Original Article
Angioplasty of Obstructed Homograft Conduits
in the Right Ventricular Outflow Tract With
Ultra-Noncompliant Balloons
Assessment of Therapeutic Efficacy and Conduit Tears
Michael R. Hainstock, MD; Audrey C. Marshall, MD; James E. Lock, MD; Doff B. McElhinney, MD
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 16, 2017
Background—Angioplasty and stent placement in right ventricle-to-pulmonary artery (RV-PA) conduits have been shown
to prolong the functional lifespan of a conduit. Safety and efficacy of angioplasty of obstructed RV-PA homografts using
ultra-noncompliant (UNC) or ultrahigh-pressure balloons are unknown.
Methods and Results—From 2004 to 2012, 70 patients underwent 76 procedures for angioplasty of RV-PA homografts
with UNC Atlas balloons. The UNC group was compared with a partially contemporaneous control cohort of 81 patients
who underwent 84 angioplasty procedures with conventional balloons. Acute hemodynamic changes after angioplasty of
homografts with UNC balloons included significantly reduced RV:Ao pressure ratio (P=0.02) and right ventricular outflow
tract gradients (P≤0.001). Balloon waist resolution was more frequently achieved with UNC balloons (P=0.04), and balloon
rupture occurred less often (P<0.001). Conduit tears of any severity occurred in 22% of patients overall and were more common
in the UNC group (P=0.001). Patients with any conduit tear had significantly greater reduction in their RV:Ao pressure ratio
(P<0.001) and right ventricular outflow tract gradient (P=0.004) than those with no tear. There were 4 unconfined tears, all in
the UNC group, with no acute decompensations or deaths and only 1 patient who required surgical management.
Conclusions—RV-PA conduit tears are common in patients undergoing angioplasty, but clinically important tears, which only
occurred during UNC angioplasty in this series, were uncommon. UNC balloons can be used to good effect with significant
reduction in right ventricular outflow tract gradient and the RV:Ao ratio when compared with conventional balloons. (Circ Cardiovasc Interv. 2013;6:00-00.)
Key Words: angioplasty ◼ conduit ◼ ventricular outflow obstruction
R
ight ventricle-to-pulmonary artery (RV-PA) homograft
conduits are widely used in surgical palliation of various forms of congenital heart disease, but they have a variable
lifespan before conduit dysfunction attributable to stenosis,
regurgitation, or both. Percutaneous techniques including balloon dilation, endovascular stenting, and transcatheter pulmonary valve (TPV) placement have been used to prolong
the functional lifespan of homograft conduits in an effort to
reduce the number of surgical conduit replacements during a
patient’s lifetime.1–6 With the advent of TPV replacement, it
will be important to achieve a better understanding of RV-PA
conduit angioplasty, including technical factors and outcomes.
Angioplasty and bare metal stent placement have become
important components of conduit preparation for a TPV7;
however, the potential for conduit disruption is an important
concern that is not well understood.8,9
Effective balloon dilation of obstructed conduits with conventional balloons has been shown to be limited because of
inability to resolve waists at rated burst pressures, balloon rupture by calcified conduits, and rupture before waist resolution
at or above rated burst pressures.5,6,10 Placement of balloon
expandable stents is reportedly superior to simple conduit
angioplasty in terms of reducing obstruction and prolonging
conduit lifespan.1,2 In the current era, a variety of angioplasty
balloons are available with differing characteristics related to
size, length, compliance, profile, and rated burst pressure, and
earlier experience may not reflect contemporary options. In
particular, material technological advances have introduced
ultra-noncompliant (UNC) angioplasty balloons that are layered with woven ultrahigh molecular weight polyethylene and
have essentially no distensibility and rated burst pressures as
high as 27 atmospheres (atm). These balloons were initially
developed for angioplasty of resistant stenosis in prosthetic
renal dialysis fistulas, for which they have shown success.11
In a previous study of our early experience using UNC balloons to dilate resistant branch PA stenosis within or near a
Received December 16, 2012; accepted October 7, 2013.
From the Department of Pediatric Cardiology, Children’s Hospital Boston, MA.
Guest Editor for this article was Lee Benson, MD.
The online-only Data Supplement is available at http://circinterventions.ahajournals.org/lookup/suppl/doi:10.1161/CIRCINTERVENTIONS.
112.000073/-/DC1.
Correspondence to Michael R. Hainstock, MD, Pediatric and Congenital Interventional Cardiology, Division of Pediatric Cardiology, University of
Virginia Health System, PO Box 800386, Charlottesville, VA 22908. E-mail [email protected]
© 2013 American Heart Association, Inc.
Circ Cardiovasc Interv is available at http://circinterventions.ahajournals.org
1
DOI: 10.1161/CIRCINTERVENTIONS.112.000073
2 Circ Cardiovasc Interv December 2013
WHAT IS KNOWN
• Angioplasty and stent placement in right ventricle to
pulmonary artery conduits has been shown to prolong the functional lifespan of a conduit.
WHAT THE STUDY ADDS
• Conduit
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 16, 2017
tears are significantly more common after
dilation with ultra-noncompliant balloons compared
with conventional balloons.
• Conduits with tears after angioplasty have a significantly greater reduction in right ventricle to pulmonary artery gradient and RV:Ao ratio than those
without tears.
• Clinically important tears were infrequent, and conduit rupture with acute decompensation was not
observed in this study.
• No therapeutic or confined tear progressed to an
unconfined tear.
previously placed stent, we found this approach to be highly
effective and generally safe.12 However, there are no published
data specifically concerning UNC balloons for angioplasty of
RV-PA conduits. Anecdotally, there is concern among some
interventional cardiologists about the safety of using high
inflation pressures to dilate RV-PA homograft conduits, which
are often calcified and do not maintain the mural architecture
of living vessels, and as a result may not behave like typical
arteries in response to angioplasty.13–15 Since 2004, we have
used UNC balloons for angioplasty of RV-PA conduits. In this
study, we review that experience, with a focus on efficacy,
safety, and factors associated with conduit tears.
Methods
Patients
The computerized database of the Cardiovascular Program at
Children’s Hospital Boston was queried for patients who underwent
balloon dilation of an RV-PA conduit with an Atlas balloon (Bard
Peripheral Vascular, Inc, Tempe, AZ) between 2004 and 2012. A control cohort was derived comprising patients who underwent RV-PA
conduit dilation between 2000 and 2012 with a conventional balloon
(ie, any non-UNC balloon) measuring ≥12 mm in diameter (12 mm
is the smallest diameter Atlas balloon). Patients in whom both UNC
and conventional balloons were used (almost always conventional
followed by UNC) were included in the UNC group. We excluded
patients who did not have a fully circumferential homograft conduit
(including those with a homograft conduit that had been surgically
augmented, a nonhomograft conduit, or no homograft at all), those
who underwent PA dilation with an UNC balloon without conduit
dilation, those who had conduits of different sizes anastomosed in series, and those in whom the largest angioplasty balloon was <12 mm.
Patients who underwent TPV placement were also excluded because
it was thought that intention to place or placement of the valve, which
is effectively a covered stent, might confound our results. Patients
who had a bare metal stent implanted after angioplasty were included.
Catheterization and Angioplasty
Demographic and hemodynamic data were obtained from catheterization reports, clinic notes, and surgical reports. When available,
postangioplasty pressure measurements were recorded after balloon
dilation but before any stent implantation (if performed). When this
information was not available, poststent pressure measurements were
used. Types and sizes of balloons were recorded, but inflation pressures were not always available.
Standard angioplasty techniques were used as described previously.16,17 Balloon choice was at the discretion of the operator. In general,
we worked from our previously cited guideline of limiting the maximum balloon diameter to no >110% of the nominal conduit diameter.2 Multiple dilations were often performed with gradual increases
in balloon inflation pressure and size, with interval angiograms to
assess for changes, including tears. When tears were identified, additional angiograms were performed to characterize the extent of the
disruption more fully. Decisions about whether and when to implant
a bare metal stent were at the discretion of the operator.
Angiographic Assessments
Angiograms and balloon inflations were imaged and acquired using
a biplane system. Angiographic and fluoroscopic measurements were
performed offline by a single investigator using digital imaging archives. Measurements were performed for both the first and final angioplasty balloons, and included diameters of the conduit and balloon
waists. The waist:nominal balloon diameter ratio was calculated as
the ratio of the first discretely identifiable balloon waist to the nominal diameter of the balloon. Successful resolution of a balloon waist
was defined as a linear border without perceivable indentation along
both edges of the balloon observed on both projections. Inflations that
were intentionally halted before full resolution of a waist could not
always be discriminated from those in which attempted waist resolution was not successful. The ratio of the nominal balloon diameter to
the nominal conduit diameter (balloon:conduit diameter ratio) was
also calculated for the first and final balloons.
Conduit tears were classified into 4 categories: (1) no tear, (2)
therapeutic tear (contrast extending <3 mm from conduit lumen;
Figures 1 and 2), (3) confined tear (contrast extending >3 mm from
conduit lumen; Figure 3), and (4) unconfined tear (contrast extravasation identified in the pericardial or pleural space or the airways;
Movie in the online-only Data Supplement). Confined tears were further subdivided into lagoon shaped (contrast elongating parallel to the
conduit lumen) or estuary shaped (contrast elongating perpendicular
to the conduit lumen) drawing from marine science descriptions.18
Conduit calcification was graded as none, mild (minimally visible
along a portion of the conduit), or severe (heavy, visible along entire
conduit).7 The relationship of the conduit to the anterior chest wall
or sternum was recorded as remote, partially apposed (<50% of conduit length apposed), or substantially apposed (>50% apposed).7 No
routine or standardized follow-up imaging was performed to evaluate
tears after the catheterization procedure.
Data Analysis
The primary outcome assessed was presence or absence of a conduit tear after balloon dilation of the conduit. Secondary outcomes
included change in the peak RV-PA pressure gradient, RV pressure,
and RV:aortic pressure ratio. Treatment for conduit tears and other
procedural complications are reported descriptively. Variables analyzed for association with outcomes included balloon type (UNC
versus conventional), age, conduit size, type of homograft (aortic
versus pulmonary), severity of conduit calcification, conduit relationship to the chest wall, diameter of the initial and final balloons, and
balloon:conduit and waist:balloon diameter ratios. For comparison of
categorical variables, Fisher exact test or χ2 analysis were used, and
for comparison of continuous data, independent-sample t test (if normally distributed) or the Wilcoxon rank-sum test was used. ANOVA
with Bonferroni post hoc testing was used to compare continuous
data between >2 groups (eg, types of tears). Data are presented as
mean±SD, median (minimum-maximum), or frequency (%). We
compared differences between preintervention and postintervention
measurements among groups (when >2 groups) with ANOVA. To account for multiple comparisons, significance was set at P≤0.01.
Hainstock et al Angioplasty of Obstructed Homograft Conduits 3
was associated with a significantly lower RV:Ao pressure ratio
and RV-PA pressure gradient after angioplasty (Figure 4).
Conduit Tears
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 16, 2017
Figure 1. Example of therapeutic tear: contrast extending <3 mm
from conduit lumen.
The study was approved by the Children’s Hospital Committee
for Clinical Investigations. The authors had full access to the data
and take full responsibility for its integrity. All authors have read and
agree to the article as written.
Results
Patients
From July 2000 to January 2012, 70 patients meeting inclusion criteria underwent 76 conduit angioplasty procedures in
which UNC balloons were used (6 patients underwent 2 separate UNC balloon procedures on different conduits), and 81
patients underwent 84 procedures in which only conventional
balloons were used (3 patients underwent 2 separate procedures on different conduits). Demographic, diagnostic, and
procedural variables are summarized in Table 1. UNC balloon
patients were slightly older than those in the conventional
group, as was the age of the conduit in UNC patients (both
P=0.08), but all other recorded demographic and preangioplasty diagnostic variables were similar between groups.
Angioplasty Outcomes
Data relating to the angioplasty procedure are summarized
in Table 1. There was no difference in the initial or final
waist:balloon diameter ratios between groups, but the initial
and final balloon:conduit diameter ratios were significantly
higher in the UNC balloon group. Among patients in the UNC
balloon group, multiple different diameter balloons were used
in 35 patients (46%), and a 4 cm long balloon was used in
24 patients (31%). The balloon waist was resolved in 65%
of UNC procedures, which was significantly higher than in
the conventional balloon group. Balloon rupture/puncture
occurred significantly more often with conventional than
UNC balloons. All 3 of the patients in whom the UNC balloon
ruptured had severely calcified conduits; the balloon rupture
in all of these cases seemed to be a point puncture with slow
contrast leakage before full inflation (ie, not at high pressure)
rather than an explosive rupture.
Hemodynamic Outcomes
Overall, both groups experienced a reduction in RV pressure
and the peak RV-PA pressure gradient. Use of UNC balloon
Data related to conduit tears are summarized in Table 2. A
conduit tear was identified angiographically in 25 patients in
the UNC balloon group and 10 in the conventional balloon
group (P=0.001). The shape and location of tears varied considerably, and no clear patterns were discerned. There were 4
unconfined tears in total, all of which occurred after UNC balloon angioplasty, none of which occurred as a frank conduit
rupture with acute decompensation. Confined tears were similarly more common in the UNC balloon group (P=0.009). In
no case did a tear first characterized as therapeutic or confined
progress to become unconfined.
Demographic, diagnostic, and procedural variables of
patients who did and did not experience a conduit tear are
summarized in Table 3. There were significantly greater
reductions in the RV:Ao pressure ratio and the peak RV-PA
pressure gradients in patients who had any conduit tear in
comparison with those with no tear, although there were no
significant differences in initial or final balloon:conduit ratio
or waist:balloon ratio. The initial waist:balloon diameter ratio
was smaller (ie, tighter waist) in patients who developed
any tear than those who did not. Patients who experienced
a confined or unconfined tear (combined) had significantly
greater reduction in the RV:Ao pressure ratio than patients
with either no tear or a therapeutic tear only (−0.35±0.28 versus −0.19±0.19; P=0.001), but there were no differences in
other hemodynamic or balloon related parameters, including
waist:balloon diameter ratios (Figure 5).
Unconfined Tears
Three of the 4 unconfined tears were recognized during the
procedure, and none occurred as frank conduit rupture with
acute decompensation. One of the cases recognized during the
procedure had a bare metal stent implanted across the tear,
and the patient was observed without hemodynamic compromise or enlargement of the tear. Two patients were observed
in the catheterization laboratory (for ≤2 hours) without hemodynamic change or enlargement of the tear, 1 of whom was
subsequently observed in the cardiac intensive care unit, and
neither underwent further intervention. The fourth patient
became hemodynamically unstable in the recovery area after
the procedure and was subsequently found to have hemothorax and an unconfined tear. The patient was returned to the
catheterization laboratory, where chest tubes were placed
and resuscitation was performed. Extracorporeal membrane
oxygenation support was initiated in the catheterization laboratory, and the patient was then taken emergently to the operating room for conduit replacement. The postoperative course
was uncomplicated, and there were no discernible neurological or other complications related to the period of hemodynamic compromise.
Of the 4 patients who developed unconfined tears after UNC
angioplasty, 2 had truncus arteriosus, 1 had double-outlet RV
with pulmonary atresia, and 1 had l-transposition of the great
arteries with pulmonary atresia. Three of the 4 patients had
an aortic homograft conduit that was substantially apposed
4 Circ Cardiovasc Interv December 2013
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 16, 2017
Figure 2. Serial images demonstrating (A) predilation, (B) balloon dilation, and (C) postdilation
­angiograms resulting in a posterior therapeutic
tear (arrow).
to the anterior chest wall. The conduits ranged in implanted
diameter from 12 to 20 mm, and 3 conduits (2 aortic, 1 pulmonary) were severely calcified. The mean initial and final
balloon:conduit diameter ratios were 0.86 and 0.93, respectively, and the mean initial and final waist:balloon diameters
ratios were 0.76 and 0.82. The initial waist:balloon diameter
ratio was ≤0.78 in all 4 patients, which was lower than the
overall population mean (and median) of 0.81, but this did not
reach statistical significance (P=0.12). In 2 of these patients, a
4 cm long UNC balloon was used, and in 2, only a single UNC
balloon was inflated. In 3 of the 4 patients, the balloon waist
was resolved on the dilation that produced the tear.
Discussion
RV-PA Conduit Dilation With
Ultra-Noncompliant Balloons
In our practice, UNC balloons have become a common tool for
angioplasty of obstructed RV-PA homograft conduits. They tend
to be effective at relieving stenoses, similar to their performance
Hainstock et al Angioplasty of Obstructed Homograft Conduits 5
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 16, 2017
Figure 3. Example of confined tears. A, Confined
lagoon tear: contrast extending >3 mm from conduit lumen and elongating parallel to the lumen
axis. B and C, Confined estuary tears: contrast
extending >3 mm from conduit lumen and elongating perpendicular to the lumen axis (images from
different patients).
in resistant PA lesions12 but result in a substantially higher frequency of angiographically identifiable conduit wall tears than
lower pressure balloons. In this study, there were a total of 4
unconfined tears, all in patients who underwent dilation with
UNC balloons. Three of these were in heavily calcified conduits that were closely apposed to the anterior chest wall, and
the waist on the initial dilating balloon was tighter than in other
patients (≤0.78 in all 4), although the small number of events
precluded robust analysis of risk factors for an unconfined tear.
In light of these findings, we think it is important to perform
angioplasty with an initial waist:balloon diameter ratio ≥0.8. If
the initial balloon waist is smaller (ie, tighter), it may be prudent
6 Circ Cardiovasc Interv December 2013
Table 1. Demographic, Diagnostic, and Procedural Variables in Patients Who Underwent RV-PA
Conduit Angioplasty With Ultra-Noncompliant and Conventional Balloons
Variable
UNC (n=76)
Conventional (n=84)
P Value
Patient age, y
13.8±10.7
11.2±8.0
0.08
Implanted conduit diameter, mm
16.1±4.2
16.5±3.8
0.5
7.9±5.0
6.5±4.3
0.08
Aortic homograft conduit
44 (58%)
55 (66%)
0.3
Conduit severely calcified
49 (65%)
55 (66%)
0.9
Conduit substantially apposed to anterior chest wall
32 (43%)
36 (43%)
0.9
RV:aortic pressure ratio preintervention
0.83±0.20
0.85±0.20
0.4
Age of implanted conduit, y
RV:aortic pressure ratio postintervention
0.59±0.16
0.66±0.20
−0.25±0.20
−0.19±0.23
RVOT gradient preintervention, mm Hg
49±21
55±20
0.08
RVOT gradient postintervention, mm Hg
28±15
39±22
<0.001
Change in RV:aortic pressure ratio after intervention
0.02
0.1
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 16, 2017
Change in RVOT gradient after intervention, mm Hg
−20±19
−15±22
0.1
Initial balloon:nominal conduit diameter ratio
0.95±0.34
0.83±0.24
0.01
Final balloon:nominal conduit diameter ratio
1.01±0.35
0.88±0.23
0.004
Initial waist:balloon diameter ratio
0.81±0.11
0.81±0.11
0.8
Final waist:balloon diameter ratio
0.82±0.08
0.81±0.10
0.2
Balloon waist resolved
49 (65%)
41 (48%)
0.04
3 (4%)*
31 (37%)
<0.001
34 (45%)
56 (67%)
0.005
Balloon rupture during conduit dilation
Bare metal stent implanted in conduit after angioplasty
during the same procedure
Data presented as mean±SD or frequency (% of column total), unless otherwise specified. Aortic homografts were
significantly more likely to have severe calcification than pulmonary homografts (88% vs 28%; P<0.001). RV-PA indicates
right ventricle-to-pulmonary artery; RVOT, right ventricular outflow tract; and UNC, ultra-noncompliant.
*In 7 other patients in the UNC group, a conventional balloon used during the same procedure ruptured.
to halt the inflation and begin with a smaller balloon, particularly when the conduit is heavily calcified or closely apposed
to the anterior chest wall. These recommendations are only
loosely based on data, as the number of unconfined tears in this
series was small, and the degree of calcification or location of
the conduit was not associated with the development of tears
in general (ie, any type of tear). We do not interpret these findings to suggest that UNC balloons should not be used for right
ventricular outflow tract conduit angioplasty, but rather that is
important to use them cautiously and appropriately.
Figure 4. Predilation and postdilation right ventricular outflow tract
(RVOT) gradient. There is a significantly greater decrease in RVOT
gradient with ultra-noncompliant (UNC) balloons where tears were
identified relative to dilations with conventional balloons.
Our definition of unconfined conduit tear, which was intended
to reflect tears that bled into the airway or a large mediastinal
space and might result in important consequences, was liberal
in this study. Namely, only 1 of the 4 cases resulted in significant extravascular fluid collection and hemodynamic instability.
The tear in this patient was not recognized in the catheterization
laboratory and was diagnosed only after hemodynamic decompensation in the recovery room, with subsequent resuscitation
using mechanical circulatory support and surgical conduit
replacement. This case highlights the importance of thorough
angiographic evaluation after balloon angioplasty of conduits.
Additional invasive procedures (ie, surgery, covered stent placement, closure of the tear with a device, or drainage of extravascular blood) were not performed in any of the other 3 cases that
met our definition of unconfined tear. We deliberately avoided
the term conduit rupture, and the frequency of unconfined tears
in this series should not be quoted as the incidence of conduit
rupture without clearly stating that only 1 of our 4 cases had
serious implications. Frank conduit rupture with acute hemodynamic decompensation, which is the most concerning scenario
that interventionalists imagine when considering the risk of
conduit injury, did not occur in this series.
In general, UNC balloons, including 4 cm long balloons, were
used to good effect in this population, with few serious adverse
events. Although tears in general and all of the unconfined tears
were in patients who underwent angioplasty using a UNC balloon, it is not necessarily the case that such balloons are more
likely to cause important tears than other types of balloons.
Hainstock et al Angioplasty of Obstructed Homograft Conduits 7
Table 2. Details of Conduit Tears in the UNC and
Conventional Balloon Groups
Variables
Any conduit tear*
Therapeutic tear
UNC
(n=76)
Conventional
(n=84)
P Value
25 (33%)
10 (12%)
0.005
6 (8%)
5 (6%)
15 (20%)
5 (6%)
4 (5%)
0 (%)
Lagoon shaped
6 (40%)
5 (100%)
Estuary shaped
9 (60%)
0 (0%)
Anterior
8 (32%)
1 (10%)
Posterior
5 (20%)
4 (40%)
Left
9 (36%)
4 (40%)
Right
3 (12%)
1 (10%)
Confined tear
Unconfined tear
with UNC balloons tended to result in greater reductions in the
RV:aortic pressure ratio and RV-PA pressure gradient.
Shape of confined tear†
0.04
Location of tear†
0.4
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 16, 2017
UNC indicates ultra-noncompliant.
*Percentages expressed as percentage of patients in the UNC or conventional
balloon groups; 4×2 comparison (no tear, therapeutic, confined, unconfined)
between balloon groups using χ2 analysis.
†Percentages expressed as percentage of confined tears in the UNC or
conventional balloon groups; 4×2 comparison (anterior, posterior, left, right)
between balloon groups using χ2 analysis.
There were other procedural differences that may have been
equally important; namely, the initial and final balloon:nominal
conduit diameter ratios were significantly higher in patients
who underwent UNC balloon angioplasty than those in the
conventional balloon group, reflecting a tendency toward more
aggressive dilation. Accordingly, compared with conventional
balloons, we found that balloon dilation of RV-PA homografts
Conduit Tears
Angiographically visible conduit tears after conduit angioplasty were common in this series, particularly when UNC
balloons were used, particularly confined and unconfined
tears (as opposed to therapeutic tears). The appropriateness
of our a priori categorization of tears is unclear. As mentioned
above, unconfined tears were anticipated to reflect that the tear
would result in hemodynamic or ventilatory consequences,
but this did not always prove to be the case. Also, our distinction between therapeutic and confined tears was based on
a somewhat arbitrary angiographic discrimination, and it is
unclear whether there is any clinical or mechanistic difference between them. It is possible that a wider definition of
therapeutic tear would have been more appropriate, as none of
the therapeutic or confined tears were observed to progress to
unconfined tears or to have other obvious clinical implications.
Studies have shown that the mechanism of action of balloon dilation of vascular stenoses is the production of tears in
the vessel intima and media.19,20 However, in situ homograft
conduits have distinctive features that are different from viable vessels, and the response to angioplasty is likely to d­ iffer
as well. For example, studies of explanted cryopreserved
allograft valves show the cellular makeup being composed
primarily of collagen, fibroblasts, and calcification with a paucity of viable cuspal cells.13 No similar pathological studies
have been published to elucidate the effects of angioplasty on
RV-PA conduits, but a similar mechanism of effective angioplasty as for vascular lesions would presumably result in the
creation of a conduit tear within/through the remodeled tissue
Table 3. Diagnostic and Procedural Variables in Patients Who Did and Did Not Develop a
Visible Tear in the Conduit After RVOT Conduit Angioplasty With Ultra-Noncompliant and
Conventional Balloons
Variables
Any Conduit Tear (n=35)
No Tear (n=125)
P Value
17.1±2.9
16.1±4.1
0.16
8.2±5.1
6.9±4.6
0.13
Aortic homograft conduit*
23 (66%)
76 (61%)
0.6
Conduit severely calcified
26 (74%)
78 (62%)
0.2
Conduit substantially apposed to anterior chest wall
13 (37%)
55 (44%)
0.5
RV:aortic pressure ratio preintervention
0.92±0.23
0.82±0.18
0.01
Implanted conduit diameter, mm
Age of implanted conduit, y
RV:aortic pressure ratio postintervention
Change in RV:aortic pressure ratio after intervention
0.59±0.19
0.64±0.19
−0.33±0.27
−0.19±0.18
0.2
<0.001
RVOT gradient preintervention, mm Hg
59±21
51±21
0.07
RVOT gradient postintervention, mm Hg
31±21
35±19
0.3
Change in RVOT gradient after intervention, mm Hg
−26±21
−15±19
0.004
Initial balloon:nominal conduit diameter ratio
0.86±0.21
0.89±0314
0.5
Final balloon:nominal conduit diameter ratio
0.94±0.18
0.95±0.33
0.9
Initial waist:balloon diameter ratio
0.77±0.10
0.82±0.10
0.02
Final waist:balloon diameter ratio
0.80±0.09
0.82±0.09
0.2
Balloon waist resolved
22 (63%)
68 (55%)
0.4
Balloon rupture during conduit dilation
10 (29%)
26 (21%)
0.03
RV indicates right ventricle; and RVOT, right ventricular outflow tract.
*All other conduits were pulmonary homografts.
8 Circ Cardiovasc Interv December 2013
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 16, 2017
Figure 5. Predilation right ventricular outflow tract (RVOT) gradient
(mm Hg) related to initial waist-to-balloon ratio. Waist resolution was
unlikely at waist-to-balloon ratio <0.75 and more likely when >0.8.
layers, and extension into the perihomograft scar tissue would
permit visualization of a tear. Absence of an angiographically
visible tear does not necessarily imply that the conduit was
unaffected by the angioplasty, and that there was no therapeutic effect. However, it seems likely that a higher frequency
of visible tears with UNC balloons is consistent with more
frequent and extensive disruption of the conduit wall, which is
an important component of relieving the obstruction.
In this study, we found that when a tear was created, it was
associated with a greater hemodynamic benefit (ie, reduction
of the RV:Ao pressure ratio and right ventricular outflow tract
gradient). We also found that tears were more common with
UNC balloons than conventional balloons, which we presume
contributes to or is related to more effective relief of stenosis
in many cases. There were no obvious conduit-related factors
(eg, calcification, ­aortic versus pulmonary origin, duration in
situ, etc) associated with a higher probability of tears in general, although 3 of the 4 unconfined tears were in severely
calcified aortic homograft conduits.
Tears can be managed but must be well defined. Our practice is
to use multiple views to characterize the location, size, shape, and
hemodynamic effect of the tear, including evidence of pericardial, pleural, or airway extension. Although we chose to exclude
patients who were catheterized for purposes of TPV placement,
these same patients are likely to derive benefit from near complete relief of RV outflow tract obstruction before valve implant.
Also, the nature of the TPV (ie, a covered stent) is an obvious but
unproven option for management of important tears. Nonvalved
covered stents of sufficient size to treat conduit tears are not currently approved for use in the United States, but it is reasonable to
presume that if and when such devices become available that they
will afford another management option for conduit tears.
Balloon Rupture
Balloon rupture may occur with particularly high frequency
during angioplasty of RV-PA conduits, which often have extensive and irregular calcification. There are several mechanisms
by which balloon rupture can complicate a dilation procedure,
including fragment embolization, air embolization, inability to resheath the balloon, and consequent vascular damage
during extraction, and possibly local vessel trauma related
to high-velocity extrusion of the fluid used to inflate the balloon. The propagation of balloon rupture in a UNC balloon is
mechanistically unique from conventional balloons. Although
conventional balloons tend to rupture abruptly in circumferential or tangential directions, UNC balloon ruptures seem to
be characterized by nonexplosive, focal/point punctures with
subsequent leaking of the contrast and saline mixture into the
surrounding vascular space. Balloon rupture occurred infrequently with UNC balloons (4%) and significantly less often
than with conventional balloons, which ruptured in more than
one third of procedures. Our data suggest that use of UNC balloons may reduce the potential for such complications. Future
studies will be needed to evaluate the safety and efficacy of
stent placement with UNC balloons in relation to the occurrence of balloon rupture.
Theoretical Advantages and Disadvantages
of UNC Balloons
A theoretical advantage of UNC balloons for angioplasty of
RV-PA conduits is that UNC balloons do not expand beyond
their nominal size even at high inflation pressures or when dilating a highly resistant lesion, unlike most other high-pressure
balloons. This ensures that the actual balloon diameter does
not exceed the intended diameter. Also, as discussed above,
UNC balloons seem to rupture less often and less explosively
than many other balloons in the difficult angioplasty substrate
of an RV-PA conduit. There are also potential disadvantages
of UNC balloons. For example, the Atlas balloon has a long
taper, which can be awkward in curved or complex geometries, as is often the case with RV-PA conduits although newer
generation balloons (Atlas Gold) have made improvements in
this design limitation. The length and high-pressure inflation
mandate close observation for vascular trauma related to the
extensive straightening of the conduit and migration of the
balloon tip that can occur during inflation. Anecdotally, some
interventional cardiologists have expressed concern about the
risk of RV-PA conduit angioplasty using high inflation pressures. Although high-pressure inflation intuitively may seem
riskier than lower pressure inflation, it is important to recognize that the high pressure to which UNC balloons can be
inflated does not necessarily impart a higher risk if the balloon
is sized appropriately. Namely, using a computerized model
of balloon dilatation, Capelli et al21 showed that balloon diameter had a greater impact on the expansion force than inflation pressure, a finding that should provide reassurance about
high-pressure angioplasty, per se, and that also highlights the
importance of appropriate selection of balloon diameter.
Limitations
This study is limited by its retrospective, nonrandomized design.
Also, there are no standard definitions of types of tears, and we
gauged our classification scale relative to estimated homograft
wall thickness, with further characterization borrowed from
marine science definitions. There was no standard practice or
criteria for balloon selection. Balloon inflation pressures were
not always available, so our results pertain to the use of UNC
Hainstock et al Angioplasty of Obstructed Homograft Conduits 9
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 16, 2017
balloons, not necessarily ultrahigh-pressure dilations, which
is an important limitation. It is worth noting that, in general,
when UNC balloons were used, it was the intention to inflate
to high pressure if necessary, but sometimes the desired result
was achieved before reaching high pressure. Some conduits
already had stents in place from a previous procedure, which
may alter the integrity of a conduit and hemodynamic results of
angioplasty. Also, balloon waist resolution may not have been
achieved because of purposeful deflation once a conduit was
seen to crack, displace, or a waist seemed to pop without complete resolution. We did not capture these fluoroscopic triggers
that may have been important to the outcome or the performance
of the procedure (eg, cessation of balloon inflation before waist
resolution), such as fracture of calcified sections of the homograft or other clear signs of therapeutic effect (eg, pop in the
balloon waist without complete resolution), conduit geometry,
etc. Many patients underwent placement of a bare metal stent
after angioplasty, so the final hemodynamic results were often
better than reported for the postangioplasty condition. However,
postangioplasty hemodynamic measurements were not always
available in the catheterization record, so poststent hemodynamic measurements were used in some cases, which may
overestimate the hemodynamic changes resulting from dilation.
The decision about when to proceed to stent placement was discretionary and may have varied according to balloon type and
era, but we were not able to assess such variability in a robust
­manner because all patients did not undergo stent implantation.
Repeated procedures on different conduits in 9 patients might
induce correlated data, but theses made up a small fraction of
total observations (18/160), so we analyzed them as independent
observations. No routine or standardized follow-up imaging was
performed to evaluate tears after the catheterization procedure,
so it is not known how tears evolved or healed, but we are not
aware of any sudden unexplained deaths or clinical events suggestive of hemorrhage in this cohort. Given the infrequency of
unconfined tears, the sample size of this study was not sufficient
to detect patient-related, or conduit-related, or procedural variables that may be associated with a higher risk of this outcome.
Conclusions
In this study, we found that UNC balloons can be used to good
effect with significant reduction of right ventricular outflow tract
gradient and the RV:Ao ratio when compared with conventional
balloons. Morbidity between the 2 groups was similar despite
more aggressive therapy. Conduit tears were common overall,
particularly among patients undergoing angioplasty with a UNC
balloon, but clinically important tears were infrequent, and frank
conduit rupture with acute decompensation did not occur. Careful
and gradual dilation of conduits with adequate interval angiography for assessment of conduit tears is advisable, and when UNC
balloons are used, we recommend that an initial waist:balloon
ratio ≥0.80. Further studies will be needed to potential risk factors associated with unconfined tears and the efficacy of covered
stents (with or without a pulmonary valve) and other methods of
treatment in such cases. With the proliferation of TPV therapy,
important conduit tears are likely to become a more prevalent
problem, even if still rare. Hopefully, the data presented in this
report will contribute to the evolving understanding of how to
achieve safe and effective outcomes with TPV replacement.
Disclosures
None.
References
1. Peng LF, McElhinney DB, Nugent AW, Powell AJ, Marshall AC, Bacha EA,
Lock JE. Endovascular stenting of obstructed right ventricle-to-pulmonary
artery conduits: a 15-year experience. Circulation. 2006;113:2598–2605.
2. Powell AJ, Lock JE, Keane JF, Perry SB. Prolongation of RV-PA conduit
life span by percutaneous stent implantation. Intermediate-term results.
Circulation. 1995;92:3282–3288.
3. McElhinney DB, Hellenbrand WE, Zahn EM, Jones TK, Cheatham JP,
Lock JE, Vincent JA. Short- and medium-term outcomes after transcatheter pulmonary valve placement in the expanded multicenter US melody
valve trial. Circulation. 2010;122:507–516.
4. Sanatani S, Potts JE, Human DG, Sandor GG, Patterson MW, Gordon
Culham JA. Balloon angioplasty of right ventricular outflow tract conduits. Pediatr Cardiol. 2001;22:228–232.
5. Sreeram N, Hutter P, Silove E. Sustained high pressure double balloon
angioplasty of calcified conduits. Heart. 1999;81:162–165.
6. Zeevi B, Keane JF, Perry SB, Lock JE. Balloon dilation of postoperative
right ventricular outflow obstructions. J Am Coll Cardiol. 1989;14:401–408,
discussion 409.
7. McElhinney DB, Cheatham JP, Jones TK, Lock JE, Vincent JA, Zahn EM,
Hellenbrand WE. Stent fracture, valve dysfunction, and right ventricular
outflow tract reintervention after transcatheter pulmonary valve implantation: patient-related and procedural risk factors in the US Melody Valve
Trial. Circ Cardiovasc Interv. 2011;4:602–614.
8. Zahn EM, Hellenbrand WE, Lock JE, McElhinney DB. Implantation of
the melody transcatheter pulmonary valve in patients with a dysfunctional
right ventricular outflow tract conduit early results from the US Clinical
trial. J Am Coll Cardiol. 2009;54:1722–1729.
9. Kostolny M, Tsang V, Nordmeyer J, Van Doorn C, Frigiola A, Khambadkone
S, de Leval MR, Bonhoeffer P. Rescue surgery following percutaneous pulmonary valve implantation. Eur J Cardiothorac Surg. 2008;33:607–612.
10. Sanatani S, Potts JE, Human DG, Sandor GG, Patterson MW, Gordon
Culham JA. Balloon angioplasty of right ventricular outflow tract conduits. Pediatr Cardiol. 2001;22:228–232.
11. Trerotola SO, Stavropoulos SW, Shlansky-Goldberg R, Tuite CM, Kobrin
S, Rudnick MR. Hemodialysis-related venous stenosis: treatment with
ultrahigh-pressure angioplasty balloons. Radiology. 2004;231:259–262.
12. Maglione J, Bergersen L, Lock JE, McElhinney DB. Ultra-high-pressure
balloon angioplasty for treatment of resistant stenoses within or adjacent
to previously implanted pulmonary arterial stents. Circ Cardiovasc Interv.
2009;2:52–58.
13. Mitchell RN, Jonas RA, Schoen FJ. Pathology of explanted cryopreserved
allograft heart valves: comparison with aortic valves from orthotopic heart
transplants. J Thorac Cardiovasc Surg. 1998;115:118–127.
14.Mitchell RN, Jonas RA, Schoen FJ. Structure-function correlations in
cryopreserved allograft cardiac valves. Ann Thorac Surg. 1995;60(2
suppl):S108–S1012, discussion S113.
15. Vogt PR, Stallmach T, Niederhäuser U, Schneider J, Zünd G, Lachat M,
Künzli A, Turina MI. Explanted cryopreserved allografts: a morphological and immunohistochemical comparison between arterial allografts and
allograft heart valves from infants and adults. Eur J Cardiothorac Surg.
1999;15:639–644, discussion 644.
16. Lloyd TR, Marvin WJ Jr, Mahoney LT, Lauer RM. Balloon dilation valvuloplasty of bioprosthetic valves in extracardiac conduits. Am Heart J.
1987;114:268–274.
17.Waldman JD, Schoen FJ, Kirkpatrick SE, Mathewson JW, George L,
Lamberti JJ. Balloon dilatation of porcine bioprosthetic valves in the pulmonary position. Circulation. 1987;76:109–114.
18. Kusky T. Encyclopedia of Earth Science. New York, NY: Facts on File,
Inc; 2005:245.
19.Nakanishi T, Tobita K, Sasaki M, Kondo C, Mori Y, Tomimatsu H,
Nakazawa M, Imai Y, Momma K. Intravascular ultrasound imaging before and after balloon angioplasty for pulmonary artery stenosis. Catheter
Cardiovasc Interv. 1999;46:68–78.
20. Edwards BS, Lucas RV Jr, Lock JE, Edwards JE. Morphologic changes in
the pulmonary arteries after percutaneous balloon angioplasty for pulmonary arterial stenosis. Circulation. 1985;71:195–201.
21. Capelli C, Nordmeyer J, Schievano S, Lurz P, Khambadkone S, Lattanzio
S, Taylor AM, Petrini L, Migliavacca F, Bonhoeffer P. How do angioplasty balloons work: a computational study on balloon expansion forces.
EuroIntervention. 2010;6:638–642.
Angioplasty of Obstructed Homograft Conduits in the Right Ventricular Outflow Tract
With Ultra-Noncompliant Balloons: Assessment of Therapeutic Efficacy and Conduit
Tears
Michael R. Hainstock, Audrey C. Marshall, James E. Lock and Doff B. McElhinney
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 16, 2017
Circ Cardiovasc Interv. published online November 19, 2013;
Circulation: Cardiovascular Interventions is published by the American Heart Association, 7272 Greenville
Avenue, Dallas, TX 75231
Copyright © 2013 American Heart Association, Inc. All rights reserved.
Print ISSN: 1941-7640. Online ISSN: 1941-7632
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circinterventions.ahajournals.org/content/early/2013/11/19/CIRCINTERVENTIONS.112.000073
Data Supplement (unedited) at:
http://circinterventions.ahajournals.org/content/suppl/2013/11/18/CIRCINTERVENTIONS.112.000073.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation: Cardiovascular Interventions can be obtained via RightsLink, a service of the Copyright
Clearance Center, not the Editorial Office. Once the online version of the published article for which
permission is being requested is located, click Request Permissions in the middle column of the Web page
under Services. Further information about this process is available in the Permissions and Rights Question and
Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation: Cardiovascular Interventions is online at:
http://circinterventions.ahajournals.org//subscriptions/
Video Legend
Angiogram demonstrates an unconfined tear with extravasation of contrast from the
proximal conduit into the mediastinum.