Critical Pulmonary Stenosis

LARRY A. LATSON, M.D.

From the Department of Pediatric Cardiology, Centerfor Pediatric and Congenital Heart Diseases, Cleveland Clinic Foundation,
Cleveland. Ohio

Critical pulmonary stenosis causes cyanosis and can be potentially lethal in the neonate. Initial treatment includes
general resuscitation and infusion of prostaglandin El to dilate the ductus. The diagnosis is usually made
ec~~ocardiographically, but a right ventriculogram in the outfzow tract may be necessary in some patients with only
a tiny valve opening. Prefoimed catheters may aid in the passage of an appropriate guidewire. Valvuloplasty
should be performed with a balloon approximately 1.2 times the annulus diameter. Most patients remain mildly to
moderately cyanotic immediately after the procedure. With right ventricular (RV) growth and improved RV
Compliance, the cyanosis eventually resolves. Some patients may require prolonged prostaglandin infusion, a
surgical shunt, or other mechanical means of maintaining systemic-to-pulmonary artery flow. Intermediate- to
long-term results are excellent. However, only 5%-10% of patients may require surgical relief of residual valve
or subvalvular stenosis. Very long-term follow-up raises concern about the significance of induced pulmonary
insufficiency. U p to 30% of patients may require repeat balloon valvuloplasty. (J Interven Cardiol
2001;14:345-350)

Introduction Presentation

Critical pulmonary stenosis with an intact ventricu-
lar septum is one of the classical congenital heart de-
fects that can cause cyanosis and death in the newborn.
Cyanosis is caused by right-to-left shunting at the level
of the foramen ovale due to reduced compliance of the
right ventricle. Poor compliance is secondary to severe
hypertrophy or to hypoplasia of the right ventricle.
Patients with systemic or suprasystemic right ventric-
ular (RV) pressure but no cyanosis are generally con-
sidered to have severe, rather than critical, pulmonary
stenosis. In centers with adequate expertise, percuta-
neous balloon pulmonary valvuloplasty is now consid-
ered the procedure of choice for the vast majority of
patients with critical or severe pulmonary stenosis.
The purpose of this article is to review the techniques,
immediate results, and long-term follow-up of percu-
taneous balloon valvuloplasty in neonates with this po-
tentially lethal condition.
Address for reprints: Larry A. Latson, M.D., Department of Pedi-
atric CardiologyM41, Cleveland Clinic Foundation, 9500 Euclid
Ave., Cleveland, OH 44195. Fax: (216) 445-3692.
Neonates with critical pulmonary stenosis often are
asymptomatic in the first hours to days after birth.
Left-to-right shunting through the patent ductus arte-
riosis (PDA) provides an additional source of pul-
monary blood flow in these patients with reduced pro-
grade flow through the critically stenotic pulmonary
’ valve. Under these circumstances, many patients have
relatively mild cyanosis and the condition goes unde-
tected until the PDA begins to close. If the severity of
the defect is unrecognized, patients may develop pro-
gressive cyanosis and even death as the ductal shunt-
ing progressively decreases. Initial treatment, there-
fore, includes both general resuscitation, if necessary,
and the administration of prostaglandin El to redilate
the ductus arteriosis. Approximately 5% of patients
with pulmonary atresia and intact ventricular septum
present with extreme cardiomegaly and signs of heart
failure. These patients may have extreme enlargement
of the right atrium and right ventricle, usually associ-
ated with severe tricuspid insufficiency and/or Eb-
stein’s malformation.’ In our experience, such patients
may benefit from percutaneous balloon pulmonary

Vol. 14, No. 3, 2001 Journal of Interventional Cardiology 345

 LATSON
valvuloplasty, but their subsequent clinical course is
determined more by the associated lesions than the
pulmonary stenosis and, therefore, will not be consid-
ered further in this article.
Diagnostic Assessment
The diagnosis of critical pulmonary stenosis is gen-
erally made echocardiographically. Two-dimensional
echocardiography shows a thickened and doming pul-
monary valve. Doppler echocardiography frequently
can detect a high velocity jet of flow through the
stenotic pulmonary valve opening. Differentiation of
critical pulmonary stenosis from pulmonary atresia
with intact ventricular septum can sometimes be diffi-
cult if there is a large amount of left-to-right flow
through the ductus arteriosis and only a tiny pulmonary
valve opening with a small amount of prograde flow.
Because of this difficulty, we recommend cardiac
catheterization with angiography in the RV outflow
tract in patients who probably have complete atresia but
who have otherwise favorable anatomy, even in centers
where percutaneous perforation of an atretic pulmonary
valve is not performed. A significant portion of such pa-
tients have a pinhole opening in the valve which may al-
low passage of a guidewire and performance of percu-
taneous balloon valvuloplasty.3 (Fig. 1).
Figure 1. Lateral right ventriculogram in a neonate with critical
pulmonary valve rtenosis. With the catheter in the RV outflow tract,
a pinhole jet of prograde flow is visible.
346 Journal of Interventional Cardiology
Patients with critical pulmonary stenosis frequently
have the appearance of a somewhat small right ventri-
cle. However, extreme hypoplasia of the tricuspid
valve and RV cavity is rare (< Tricuspid in-
sufficiency is common and is frequently moderate or
even severe in the presence of a normal or slightly
small right ventricle. Coronary artery to right ventricle
fistulas are rare (approximately 2%).
Balloon Valvuloplasty Technique
The first step in percutaneous pulmonary balloon
valvuloplasty is to advance the catheter to the right
ventricle for appropriate initial assessment of the pres-
sure and anatomy. Many operators prefer to advance
the catheter through the umbilical vein, if available.
We find that maneuvering is slightly better from a
femoral venous approach. In rare situations, it may be
necessary to advance a catheter from the jugular or
transhepatic routes. The initial angiogram is per-
formed best with the catheter in the RV outflow tract.
This improves the likelihood of detecting a pinhole
opening through the valve.3
Since the opening in the valve in patients with criti-
cal pulmonary stenosis is typically quite small, it is of-
ten not possible to advance a catheter directly through
the pulmonary valve, The use of preformed catheters,
such as the JR catheter, neonatal multipurpose
catheter, or cerebral catheter, will help direct a soft
guidewire to the pinhole ~ p e n i n g The
. ~ use of an
overly large or stiff guidewire increases the likelihood
of the guidewire perforating the RV outflow tract an-
teriorly rather than coursing posteriorly through the
pulmonary valve opening and toward the spine. If pos-
sible, the guidewire is directed through the ductus ar-
teriosis to the descending aorta. This provides the
straightest course with a good length of support. If the
guidewire is directed into either the right or left pul-
monary arteries, it is essential to utilize a guidewire
with the shortest floppy tip so that the stiff portion of
the guidewire can be positioned well past the pul-
monary valve annulus to provide support for the ad-
vancing catheter. Advancing the diagnostic catheter it-
self through the pulmonary valve is usually optional.
In patients with critical stenosis who do not have an
adequate PDA, the saturation may drop precipitously
if the catheter completely occludes the small opening
in the valve. However, advancing the diagnostic
catheter to the pulmonary artery may allow placement
Vol. 14, No. 3, 2001
 CRITICAL PULMONARY STENOSIS
of a different guidewire or enhance positioning of the
guidewire, if necessary.
The next step is to advance the balloon valvulo-
plasty catheter over the guidewire. The lowest profile
balloon with an adequate diameter should be utilized.
If the guidewire stability is somewhat tenuous, it may
be helpful to advance a 3-6-mm diameter coronary
balloon angioplasty catheter as the initial dilating
catheter. Many of these have excellent flexibility and a
very low profile. However, if the guidewire stability is
excellent, the valvuloplasty can be performed with a
single appropriate-sized balloon. Excellent stability
can be guaranteed by snaring the distal tip of the
guidewire in the descending aorta to create a
(Fig. 2). The one-step dilation may save time and the
expense of more than one valvuloplasty catheter.
Most authors agree that the final balloon utilized
should be 100%-130% of the pulmonary annulus di-
ameter. The annulus is best measured in very early sys-
tole from a lateral angiogram of the pulmonary out-
flow tract or from an echocardiogram illustrating the
long axis of the RV outflow tract, pulmonary valve,
and proximal main pulmonary artery. In a neonate, the
Figure 2. Lateral right ventriculogram demonstrating the trans-
ductal guidewire “rail.” A guidewire has been advanced through the
catheter in the RV outflow tract to the main pulmonary artery and
then through the ductus to the descending aorta. The guidewire has
been snared by a catheter advanced to the descending aorta from the
femoral artery. By maintaining the position of the distal end of the
guidewire, a relatively large balloon valvuloplasty catheter can be
advanced even through a tiny opening in the pulmonary valve with
the guidewire held firmly at both ends to create the ‘‘rail.’’
Vol. 14, No. 3 , 2001 Journal of Interventional Cardiology
most commonly utilized balloon sizes are 6-8 mm,
with occasional patients requiring smaller or larger
balloons. We have found that a balloon length of 2 cm
is ideal and balloons longer than 3 cm should be
avoided because of the risk of damage to the tricuspid
annulus, ductus arteriosis, or proximal pulmonary ar-
teries. If the PDA is widely open, inflation of the bal-
loon usually causes little change in the hemodynamic
status of the patient. If the ductus is closed or very
small, then the pulmonary valve blockage by the
catheter and balloon must be kept to a minimal time or
the patient can become critically cyanotic.
Immediate Results
The probability of an initial successful balloon
valvuloplasty has clearly improved over the years and
tends to be better in large-volume centers.2 Several
significant series reported before 1993, or including
fewer than 20 patients, showed failure rates of
20%~45%.’-~ In our own experience and in two more
recent series from large centers, the rate of failure to
cross the pulmonary valve and achieve an initial bal-
loon valvuloplasty is closer to 5%-10%.’0.’ I
In most instances, the RV pressure is suprasystemic
prior to balloon valvuloplasty. RV-to-systemic pres-
sure ratios of 1.3-1.5 are common. Gradients across
the pulmonary valve are difficult to interpret since fre-
quently there is only a small amount of prograde flow
through the pulmonary valve and the pulmonary artery
pressure often is variably elevated in the neonate with
a large PDA. Balloon valvuloplasty results in an im-
mediate drop in the RV pressure, but the RV-to-sys-
temic pressure ratio generally remains around
0.75-0.85.10s1I
Residual gradients across the RV outflow tract are
expected. In many cases, it is not possible to differen-
tiate between a residual gradient at the pulmonary
valve level and a gradient in the RV outflow tract due
to dynamic subvalvular stenosis. Angiography in the
RV outflow tract and use of an end-hole catheter for
withdrawal pressure measurements may be helpful in
estimating the significance of obstruction at the valve
versus subvalvular level. Residual total gradients of
20-55 mmHg from the main pulmonary artery to the
RV body are common. In our experience, if the
transvalvular gradient can be specifically measured, it
typically is only 5-10 mmHg. If the balloon has been
well-positioned and a distinct waist in the balloon has
341
 LATSON
been eliminated with inflation, repeated balloon infla-
tions or the use of significantly oversized balloons is
generally not helpful. Time is necessary to determine
whether even impressive dynamic subvalvular steno-
sis will gradually regress. Most cyanotic patients will
show a significant immediate increase in systemic sat-
uration. However, some degree of desaturation re-
mains in the majority of patients for a period of time
after the ductus arteriosis closes.
Procedural Complications
A wide variety of complications have been reported
over the years in attempted treatment of critical pul-
monary valve stenosis with balloon pulmonary valvu-
loplasty. In early reports from the 1980s, available
balloon valvuloplasty catheters were considerably
larger and stiffer than the present generation of
catheters. Vascular complications, including retro-
peritoneal hematomas and avulsions of the iliac vein,
were seen. These complications should be extremely
rare with the newer generation of valvuloplasty
catheters. Perforation of the RV outflow tract has been
reported in most large series. The frequency of this
complication seems to have decreased over the years
and should be less than 5%. Perforation of the RV out-
flow tract by a small wire may not necessarily require
surgical intervention. Perforation by larger wires or
catheters, however, is likely to result in significant
hemopericardium and tamponade. Transiently in-
creased cyanosis, hypotension, and arrhythmias dur-
ing catheter manipulations are frequently seen but not
generally reported as significant complications. Since
some patients are critically ill at presentation, it is not
unexpected that necrotizing enterocolitis and sepsis
have been seen in a small percentage of patients who
have undergone balloon valvuloplasty. Although
gentle maneuvers through the ductus arteriosis in pa-
tients on prostaglandin El seem to be well-tolerated, a
small number of patients have experienced abrupt
ductal closure." In a recent review of the data con-
tained in The Society for Cardiac Angiography and In-
terventions (SCAI) pediatric catheterization database,
there were 1,562 catheterization procedures reported.
Of these, 12 were balloon pulmonary valvuloplasties
in the neonatal age group. The procedural results were
considered to be good or excellent in all patients and
no complications were reported to have occurred dur-
ing the procedure.
348 Journal of Interventional Cardiology
Short- to Intermediate-TermResults
After successful balloon pulmonary valvuloplasty
for critical pulmonary stenosis, time is required for
restoration of several variables to near normal condi-
tions. As many as 25%-30% of patients may have
systemic or suprasystemic pressure due to dynamic
infundibular stenosis and/or elevated pulmonary artery
pressures.'' If the valve opening has been increased,
prostaglandins are generally weaned or discontinued
immediately after the catheterization as an initial trial.
Most patients with critical stenosis will continue to
have at least mild desaturation when the ductus
completely closes due to continued right-to-left atrial
level shunting. Approximately 25% will have unac-
ceptably low saturations, with ductal closure in the
first week after balloon valvuloplasty. The reinitiation
of prostaglandins has uniformly improved saturations
to an acceptable level in our experience, but rare
patients required emergency procedures because of
unresponsive rapid closure of the d u c t u ~ . Contin-
~~'~
ued right-to-left shunting is usually caused by poor RV
compliance related to RV hypertrophy or mild RV hy-
poplasia, a tricuspid insufficiency jet directed toward
the patent foramen ovale, or rarely, to a hypoplastic or
stenotic tricuspid valve.'.'
RV compliance clearly improves with time if the pul-
monary valve opening has been significantly increased.
In the majority of patients, an acceptable systemic arte-
rial saturation is achieved, with the ductus closed,
within 5-10 days after the valvuloplasty procedure. If
saturations remain unacceptably low with discontinua-
tion of prostaglandins after 1 week, a small surgical sys-
temic-to-pulmonary artery shunt can be performed. Al-
ternatively, prostaglandins can be continued, but in
some instances, the prostaglandins may be necessary
for over 1 month. In either case, eventual normal or
near-normal saturation is expected long term.
Alternative methods to avoid surgery in patients
with prolonged cyanosis have been attempted in a
small number of patients. Simple angioplasty of the
ductus arteriosis with a relatively large balloon (10
mm) was successful in maintaining ductal patency for
3 months in two patients.12Schneider et al. reported the
use of a stent to maintain ductal patency in eight pa-
tients with critical pulmonary stenosis or pulmonary
atresia. The stents were delivered to the ductus via the
venous route and dilated with a 4-mm or 6-mm diame-
ter balloon. In such patients, intimal proliferation
within the stent is commonly seen and complete spon-
Vol. 14, No. 3,2001
 CRITICAL PULMONARY STENOSIS

A B
Figure 3. (A) Lateral descending aortogram in a patient with a Palmaz stent inserted as a neonate to maintain ductal patency.
There is significant intimal proliferation narrowing the lumen at the aortic end. The patient's saturations remained excellent
when the PDA was test occluded. (B) Lateral descending aortogram after placement o f a 5-rnm Gianturco coil in the Palmaz
stenr. There is complete occlusion of the stented PDA.

taneous closure of the ductus may occur in approxi-
mately 50% of patients by 4.5-17 months." In patients
with continued significant shunting, the stent can be
coil-occluded (Fig. 3). Transcatheter closure of the
atrial septa1 defect (ASD) has been successful in elim-
inating cyanosis in one infant who did well with initial
balloon test occlusion of the ASD.I4 As availability of
transcatheter ASD occlusion devices increases, this
may become a more frequently utilized option. Some
operators have utilized beta blockers when there is ev-
idence of significant dynamic subvalvular stenosis.
Gala1 et al. found that the addition of the alpha blocker
phentolamine was helpful in two patients in whom beta
blockade alone was unsuccessful. I s
Approximately 15%-30% of patients with an ini-
tially successful balloon pulmonary valvuloplasty may
show evidence of significant residual or recurrent RV
outflow tract obstruction within 1-2 years. Repeat bal-
loon pulmonary valvuloplasty is successful in many of
these patients. However, up to 10% of patients may
eventually require surgical RV outflow tract resection
or pulmonary valve rese~tion.'.'~.' '3"
Intermediate-term follow-up has shown excellent
results in patients who underwent successful balloon
pulmonary valvuloplasty . Residual echocardiographic
gradients are nearly uniformly less than 30 mmHg."' '
Evaluations of RV size demonstrated that the right
ventricle and pulmonary valve grow at a rate that par-
allels or exceeds the rate of somatic growth.'"I8 In the
majority of patients, balloon pulmonary valvuloplasty
is the only intervention that is required. In a small num-
ber of patients, a repeat interventional catheterization
or surgical procedure may be necessary to close a per-
sistently patent PDA, surgical shunt, or ASD. We gen-
erally postpone such interventions until patients are
3-5 years old if the patient is otherwise clinically well.
Long-Term Follow-Up
Truly long-term follow-up for balloon valvuloplasty
in patients with critical pulmonary stenosis is sparse.
Small series, anecdotal reports, and the author's expe-
rience indicate that nearly all patients followed for up
to 20 years are clinically well. It is very unusual to have
further progression of pulmonary stenosis after several
years of age. Rather, the biggest concern after 5-10
years may be increasing significance of pulmonary

Vol. 14, No. 3, 2001 Journal of Interventional Cardiology 349

 LATSON
insufficiency. Moderate or even severe pulmonary in-
sufficiency tends to remain relatively asymptomatic
for many years. Progressive RV dilation is seen
echocardiographically long before the appearance of
overt symptoms. Berman et al. found that 6 of 107 con-
secutive patients developed severe pulmonary insuffi-
ciency during a mean follow-up time of 7.2 years.”
Three of these patients required pulmonary valve re-
placement and three others were likely to require pul-
monary valve replacement during childhood. The au-
thors believed that younger age, a higher degree of ini-
tial obstruction, and the use of oversized balloons were
significant contributing factors to worsened pulmonary
insufficiency. We agree with them that it is probably
prudent to use balloons closer to 1.2 times the annulus
diameter and to avoid the use of balloons greater than
1.3 times the annulus diameter. Although a small num-
ber of patients treated with slightly smaller balloons re-
quire repeat balloon pulmonary valvuloplasty, a repeat
balloon procedure is considerably better than the even-
tual need for pulmonary valve replacement.”
Summary
Balloon pulmonary valvuloplasty is accepted as the
treatment of choice for critical pulmonary valve steno-
sis in centers with significant experience. Improvements
in technique and equipment should allow for use of this
procedure in over 90% of patients in most centers. Ini-
tial transiently persistent mild-to-moderate cyanosis af-
ter a successful procedure is common. A number of op-
tions are available to patients who remain unacceptably
cyanotic more than a week after the procedure. Interme-
diate- and long-term results are excellent, with nearly all
patients eventually becoming asymptomatic. There is
growing concern about the significance of induced pul-
monary insufficiency in the very long-term follow-up,
and the trend is to recommend the use of slightly smaller
balloons than were recommended in the past because of
this potential problem.
References
1. Sommer RJ, Rhodes JF, Pamess IA. Physiology of critical pul-
monary valve obstruction in the neonate. Cathet Cardiovasc
Interven 2000;50:473479.
2. Hanley FL, Sade RM, Freedom RM, et al. and The Congenital
Heart Surgeons Society. Outcomes in critically ill neonates
350 Journal of Interventional Cardiology
with pulmonary stenosis and intact ventricular septum: A
multi-institutional study. J Am Coll Cardiol 1993;22:183-192.
3. Walsh KP, Abdulhamed JM, Tometzki AJP. Importance of
RV outflow tract angiography in distinguishing critical pul-
monary stenosis from pulmonary atresia. Heart 1997;77:
45H60.
4. Klewer SE, Zamora R. Use of a cerebral angiographic catheter
facilitates crossing the pulmonary valve in neonates with crit-
ical or severe pulmonary valve stenosis. Am J Cardiol 1997;
80:1245-1247.
5. Latson L, Cheatham J, Froemming S, et al. Transductal
guidewire “rail” for balloon valvuloplasty in neonates with
isolated critical pulmonary valve stenosis or atresia. Am J Car-
diol 1994;73:713-714.
6. Weber HS, Cyran SE. Effectiveness of an umbilical artery
“snare assisted” approach for critical pulmonary valve steno-
sis or atresia in the neonate. Am J Cardiol 1997;80:1502-1504.
7. Caspi J, Coles JG, Benson LN, et al. Management of neonatal
critical pulmonic stenosis in the balloon valvotomy era. Ann
Thorac Surg 1990;49:273-278.
8. Talsma M, Witsenburg M, Rohmer J, et al. Determinants for
outcome of balloon valvuloplasty for severe pulmonary steno-
sis in neonates and infants up to six months of age. Am J Car-
diol 1993;7l :1246- 1248.
9. Gildein HP, Kleinert S, Goh TH, et al. Treatment of critical
pulmonary valve stenosis by balloon dilatation in the neonate.
Am Heart J 1996;131:1007-1011.
10. Gournay V, Piechaud JF, Delogu A, et al. Balloon valvotomy
for critical stenosis or atresia of pulmonary valve in newborns.
JACC 1995;26: 1725-1731.
11. Colli AM, Perry SB, Lock JE, et al. Balloon dilation of critical
valvar pulmonary stenosis in the first month of life. Cathet
Cardiovasc Diagn 1995;34:23-28.
12. Walsh KP, Abrams SE, Arnold R. Arterial duct angioplasty as
an adjunct to dilation of the valve for critical pulmonary steno-
sis. Br Heart J 1993;69:26&262.
13. Schneider M, Zartner P, Sidiropoulos A, et al. Stent implanta-
tion of the arterial duct in newborns with duct-dependent cir-
culation. Eur Heart J 1998;19:1401-1409.
14. Nugent AW, Menahem S, Goh TH, et al. Device closure of an
atrial septa1 defect following successful balloon valvuloplasty
in a neonate with critical pulmonary valve stenosis and persis-
tent cyanosis. Pediatr Cardiol2000;21: 17&171.
15. Gala1 0, Kalloghlian A, Pittappilly BM, et al. Phentolamine
improves clinical outcome after balloon valvoplasty in
neonates with severe pulmonary stenosis. Cardiol Young
1999;9: 127-128.
16. Tabatabaei H, Boutin C, Nykanen DG, et al. Morphologic and
hemodynamic consequences after percutaneous balloon
valvotomy for neonatal pulmonary stenosis: Medium-tern fol-
low-up. JACC 1996;27:473478.
17. Gildein HP, Kleinert S, Goh TH, et al. Pulmonary valve annu-
lus grows after balloon dilatation of neonatal critical pul-
monary valve stenosis. Am Heart J 1998;136:276-280.
18. Kovalchin JP, Forbes TJ, Nihill MR, et al. Echocardiographic
determinants of clinical course in infants with critical and se-
vere pulmonary valve stenosis. JACC 1997;29:1095-1101.
19. Berman W Jr, Fripp RR, Raisher BD, et al. Significant pul-
monary valve incompetence following oversize balloon pul-
monary valvuloplasty in small infants: A long-term follow-up
study. Cathet Cardiovasc Interven 1999;48:6145.
20. Rao PS. Late pulmonary insufficiency after balloon dilatation
of the pulmonary valve. Letter to the Editor. Cathet Cardiovasc
Interven 1999;48:118-119.
Vol. 14, No. 3,2001