band” was placed.
Percutaneous valve implant in the aortic
position
The bovine jugular vein valve mounted within a stent also
provides the possibility for a “percutaneous” aortic valve
replacement. This concept has been tried in animals and
performed at least once in a human
5–7
. The percutaneous
aortic valves are mounted in much shorter and stronger
stents. There still are problems in placing such a stent/
valve in the aortic position, which are significantly
greater than in the pulmonary position. The stent/valve

implanted in the aortic root must not occlude the orifices
of the coronary arteries. In most cases, particularly in the
older adult patient, dilation of the aortic sinuses displaces
the orifices of the coronary arteries far laterally and away
from a stent placed in the aortic annulus. In addition to the
problem of the coronary ostia, the very large stent/valve
delivery system is introduced into an artery, which at the
present size and configuration of the stent/valve requires
a surgical cut-down on the femoral or even iliac artery. An
alternative is the delivery of the stent-mounted valve pro-
grade through a transseptal puncture, the mitral valve
and the left ventricle. Although the larger delivery system
can be introduced percutaneously into a vein, it then
requires the extensive manipulation of the large and stiff
delivery system/stent/valve through a circuitous course
through the left heart. The aortic roots of patients with
aortic regurgitation usually are markedly dilated and far
larger in diameter than the currently available bovine
or other tissue valves. At the same time the percutaneous
CHAPTER 32 Innovative, new therapeutic procedures
863
aortic valve is placed within the thickened (calcified) ori-
ginal valve, which reduces the diameter significantly and
allows a very secure implant.
In spite of the present obstacles, this concept for a trans-
catheter replacement of the aortic valve appears very
promising for future developments.
Prospective catheterization laboratory
completion of lateral tunnel/Fontan
circuits

In spite of the lack of prospective and/or planned com-
mercial development in this area, the most innovative
uses of covered stents in humans to date have been in
pediatric/congenital heart lesions. Covered stents were
used to “rebuild” disrupted intra-atrial, venous chan-
nels/baffles in several complex patients with single
ventricles who had undergone “Fontan” cavopulmonary
type single ventricle repairs, and where the baffles and/or
venous channels were disrupted and leaking significant-
ly. Two very different types of covered stent were used
initially for this purpose in two similar, very sick patients,
and both under very extenuating circumstances. The cov-
ered stents were placed in disrupted lateral tunnel chan-
nels to percutaneously repair the “tunnels”. The covered
stents extended from the inferior vena cava, at the caudal
end of the right atrium, through the area of the “true”
right atrium and superiorly into the base of the caval–
pulmonary anastomosis with the right pulmonary artery.
In doing so the channels that were created with the cov-
ered stents directed all of the systemic venous flow from
the inferior cava to the pulmonary arteries and eliminated
all of the intra-atrial leaking
8
.
With some collaborative cardiology–surgery “pre-
planning” in prospective studies and with very little
change in either the surgical or the catheterization pro-
cedure, the use of covered stents in “Fontan” patients is
being extended to provide an elective final phase of the
“Fontan completion” in the catheterization laboratory

9,10
.
The “catheter completion” still requires further improve-
ments in the design of the covered stents and a specifically
“pre-planned” second stage, “bi-directional Glenn”, using
one of two different proposed methods.
When performing the bi-directional Glenn, the surgeon
creates a contiguous “floor” or, at the most, a very small
restrictive opening, between the caudal surface of the
right pulmonary artery and the most cephalic “roof” of
the right atrium. Ideally, the circumference of this “roof ”
and the orifice of the inferior vena cava into the right
atrium both are outlined with opaque sutures to assist the
later catheter steps in the procedure. A wide open interatrial
communication (common atrium!) also must be ensured
at the time of the bi-directional Glenn anastomosis. For the
“completion” of the Fontan in the catheterization labor-
atory, the “roof/floor” of the right atrium/right pul-
monary artery is punctured with a “transseptal” type
puncture from either the right atrium into the pulmonary
artery or from the superior vena cava/pulmonary artery
into the right atrium, and the opening is dilated to a dia-
meter 2–3 mm smaller than the diameter of the stent/graft
which will be implanted. A channel is created from the
inferior vena cava to the opening in the right pulmonary
artery with a large, long, specifically manufactured,
covered stent, which is implanted extending from the
cephalic, newly created opening in the floor of the right
pulmonary artery, through the cavity of the right atrium
and caudally, well into the inferior vena cava. This creates

a separate, confined inferior caval to pulmonary artery
channel. This technique requires relatively little change in
the “second-stage” surgical procedure, but is very chal-
lenging for the interventionist in the catheterization labor-
atory. The interposed covered stent must create a channel
large enough in diameter to carry two thirds of the sys-
temic venous blood flow and, after the “stent” shrinks in
length during the implant, still be long enough to extend
exactly from the IVC through the right atrium and into the
pulmonary artery. An even greater challenge for this pro-
cedure is generated by the subsequent eventual growth of
the patients. The new “tube graft” must be large enough in
diameter to accommodate the increasing volume in blood
flow and the growth of the patient and the patient’s heart,
in particular the length of the right atrium! The diameter
of some stent grafts may be adjustable with subsequent
balloon dilation, but the implant length is fixeda or pos-
sibly it could shrinkawith further dilation of the stent that
creates the channel!
An alternative technique requires an open-heart, surg-
ical procedure on bypass by the surgeon with opening of
the right atrium at the earlier “Glenn” stage, but currently
makes the procedure more suitable for patients of all sizes
and makes the final stage, which is performed in the
catheterization laboratory by the interventionist, much
more straightforward. During the surgical “bidirectional
Glenn” stage of the procedure, the surgeon ensures that
there is an adequate intra-atrial communication and then
establishes the complete lateral tunnel, caval–pulmonary
baffle within the right atrium, however, with two very

significant variations from the standard lateral tunnel. First,
a large (15–20 mm) window or “fenestration”. Which com-
municates with the right atrium, is created in the medial
wall of the baffle of the lateral tunnel. Secondly, the ceph-
alad end of the newly created lateral tunnel is attached
to the caudal surface of the intact right pulmonary artery,
but no opening or communication is made between the right
atrium/lateral tunnel and the pulmonary artery at the
time of the bidirectional “Glenn” surgical procedurea
i.e. the inferior vena cava/right atrial blood flow is not
CHAPTER 32 Innovative, new therapeutic procedures
864
opened into the pulmonary artery and, as a consequence,
the patients function as though they only have had the
bidirectional Glenn anastomosis!
For the “completion” of the “Fontan” the interventional
cardiologist in the catheterization laboratory punctures
through the intact roof/floor between the right atrium
and the right pulmonary artery, into the right pulmonary
artery at the cephalad end of the lateral tunnel, dilates the
tunnel to right pulmonary artery communication and
implants a large diameter, but short stent (covered?) to
widen the opening and to keep this newly created channel
open. This stage of the procedure is performed from a per-
cutaneous femoral venous approach. The large “fenestra-
tion” that the surgeon has created in the medial wall of the
lateral tunnel is closed with a percutaneous atrial septal
occlusion device! The advantage of this version of the
“Fontan completion” would be that the native tissue of the
“lateral tunnel” will grow in “diameter” to accommodate

the patient’s growth and increased blood flow with the
growth. The partial native tissue lateral tunnel also has a
better chance of accommodating the growth of the patient
in “length” as occurs with the current lateral tunnel,
cavopulmonary anastomoses.
The capability of having custom-designed, large, cov-
ered stents, which are made to fit specific lesions/patients,
would make it conceivable to perform a conversion of a
failed, classic, right atrial to pulmonary artery “Fontan” to
a lateral tunnel cavopulmonary “Fontan” in the catheter-
ization laboratory. These special covered stents could
be modifications of the aortic “stent grafts”, which already
are used for the catheter treatment of aortic aneurysms
11
.
The stent graft would have to be built to the specific
dimensions of each individual patient (e.g. length of RA,
diameter of IVC–RA junction and proposed SVC–PA
diameters).
In the catheterization laboratory, first the original atrial
septum/septal patch would require reopening widely in
order to allow the coronary sinus blood to return back into
the circulation once the “lateral tunnel” is completed. The
cephalic end of the catheter-implanted, stent graft “lateral
tunnel” could be implanted in the original direct right
atrial to pulmonary artery connection or a puncture could
be performed from the most cephalic portion of the stump
of the right atrium through the bottom (caudal surface) of
the adjacent right pulmonary artery and then the cephalic
end of the custom “lateral tunnel” stent graft implanted

directly into the pulmonary artery. The stent graft would
form a tunnel through the right atrium between the
inferior vena cava and the right pulmonary artery sim-
ilar to the catheterization “completion” of the “Fontan”
described above. If the cephalad end of the catheter-
implanted “lateral tunnel” was not implanted through the
original connection between the right atrium and pul-
monary artery connection, then this original right atrial to
pulmonary artery connection would be occluded separ-
ately with a catheter-delivered device. The superior vena
cava still would need connecting to the pulmonary artery.
Once the cephalic end of the stent graft, which passes
through the right atrium to the right pulmonary artery,
was anchored into the right pulmonary artery, the super-
ior vena cava still would have its native connection to the
right atrium and be draining past the pulmonary artery,
around and outside of the stent graft, with the result that
the superior vena cava blood still would pass through the
atrial septal defect into the systemic output. An additional
puncture would be necessary from the superior vena cava
into the top (cephalad surface) of the right pulmonary
artery. A second, short, 18–20 mm diameter covered stent
would be implanted in this communication, which would
divert all of the superior vena caval blood into the pul-
monary artery. Since the surrounding area would be
so densely scarred from two previous cardiac surgeries
(at least one systemic to pulmonary shunt and/or a
“Glenn” plus the original “Fontan”), this extra vascular
vessel-to-vessel puncture would not result in a significant
extravasation of blood into the surrounding thorax.

Although “catheterization revision” would represent a
very extensive procedure in the catheterization labor-
atory, with the proper equipment, it should be very doable
and still would be less of a procedure than the current
surgical revision of a classic “Fontan”.
Perforation of vessel walls with creation
of a vessel-to-vessel communication
and/or shunt
Dr Kurt Amplatz reported the possibility of percuta-
neously creating communications between native, vascu-
lar structures using a modification of the Nitinol™ mesh,
Amplatzer™ occluders from a preliminary study in ani-
mals. Dr Chigogidze from the Bakuolev Scientific Center
of Cardio-Vascular Surgery in Russia also has performed
some animal work on the successful percutaneous cre-
ation in the cardiac catheterization laboratory of vascular
shunts between adjacent major vascular structures. Com-
munications were created between adjacent venous struc-
tures, the aorta and vena cava, the pulmonary artery and
the superior vena cava and even between the aorta and a
pulmonary artery. The minute details of the procedure are
not available to this author, however, the general concept
was presented and represents a very exciting develop-
ment in the area of transcatheter therapy
12
.
Tiny intravascular magneto-mechanical devices are
placed in the adjacent vessels ostensibly to pull the vessels
together and to direct a flexible “kinetic” needle as it punc-
tures from one vessel to the other. Once the magnets are in

place in the adjacent vessels, a puncture from one vessel
CHAPTER 32 Innovative, new therapeutic procedures
865
into the adjacent vessel is performed with the special
“kinetic” needle attached to a flexible wire. The exact
mechanism of the needle puncture is unclear. Once the
needle has entered the adjacent vessel, it is captured
with a snare catheter and the needle with the attached
wire is withdrawn through the peripheral entry site of
this vessel. The exteriorization of this wire creates a
through-and-through “rail” wire. A large sheath/dilator
set is advanced over the wire and pulled through the new
communication between the vessels. Presumably, the
presence of the large sheath/dilator within the newly
created openings in the walls of both of the vessels pre-
vents exsanguination from the now large puncture sites in
the walls of these vessels.
Once the sheath has passed through the puncture site
into the adjacent vessel, the dilator is removed over the
wire. A specially prepared, Nitinol™ self-expanding, cov-
ered, stent graft, which is flared widely at both ends, is
implanted in the area between the two vessels by with-
drawing the sheath off the stent graft. The Nitinol™ stent
graft is covered with polyurethane and has a central dia-
meter of the desired opening between the two vessels. Once
implanted, the covered stent creates a channel between
the two vessels while at the same time sealing the open-
ings in the walls of the vessel.
Reportedly, vena cava to portal vein, superior vena
cava to pulmonary artery and aorta to pulmonary artery

communications all have been created successfully in ani-
mals! In addition, with an extension of this technique, a
normal heart in an animal was converted to a completed
“Fontan” circuit by the creation of: (1) a superior vena
cava to right pulmonary artery communication with one
stent graft (i.e. a “Glenn” shunt); (2) exclusion of the main
pulmonary artery trunk and the pulmonary valve from
the circulation by a covered stent which bridged across
the main pulmonary artery from the right to the left pul-
monary artery; and (3) the completion of the “Fontan
atrial lateral tunnel” with a covered stent, which extended
from the inferior vena cava and through the right atrium,
connecting into the right pulmonary artery/superior vena
caval channel. To date, these procedures have been per-
formed only in animals, but the concept certainly is an
exciting evolution into the immediate future of interven-
tional cardiology.
Completion of the first stage “Norwood”
procedure for hypoplastic left heart
syndome in the catheterization laboratory
By the combination of implanting a stent in the ductus
arteriosus to maintain its patency and the implant of “flow
restrictors” in the proximal branch pulmonary arteries to
reduce pulmonary artery flow and pressure, Drs Boucek
and Chan in conjunction with AGA Medical Corporation
(AGA Medical Corp., Golden Valley, MN) have devel-
oped a procedure for the initial palliation of patients with
hypoplastic left heart syndrome entirely in the cardiac
catheterization laboratory
13

.
The special flow restricting devices (AGA Medical
Corp., Golden Valley, MN) are very similar to the
Amplatzer™ Intravascular Plugs (AGA Medical Corp.,
Golden Valley, MN), however the flow restrictors have
holes of a specific diameter passing through the devices.
After very accurate sizing of both the proximal right and
left pulmonary arteries, flow restrictors that are several
millimeters larger in diameter than each pulmonary
artery are placed precisely in each proximal branch pul-
monary artery. If the flow restrictor devices are under-
sized, they can migrate distally and/or change their
orientation in the vessel and totally obstruct flow to the
entire lung and/or a major branch. If the flow restrictor
is too large for the particular vessel, the openings in the
central portion of the device may not open entirely or
sufficiently to allow any flow through the device.
Once the flow restrictors are placed in the separate right
and left pulmonary arteries, the ductus arteriosus is
stented with a self-expanding stent, which again, is sev-
eral millimeters larger in diameter than the existing duc-
tus. The stent also must be long enough to cover all of the
ductal tissue, but at the same time not so long as to inter-
fere with the pulmonary valve and/or compromise the
entrance of the transverse aortic arch into the descending
aorta at the area of the ductus. Presently, the Precise™
or Smart™ stents (Cordis Corp., Miami Lakes, FL) or the
Protégé™ stents (ev3, Plymouth, MN) appear to be the
most suitable for this use. The Protégé™ stents have
the advantage of having a separate attach–release mechan-

ism, which allows withdrawal or repositioning of the
stent up until the moment of final deployment. Because of
the limited lengths of these stents that are available, occa-
sionally several stents must be overlapped in the ductus to
cover all of the ductal tissue adequately.
At present, the flow restrictor devices are custom manu-
factured for the specific investigational study and are not
available commercially. The procedure is technically very
challenging in small sick hypoplastic left heart syndrome
patients, however, certainly it is less traumatic than the
comparable “standard” surgical “first stage Norwood”
procedure. As the equipment is developed and improves,
this innovative approach should replace the initial sur-
gery for many (most) of these patients.
Radio-frequency tissue desiccation
Sigwart introduced the concept of “ablation” of an abnor-
mal portion of the ventricular septum for the treatment of
CHAPTER 32 Innovative, new therapeutic procedures
866
obstructive hypertrophic cardiomyopathy by the sub-
selective infusion of alcohol into a septal perforating coron-
ary artery
14
. The relief of the obstruction is accomplished
by the purposeful creation of a localized myocardial in-
farction in the abnormal septal tissues. The procedure is
very effective for the relief of left ventricular outflow tract
obstruction, however the procedure seldom is used in
pediatric patients and, as a consequence, is technically
very challenging for the majority of physicians who

perform therapeutic interventions primarily on pediatric
and/or congenital heart patients. The alcohol septal ablation
procedure definitely is not without some significant
complications.
More recently the ablation of left ventricular septal tis-
sue has been performed using radio-frequency (RF)
energy. “Excavation” of the left ventricular outflow tract
with reduction of the gradient across the outflow tract
has been accomplished using an RF catheter with an 8 or
10 mm electrode tip, a high output 100-w RF generator
(Boston Scientific, Natick, MA), which is set at 60 watts for
1 minute with the tip cooled with a continuous saline flush
of 300 to 600 ml/hour while the RF is active. The catheter
tip, which is deflected against the tissue while the energy
is applied, is drawn along the area of abnormal muscle in
the outflow tract in order to create a channel or groove
in the tissues. The RF energy apparently “excavates” the
area of the septum by desiccation of the tissues without
any resultant adverse effects from the disseminated
(embolized) materials (gases?). This particular use of
radio-frequency energy is very new and the long-term
effects/consequences certainly are not known.
Hypertrophic cardiomyopathy with left ventricular
obstruction is a relatively rare lesion in the pediatric popu-
lation, however, if RF energy is effective in relieving and
sustaining the relief of the left sided muscular obstruction,
then this procedure should have extensive application for
the innumerable congenital heart patients who have mus-
cular right ventricular outflow tract obstruction with or with-
out other lesions.

Sideris “frameless” transcatheter patch
occlusion devices
The latest, and completely different entry into the pro-
cedures and devices for the occlusion of intracardiac de-
fects are the Sideris™ “frameless” Transcatheter Patches
(Pediatric Cardiology Custom Medical Devices, Athens,
Greece), which were developed to eliminate the perceived
problems of the metal frames that are present in all of
the current intracardiac occlusion devices
15
. The Sideris
“Patches” and techniques are based on experimental
information that the porous material of polyurethane
foam, when firmly fixed in place against a surface within
the circulation for the relatively short time of 23–48 hours,
stimulates adhesions, which adhere to tissues securely
enough to hold the patch in place permanently in that
length of time. The fixation is secure enough to withstand
displacement from the pressures/flow within or through
an intracardiac defect
16,17
. This has been demonstrated in
animal models and is reported to be successful in atrial
septal defects, ventricular septal defects and very large
patent ductus in humans.
ASD frameless patch device
The occlusion of atrial septal defects with a “frameless”
patch was performed in animals and, on a compassionate
use basis, in humans outside of the United States
18

. The
delivery system consists of two relatively large, inflatable
spherical latex balloons attached in series, immediately
adjacent to each other, at the distal end of a triple lumen,
end-hole catheter (Pediatric Cardiology Custom Medical
Devices, Athens, Greece). The most distal balloon, which
communicates with one lumen of the catheter, is covered
completely with a “sleeve patch” or “sack”, which is com-
posed of a thin layer of polyurethane foam. The sheet of
foam is wrapped over the distal balloon so that the “open
end” of the folded patch extends proximally against the
second, more proximal balloon. The second balloon is
mounted on the triple lumen catheter immediately prox-
imal to the first balloon and communicates with the second
lumen. The central (third) lumen of the catheter extends
through the distal end of the catheter and, in turn, through
the center of the patch and allows the balloons/patch to be
delivered over a wire.
The polyurethane patch has a radio-opaque suture run-
ning through the proximal margins of the patch for radio-
graphic visualization of the general area of the patch.
During implant and while the patch is “fixing” on the sep-
tum, there also is a very long loop of “retrieval” nylon
suture passing through an edge of the “patch” material.
Both ends of this suture extend from the patch, which is
positioned in the atrial septum, and exit out of the body in
the inguinal area adjacent to the catheter. This suture
serves as a retrieval mechanism should the patch not
adhere to the particular area after the balloon is deflated.
The atrial defect is measured very accurately with a

static sizing balloon. A short tipped, stiff exchange length
guide wire is positioned across the defect and into a left
upper pulmonary vein similarly to the delivery of other
ASD occlusion devices. A large, long sheath/dilator that
will accommodate the delivery balloon with the covering
patch is advanced over a wire to the area of the right
atrium, the dilator alone is removed and the sheath is
cleared passively of air/clot and placed on a continuous
flush. The special balloon catheter carrying the collapsed
polyurethane “patch” on the deflated latex balloon is
CHAPTER 32 Innovative, new therapeutic procedures
867
introduced over the wire, into the large, long sheath,
advanced out of the sheath and into the left atrium over the
wire. The two ends of the long suture, which is attached
to the “patch” material, extend back through the long
sheath, along the side of the shaft of the balloon catheter
and out of the proximal end of the long sheath. The distal
balloon with the “patch” covering it is inflated to a dia-
meter “several” millimeters larger than the measured
diameter of the defect. The inflated distal balloon, which is
covered with the “patch”, is pulled back against and into
the defect by withdrawing the balloon catheter while
observing the balloon and septum on TEE or ICE. If the
balloon/patch pulls through the defect, the balloon is
deflated, re-advanced into the left atrium, refilled with
slightly more fluid and repositioned back against/into the
septal defect. The position and degree of occlusion are
confirmed with TEE/ICE. If there still is leakage around
the balloon, the balloon is inflated even further. Once

the distal balloon is against the septum and completely
occluding the defect, the second, more proximal balloon is
inflated. This balloon inflates in the right atrium and helps
to “stabilize” the distal, left atrial balloon with the “patch”
against the rim of the defect. The “patch” material, which
is covering the distal balloon, now should be in firm con-
tact with the entire rim of the atrial septal defect. The
polyurethane sheet of material is wrapped around the
balloon and the catheter, but not fastened to the shaft of
the catheter. Because of the extensive stretchability of the
polyurethane foam and of the latex balloons, only three
sizes of balloon/patch combinations reportedly are suit-
able for all defects.
The constant pressure of the polyurethane against
the septum, which is necessary for the fixation of the
polyurethane “patch”, is accomplished by continual ten-
sion against the distal balloon, which is maintained on the
shaft of the balloon catheter by fastening the proximal shaft
of the catheter against the skin “outside” of the vascula-
ture system for 48 hours! The second large latex balloon,
which is inflated on the right side of the septum, helps to
stabilize the “patch” in position against the rim of the
atrial septal defect for the 48 hours, but does not hold the
distal balloon against the septum.
Enough traction is applied to the balloon/delivery
catheter at the skin surface to maintain the distal (left
atrial) balloon against, and partially into, the atrial defect
without pulling the balloon through the defect. The bal-
loon catheter is fixed securely against the patient’s leg at
the puncture site with several sutures in order to maintain

the “traction” on the catheter/balloon against the atrial
septum. The patient is returned to the hospital ward but is
kept at strict bed rest with their leg extended straight for 48
hours. All of this time the two balloons are inflated within
the heart and the proximal end of the balloon catheter
extends out of the femoral puncture site under some
tension! The patient is returned to the catheterization
laboratory in 24 hours to check the balloons’ position and
stability. When the distal balloon remains in its proper
position, it moves only in synchrony with the atrial septal
motion and should not be “bobbling” in the left atrium.
After 48 hours the patient is returned to the catheteriza-
tion laboratory. The tension is released from the catheter
at the skin puncture site and the left atrial balloon, which
is covered with the “patch”, is deflated. This allows the
originally spherical patch material to collapse and shrink
into a flat, but crumpled “patch” across the atrial defect
while the deflated and collapsed balloon now is posi-
tioned on the right atrial side of the patch. The area is inter-
rogated by TEE for any residual leak or displacement
of the patch. When comfortable with the position and
fixation of the patch, the more proximal balloon is deflated
and the balloon catheter with the two deflated balloons is
withdrawn very gently away from the septum. The sep-
tum and the now free “patch” again are interrogated with
the TEE/ICE for security on the septum, position and any
leak. When satisfied with the implant, the long “retain-
ing” suture is withdrawn slowly and carefully from the
“patch” material by pulling one end while releasing the
opposite end of the suture.

Because the “patch” material is pushed against the
inner edge of the atrial defect and adheres to the septum
by that contact, theoretically, atrial defects that have no
rim and which are very large, can be occluded with
this device. Further multicenter, monitored trials of this
device with controlled rigid follow-up are necessary to
establish the utility of this very innovative but radical
approach to ASD closure.
Frameless VSD patch
The same “frameless” patch concept has been reported for
the closure of perimembranous ventricular septal defects
(VSD). The patch is similar (identical?) to the frameless
patches for atrial septal defects (Pediatric Cardiology
Custom Medical Devices, Athens, Greece). The ventricu-
lar septal defect is crossed from the left ventricle into the
right ventricle and a through-and-through wire “rail” cre-
ated exactly as described for other techniques for per-
imembranous ventricular defect closures (Chapter 30).
Once the “rail” is created, the frameless VSD device,
which is mounted on the latex balloon, is delivered
through a long sheath introduced from the femoral vein
and pre-positioned across the VSD. In the report on the
VSD patches, the patches are pre-soaked in the patient’s
blood, which accelerates the development of adhesions,
and the duration of the traction against the patch neces-
sary to fix the patch in place is only 23 hours. The frame-
less patch has been used with reported success in a
number of patients outside of the United States
19
. This is

CHAPTER 32 Innovative, new therapeutic procedures
868
another innovative concept with some very favorable
points, but like the frameless ASD patch, the VSD patch
also requires a well controlled trial before it becomes an
accepted procedure.
Frameless PDA patch
The same concept used for the frameless ASD and VSD
patches has been applied successfully to occlude the large
patent ductus arteriosus (PDA)
20
. Like the VSD patch, the
PDA patch is pre-clotted and only requires 23–24 hours of
balloon fixation before the implanting/fixation balloons
can be removed. Also, like the other devices, the PDA
frameless patch has not been used in the United States or
in any regulated and/or controlled clinical trials.
Dr Sideris has been experimenting with several “sur-
gical glues” in addition to the pre-clotting to enhance and
accelerate fixation of the patches to the tissues, which in
turn would eliminate the undesirable relatively long
period of fixation and immobilization of the patients. The
frameless patches certainly are imaginative and possibly
indicative of the potential for transcatheter occlusion ther-
apy in the future.
The future of interventional/therapeutic
catheterizations
The various procedures and devices described in this
chapter provide a glimpse into the future of transcatheter
therapeutic procedures for pediatric and congenital heart

patients. Although all of these ideas may not evolve into
clinically useful procedures, they do illustrate the con-
tinued imagination and innovation of the current pedi-
atric/congenital interventionists and provide a challenge
for the future. Patient care certainly will continue to be
improved, not only by procedures performed by catheter-
ization rather than surgery, but also by collaborative
and/or “hybrid” surgical/catheterization procedures
21
.
References
1. Bonhoeffer P et al. Percutaneous replacement of pulmonary
valve in a right-ventricle to pulmonary-artery prosthetic con-
duit with valve dysfunction. Lancet 2000; 356(9239): 1403–1405.
2. Khambadkone S et al. Percutaneous pulmonary valve
implantation for right ventricular outflow tract lesions after
congenital heart surgery. Cardiol Young 2003; 13(Suppl 1
(Abstracts ’03 AEPC) ): 32 (Abstract #92).
3. Boudjemline Y et al. Percutaneous pulmonary valve replace-
ment in large right ventricular outflow tract: an experimental
study. Cardiol Young 2003; 13(Suppl 1 (abstracts of ’03 AEPC
meeting)): 30 (Abstract #84).
4. Boudjemline Y et al. Percutaneous pulmonary valve replace-
ment in a large right ventricular outflow tract: an experimen-
tal study. J Am Coll Cardiol 2004; 43(6): 1082–1087.
5. Boudjemline Y and Bonhoeffer P. Percutaneous implantation
of a valve in the descending aorta in lambs. Eur Heart J 2002;
23(13): 1045–1049.
6. Boudjemline Y et al. Percutaneous implantation of a biolo-
gical valve in the aorta to treat aortic valve insufficiencya a

sheep study. Med Sci Monit 2002; 8(4): BR 113–116.
7. Cribier A et al. Percutaneous transcatheter implantation of an
aortic valve prosthesis for calcific aortic stenosis: first human
case description. Circulation 2002; 106(24): 3006–3008.
8. Richens T et al. Interventional treatment of lateral tunnel
dehiscence in a total cavopulmonary connection using a
balloon expandable covered stent. Catheter Cardiovasc Interv
2000; 50(4): 449–451.
9. Konertz W et al. Modified hemi-Fontan operation and subse-
quent nonsurgical Fontan completion. J Thorac Cardiovasc
Surg 1995; 110(3): 865–867.
10. Hausdorf G, Schneider M, and Konertz W. Surgical precon-
ditioning and completion of total cavopulmonary connec-
tion by interventional cardiac catheterization: a new concept.
Heart 1996; 75(4): 403–409.
11. Diethrich EB. AAA stent grafts: current developments.
J Invasive Cardiol 2001; 13(5): 383–390.
12. Chigogidze NA, Avaliani MV, and Cherkasov VA. New per-
cutaneous technology of vascular shunting. Cardiol Young
2003. 13(Suppl 1 (abstracts of ’03 AEPC meeting)): 30
(abstract #86).
13. Mitchell MB et al. Mechanical limitation of pulmonary blood
flow facilitates heart transplantation in older infants with
hypoplastic left heart syndrome.
Eur J Cardiothorac Surg 2003;
23(5): 735–742.
14. Sigwart U. Non-surgical myocardial reduction for hyper-
trophic obstructive cardiomyopathy. Lancet 1995; 346(8969):
211–214.
15. Sideris EB et al. From disk devices to transcatheter patches:

the evolution of wireless heart defect occlusion. J Interv
Cardiol 2001; 14(2): 211–214.
16. Sideris EB et al. Transcatheter patch occlusion of experimen-
tal atrial septal defects. Catheter Cardiovasc Interv 2002; 57(3):
404–407.
17. Sideris EB et al. Transcatheter atrial septal defect occlusion
in piglets by balloon detachable devices. Catheter Cardiovasc
Interv 2000; 51(4): 529–534.
18. Sideris A et al. Transcatheter patch correction of atrial septal
defects: experimental validation and early clinical experi-
ence. Cardiol Young 2000; 10: 13.
19. Sideris EB et al. Transcatheter patch occlusion of perimem-
branous ventricular septal defects. J Am Coll Cardiol 2003;
41(6 Suppl B): 473.
20. Sideris A et al. Accelerated transcatheter patch occlusion of
large patent ductus arteriosus. Cardiol Young 2003; 13(Suppl
1 (Abstracts of ’03 AEPC)): 35 (abstract #99).
21. Bacha EA et al. New Therapeutic Avenues with Hybrid
Pediatric Cardiac Surgery. Heart Surg Forum 2004; 7(1):
33–40.