Myocardial Infarction

Transcatheter Closure of Post-myocardial Infarction

Ventricular Septal Rupture
Gabriele Egidy Assenza, MD; Doff B. McElhinney, MD; Anne Marie Valente, MD;
Disty D. Pearson, PA-C; Massimo Volpe, MD; Giuseppe Martucci, MDcM;
Michael J. Landzberg, MD*; James E. Lock, MD*

Background—Ventricular septal rupture (VSR) after acute myocardial infarction (AMI) is a potentially lethal mechanical
complication of acute coronary syndromes. Given high surgical mortality, transcatheter closure has emerged as a potential
strategy in selected cases. We report our single-center experience with double-umbrella device percutaneous closure of
post-AMI VSR.
Methods and Results—In this single-center, retrospective, cohort study, patients who underwent transcatheter closure of post-
AMI VSR between 1988 and 2008 at Boston Children’s Hospital were included. Data were analysed according to whether
the patients underwent direct percutaneous VSR closure or closure of a residual VSR after a previous surgical approach.
Primary outcome was mortality rate at 30 days. Clinical predictors of primary outcome were investigated using univariate
logistic regression. Thirty patients were included in the study (mean age, 67±8 years). A total of 40 closure devices were
implanted. Major periprocedural complications occurred in 4 (13%) patients. Cardiogenic shock, increasing pulmonary/
systemic flow ratio, and the use of the new generation (6-arm) STARFlex device all were associated with higher risk of
mortality. The Model for End-Stage Liver Disease Excluding international normalized ratio (MELD-XI) score at the time
of VSR closure seemed to be most strongly associated with death (odds ratio, 1.6; confidence interval, 1.1–2.2; P<0.001).
Conclusions—Transcatheter closure of post-AMI VSR using CardioSEAL or STARFlex devices is feasible and effective.
The MELD-XI score, a marker of multiorgan dysfunction, is a promising risk stratifier in this population of patients.
Early closure of post-AMI VSR is advisable before establishment of multiorgan failure.  (Circ Cardiovasc Interv.
2013;6:59-67.)
Downloaded from http://ahajournals.org by on June 28, 2022

Key Words: catheterization ◼ interventional closure ◼ myocardial infarction ◼ ventricular septal defect

V entricular septal rupture (VSR) after acute myocardial

infarction (AMI) is a potentially lethal mechanical
complication of acute coronary syndromes.1 VSR classically
occurs as a bimodal presentation with a higher incidence in
the first 24 hours and then again 3 to 5 days after AMI. The
incidence of VSR has substantially decreased since the
widespread implementation of early reperfusion strategies,
ranging between 0.2% and 0.34% in the modern era. However,
when it does occur, VSR clinical presentation is often ominous,
is commonly associated with extensive comorbidities, and
results in poor cardiac output, multiorgan failure, and death.2
Early surgical closure is recommended. This current indication,
irrespective of hemodynamic status, is based on natural history
data that suggest a 90% mortality within 2 months of diagnosis
in patients who do not undergo defect closure.3
Since the original description of post-AMI VSR in 1957,
surgical closure remains a challenging procedure with mor-
tality rates between 20% and 87%.4–6 Expert opinion and pro-
fessional guidelines suggest that immediate closure should
be considered to reduce the duration of poor systemic perfu-
sion that results from left-to-right shunting, pulmonary over
circulation, and systemic hypoperfusion that may eventuate
in refractory multiple organ failure and death.7 Despite this
awareness, there is a tendency for surgeons to wait several
weeks before operating to allow tissue healing and more
complete rupture remodeling, contributing to high interstage
mortality and positive selection of more favorable cases. In
this setting, transcatheter approaches for closure of post-AMI
VSR have been introduced as a less invasive means of reduc-
ing or eliminating shunt and improving antegrade perfusion,

Received June 27, 2012; accepted December 13, 2012

From the Department of Cardiology, Boston Children’s Hospital, Harvard Medical School, Boston, Massachusetts (G.E.A., D.B.M., A.M.V., D.D.P.,
M.J.L., J.E.L); Department of Medicine, Division of Cardiology, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts
(G.E.A., A.M.V., D.D.P., M.J.L.); Department of Clinical and Molecular Medicine, Sant’Andrea Hospital, “Sapienza Universita’ di Roma” University,
Rome, Italy (G.E.A., M.V.); IRCCS Neuromed, Pozzilli, Italy (M.V.); and McGill University Health Center, McGill University, Montreal, Canada (G.M.).
*Drs Landzberg and Lock contributed equally to the study.
Guest Editor for this article was Jeff A. Brinker, MD.
The online-only Data Supplement is available at http://circinterventions.ahajournals.org/lookup/suppl/doi:10.1161/CIRCINTERVENTIONS.
112.972711/-/DC1.
Correspondence to Michael J. Landzberg, MD, Boston Children’s Hospital, Department of Cardiology, 300 Longwood Ave, Boston, MA 02115. E-mail
mike.landzberg@cardio.chboston.org
© 2013 American Heart Association, Inc.
Circ Cardiovasc Interv is available at http://circinterventions.ahajournals.org DOI: 10.1161/CIRCINTERVENTIONS.112.972711

59
 60  Circ Cardiovasc Interv  February 2013
WHAT IS KNOWN
• Ventricular septal rupture after acute myocardial in-
farction is a lethal mechanical complication of acute
coronary syndromes.
• Early surgical closure is recommended, but imple-
mentation of such an indication in clinical practice is
heterogeneous among centers because of the exces-
sive surgical risk perceived by operators
• Transcatheter closure of ventricular septal rupture
emerged as a possible alternative to surgery in se-
lected cases. However, limited information is avail-
able for such a procedure, which is currently an off-
label indication.
WHAT THE STUDY ADDS
• This is a single-center series reviewing transcatheter
closure of post-myocardial infarction ventricular
septal rupture in 30 patients.
• We found that this procedure is technically feasible
and can be accomplished with relatively low risk of
major periprocedural complication, including death.
• Multiorgan failure dictates high risk of 30-day mor-
tality, irrespective of closure status.
as definitive single therapy, as a bridge to surgery, or for
closure of residual defects after surgical repair in selected
Downloaded from http://ahajournals.org by on June 28, 2022
cases.8–10 To date, the majority of studies of transcatheter clo-
sure of post-AMI VSR have reported on the use of Amplatzer
atrial septal defect and ventricular septal defect occluders
(AGA Medical Corporation, Plymouth, MN).8–10
Recent studies have investigated the utility of the Model for
End-Stage Liver Disease (MELD) score as a potential predictor
of poor outcomes in patients with heart disease.11,12 A logarith-
mic function of creatinine, total bilirubin, and international nor-
malized ratio, the MELD score in many ways serves as a simpler
marker of multiple organ dysfunction. The modified MELD
score Excluding international normalized ratio (MELD-XI
score) was validated against the original MELD score and has
been shown to be equally able to predict mortality in patients
with liver cirrhosis waiting for liver transplantation.13
In this study, we report the outcomes of transcatheter
closure of post-AMI VSR using different double-umbrella
devices in a consecutive cohort of patients referred during the
course of 2 decades to our institution.
Methods
Population
In this single-center, retrospective, cohort study, patients referred to
our institution for transcatheter closure of post-AMI VSR between
1988 and 2008 were included. Patients were excluded if they pre-
sented with congenital heart disease or a ventricular septal defect re-
sulting from a previous cardiac surgical procedure. Clinical data were
collected retrospectively from medical records, electronic charts,
procedure reports, discharge summaries, and clinical correspondence
with the referring physician or institution. The Social Security Death
Index was used to confirm the 30-day clinical status for each patient.
This study was compliant with the Health Insurance Portability and
Accountability Act, and a waiver of informed consent was approved
by the Institutional Review Boards of Boston Children’s Hospital,
and Brigham and Women’s Hospital.
Patients were classified into 2 groups according to whether they
underwent primary VSR closure with no surgical attempt before the
index transcatheter procedure (primary VSR) or closure of a recur-
rent defect after a previous surgical VSR closure (residual VSR).
In all patients the indication for closure relied on the presence of
severe heart failure and pulmonary over circulation not amenable
to medical therapy. Although prospective criteria were not used to
guide the eligibility for transcatheter VSR closure, a surgical option
was ­denied in all patients by consensus of experienced cardiologists
and cardiac surgeons.
Outcomes and Definitions
All-cause mortality at 30 days was defined as the primary outcome.
Cardiogenic shock at the time of postinfarction VSR closure was
defined as need for intravenous vasopressor support to maintain a
systolic blood pressure >90 mm Hg, need for an intra-aortic balloon
pump to sustain an adequate circulation, and evidence of poor end-
organ perfusion (eg, increasing creatinine, increasing lactate, urine
output <30 mL/h). Time to admission was defined as the time be-
tween symptom onset and primary hospital admission. Time to VSR
closure was the time between symptom onset and VSR closure. In
patients who underwent surgical closure as the first option (residual
VSR group), time to surgery was the time between symptom onset
and surgical VSR closure.
The MELD-XI score was calculated using creatinine and total bili-
rubin at the time of VSR closure according to the following formula:
5.11×ln (bilirubin mg/dL) + 11.76×ln (creatinine mg/dL) + 9.44.13
Transcatheter VSR Closure Procedure
Catheterization and VSR closure were performed under fluoro-
scopic guidance. In selected cases, intraprocedural transesophageal
echocardiography was used to improve visualization of intracardiac
anatomy. Invasive hemodynamic measurements were obtained for
every patient before VSR closure, including calculation of the car-
diac index, pulmonary/systemic flow ratio (using Fick equation),
and direct measurement of the left ventricular (LV) end-diastolic
pressure. Vascular access was generally obtained in the femoral ar-
tery and femoral vein. Internal jugular venous access was used for
recovery of the venous arm of the arterio-venous or veno-venous
loop to reduce torsional tension of the wire loop in the setting of pos-
terior or very apical defects. In many cases a veno-venous loop was
established using trans-septal puncture (Brockenbrough procedure)
to reach the LV cavity (Figure 1). The defect was typically crossed
from the LV cavity to the right ventricular cavity using a balloon-
tipped, Judkins, Amplatzer, or multipurpose catheter and a long, soft
guide wire. After advancing the catheter to the pulmonary artery, the
wire was snared from the venous access and subsequently exterior-
ized through the femoral or the internal jugular vein to establish a
loop (Figure 2, Video I in the online-only Data Supplement).
An appropriate long sheath was then advanced over the wire and
across the defect (Figure 2). The approach for sheath delivery and
deployment depended on the location of the defect and other factors
specific to individual anatomic constraints. In our early experience,
the device chosen was 1.5-times to 2.2-times the maximal balloon,
angiographic, or echocardiographic size, although in our more recent
experience, we chose the largest device available that could fit in the
specific anatomic location. After deployment and complete expan-
sion of the proximal ventricular arms, LV angiography usually was
performed through a separate catheter to assess the position of the
device and assess for residual intracardiac shunt, valve interference,
and device position.
Devices
Three generations of double-umbrella closure devices were deployed
during this study: the first-generation Clamshell device (C.R. Bard,
 Egidy Assenza et al   Post-AMI VSR Transcatheter Closure   61
Figure 1.  Schematic representation of railway loop establish-
ment, using transseptal and transaortic approaches. A, Transsep-
tal puncture is used to advance a catheter into the left ventricle.
Downloaded from http://ahajournals.org by on June 28, 2022
A long, soft guide wire is advanced across the ventricular septal
rupture into the pulmonary artery, where it is snared (B) to avoid
potential ensnaring of the tricuspid valve subvalvar apparatus.
C, Transaortic approach is used to cross the ventricular septal
rupture and to advance the long soft guidewire into the pulmonary
artery. D, Alternative route is used to snare the guide wire in the
pulmonary artery and to exteriorize the venous arm from the infe-
rior vena cava (see Video I in the online-only Data Supplement).
Murray Hill, NJ), the second-generation CardioSEAL device, and the
subsequent modification third-generation STARFlex device (NMT
Medical, Boston, MA). These devices consisted of 2 opposing self-
expanding umbrellas, scaffolded by single spring-loaded stainless
steel arms that were covered with a Dacron meshwork. The STARFlex
modification of the CardioSEAL device incorporated MP35N alloy
for the umbrella arms, an internally suspended self-adjusting nitinol
microspring meshwork, a self-centering mechanism, and a smaller
delivery system profile.
Data Analysis
Between-group comparisons for clinical and outcome variables were
performed using independent samples t test, Wilcoxon rank-sum test,
χ2 analysis, or Fisher exact test. Receiver–operator characteristic
curves were used to determine optimal thresholds for dichotomizing
predictor variables. Univariate logistic regression was used to identify
predictors of composite outcome. The following predictors have been
included in the analysis: type of defect (residual versus primary); LV
ejection fraction; cardiogenic shock at the time of closure; hemo-
dynamics at the time of closure (LV end-diastolic pressure, cardiac
index, and the ratio between pulmonary and systemic blood flow);
MELD-XI score at the time of closure; white blood cell count at the
time of closure; fever at the time of closure; time to closure; num-
ber of coronary arteries involved; and device type. Because of the
small number of events, multivariable analysis of outcomes was not
performed. Data are reported as mean±SD, median (first and third
quartiles), or frequency (%). Odds ratios are presented with 95% con-
fidence intervals. All tests are 2-sided. Given the exploratory nature
of the study, no adjustment for multiple comparisons was performed.
Results
Patients
Thirty patients underwent transcatheter closure of post-AMI
VSR during the study period. Table 1 summarizes demo-
graphic, clinical, and laboratory features of the entire study
population and the 2 clinical subgroups. Twelve patients
underwent primary VSR closure, and 18 had double-umbrella
devices placed after an initial surgical procedure after a
median of 52 days from the original operation (first and third
quartile range, 21–170 days). On average, patients were in the
sixth decade of life and had multiple coronary artery disease
risk factors. The majority of patients received aspirin and
heparin before VSR closure, but only a minority were treated
with β-blockers or systemic thrombolysis. Cardiogenic
shock was present in 17 patients (57%) at the time of VSR
closure. Patients who underwent primary VSR closure had a
shorter time to transcatheter closure than postsurgical closure
patients. An abnormal serum creatinine level was common;
the calculated MELD-XI score was significantly higher in
those undergoing primary closure.
VSR was the result of myocardial necrosis attributable to
ST elevation myocardial infarction in all but 1 patient. ST ele-
vation myocardial infarction localization involved the anterior
wall of the left ventricle in 15 patients (50%) and the infe-
rior wall of the LV in 11 patients (39%). Table 2 summarizes
hemodynamic, angiographic, and reperfusion strategy fea-
tures of the cohort. In patients who underwent primary VSR
closure, the VSR diameter by angiography was significantly
larger than in patients with residual postsurgical VSR. In the
26 patients in whom coronary angiograms were available
for review, 21 had stenoses or occlusions in 1 or 2 epicar-
dial coronary arteries, and only 5 had multivessel obstruction.
Myocardial reperfusion with percutaneous coronary interven-
tion or coronary artery bypass surgery before VSR closure was
performed in 21 patients. Percutaneous coronary intervention
was never simultaneous with transcatheter VSR closure. In all
but 1 of the patients who underwent coronary artery bypass
grafting, surgical closure of the VSR was performed at the
time of coronary surgery. In 3 other patients, surgical VSR
closure was performed without concomitant coronary artery
bypass grafting.
VSR Closure and Procedural Outcome
A total of 40 closure devices were implanted in the 30 study
patients. Clamshell devices were used in our first experience
with this procedure (between 1988 and 1997), CardioSEAL
devices were used between 1994 and 2008, and StarFLEX
device were used in the latest era of the study (after 2001).
In 8 patients, multiple devices were implanted because of
persistence or recurrence of large-volume shunts after the
first device was placed. In 5 patients, additional devices were
implanted at the time of the first VSR closure procedure.
In 2 patients, 1 additional procedure was performed a few
days after the first intervention to implant a second device.
In another patient, 2 additional procedures were performed
 62  Circ Cardiovasc Interv  February 2013
(1 week and 11 months later, respectively), implanting a
total of 3 devices because of a large residual shunt. Tables 3
and 4 summarize procedural details. Major periprocedural
Downloaded from http://ahajournals.org by on June 28, 2022
complications occurred in 4 patients (13%), as listed in
Table 4. There was 1 procedural death occurring in a patient
who had development of electromechanical dissociation
during intracardiac manipulation of the delivery sheath,
and who was not able to be restored to a perfusing rhythm.
This patient presented with a large anterior ST elevation
myocardial infarction and was treated with primary
angioplasty a few hours before VSR closure. One other
patient had development of transient complete heart block
during manipulation of the long sheath and was treated with
temporary pacing; heart block resolved after few minutes
and a 38-mm STARFlex device was successfully implanted.
A permanent pacemaker was not implanted.
In 2 patients, the device embolized from the optimal posi-
tion. In 1 patient, a 40-mm Clamshell device dislodged from
the initial position and became entangled in the tricuspid sub-
valvar apparatus. This device was retrieved a few days later
using a transcatheter approach; the patient ultimately died a
few days after this procedure because of multiorgan failure.
The VSR diameter by angiography was 18 mm. In another
patient, a 33-mm Clamshell device dislodged from the ini-
tial position but was retrieved and a 40-mm Clamshell device
was then successfully implanted. In this patient, before the
implantation of the second device, the VSR was tested with
low-pressure sizing balloon inflation, giving a stretched
diameter >20 mm.
In 3 patients (all presenting with primary VSR), surgi-
cal repair of VSR was performed after transcatheter closure
because of persistence of large-volume shunt. In 1 case,
Figure 2.  Right anterior oblique fluoroscopic view
demonstrating (A) transseptal position of the long
guiding sheath (big black arrow) into the left ven-
tricle; (B) established arterio-venous loop using a
balloon-tipped catheter*; (C) delivery sheath across
the ventricular septal rupture through transseptal
approach with a fully loaded 33-mm StarFLEX
device (small black arrow); and (D) deployment of
the 33-mm StarFLEX device (dotted black arrow).
surgical repair was accomplished the same day of the failed
percutaneous repair. In the other 2 patients, surgical repair
was completed 1 and 4 months after percutaneous closure,
respectively.
30-Day Mortality
The primary outcome, death within 30 days of VSR closure,
occurred in 7 patients (23%). Comparing the outcome rate in
the first consecutive 15 patients with the rest of the cohort,
no difference was noted (20% versus 27%; P=0.7). All deaths
(with the exception of the intraprocedural event) were related
to multiorgan failure in the setting of systemic hypoperfusion
despite aggressive medical therapy, inotropic support, and
intra-aortic balloon pump. There was a trend toward a higher
30-day mortality in patients who underwent primary VSR clo-
sure (5 of 12; 42%) compared with patients with a residual
VSR after surgical closure (2 of 18; 11%; P=0.08). Factors
significantly associated with 30-day mortality at a 2-sided
0.05 level of significance are summarized in Table 5.
Cardiogenic shock, increasing pulmonary/systemic flow
ratio (at the time of percutaneous closure), and use of the
43-mm (6-arm) STARFlex device were all related to increased
risk of mortality. The baseline MELD-XI score at the time
of VSR closure was strongly associated with increased risk
of death. On receiver–operator characteristic curve analysis, a
MELD-XI score of 20 was associated with 100% sensitivity
and 87% specificity for predicting the primary outcome. The
association between baseline MELD-XI score and outcome
remained significant even after repeated the analysis in the 2
clinical subgroups (primary VSR closure versus residual VSR
closure), although with marginal probability values because
of the small sample size (primary VSR group: dds ratio, 1.49
 Egidy Assenza et al   Post-AMI VSR Transcatheter Closure   63

Table 1.  Demographics of the Study Population and the Two Clinical Subgroups
Overall Cohort Primary Ventricular Residual Ventricular P
(n=30) Septal Rupture (n=12) Septal Rupture (n=18) Value
Male, n (%) 12 (41) 8 (67) 8 (44) 0.2
BSA, kg/m2 1.84±0.2 1.9±0.2 2.2±0.2 0.1
Age, y 67±8 68±6 67±9 0.6
Systemic arterial hypertension, n (%) 12 (41) 7 (58) 5 (29) 0.1
Hypercholesterolemia, n (%) 12 (41) 7 (58) 5 (29) 0.1
Diabetes mellitus, n (%) 4 (14) 2 (17) 2 (12) 0.7
Active smoking, n (%) 4 (14) 3 (25) 1 (6) 0.1
Family history of CAD, n (%) 5 (17) 3 (25) 2 (12) 0.4
History of CAD, n (%) 4 (14) 2 (17) 2 (12) 0.7
History of stroke, n (%) 2 (7) 1 (8) 1 (6) 0.8
ASA before VSR closure, n (%) 20 (67) 9 (75) 11 (55) 0.4
Heparin before VSR closure, n (%) 18 (60) 8 (67) 10 (57) 0.5
GpIIb-IIIa before VSR closure, n (%) 2 (7) 1 (8) 1 (6) 0.8
Beta-blocker before VSR closure, n (%) 9 (30) 3 (25) 6 (67) 0.6
Systemic thrombolysis before VSR closure, n (%) 4 (13) 1 (8) 3 (17) 0.5
LVEF by echocardiogram, % 48±10 50±14 47±8 0.3
VSR diameter by echocardiogram, mm 13.6±6.7 16.7±9.1 11.2±3 0.2
At least moderate mitral regurgitation by echocardiogram, n (%) 3 (10) 2 (17) 1 (6) 0.3
Cardiogenic shock at the time of VSR closure, n (%) 17 (57) 7 (58) 10 (56) 0.6
IABP at the time of VSR closure, n (%) 14 (47) 5 (42) 9 (50) 0.7
Inotropic support at the time of VSR closure, n (%) 13 (43) 5 (42) 8 (44) 0.9
Time to admission, d 0 (0–8) 7 (0–22) 0 (0–5) 0.049
Time to surgery, d NA NA 4.5 (0–10) NA
Downloaded from http://ahajournals.org by on June 28, 2022

Time to VSR closure, d 27 (17–172) 19 (11–27) 54 (22–173) 0.045

Creatinine at the time of VSR closure, mg/dL 1.6 (1.2–2.45) 2.3 (1.2–3.6) 1.4 (1.15–1.7) 0.1
Bilirubin at the time of VSR closure, mg/dL 1.2 (0.7–1.5) 1.5 (0.7–1.7) 1.2 (0.7–1.2) 0.03
MELD-XI score at the time of VSR closure 16 (12–21) 19 (16–26) 14 (12–18) 0.03
WBC 103/μL 12.72 (7.32–14.93) 13.28 (7.36–16.23) 12.15 (7.28–14.47) 0.9
Fever at the time of VSR closure, n (%) 8 (29%) 2 (18%) 6 (35%) 0.4
Data are presented as as mean±SD, median (first and third quartiles), or frequency (%).
ASA indicates aspirin; BSA, body surface area; CAD, coronary artery disease; IABP, intra-aortic balloon pump; LVEF, left ventricular ejection fraction; MELD-XI, Model
for End-stage Liver Disease Excluding International normalized ratio; NA, not applicable; VSR, ventricular septal rupture; and WBC, white blood cell.

per 1-unit difference; confidence interval, 0.99–2.26; P=0.01;
residual VSR group: odds ratio, 1.65 per 1-unit difference;
confidence interval, 0.7–3.9; P=0.07).
Discussion
Post-AMI VSR: Management Considerations
Our study reports on the use of double-umbrella devices
(Clamshell, CardioSEAL, and STARFlex devices) in the
transcatheter closure of post-AMI VSR. Our experience
suggests that these devices can be used in these challeng-
ing interventional procedures achieving reasonable techni-
cal outcome and low rate of major complications. The high
30-day mortality (23%), which was attributable to multior-
gan failure and poor systemic perfusion, reflects the burden
of cardiac dysfunction and extensive comorbidity, which is
usually present in this group of patients.8,9 As previously
reported, we confirmed that cardiogenic shock at the time of
clinical presentation and the hemodynamic burden imposed
by the left-to-right shunt are important predictors of unfa-
vourable outcomes.9
In an effort to better understand the risks associated with
multiorgan system dysfunction in this setting, we examined
the MELD-XI score13 and found it to be strongly associated
with 30-day mortality. Assuming normal total bilirubin and
creatinine values, the MELD-XI score is between 9.4 and
12 in normal subjects. In our study, 62% of patients with a
MELD-XI score >20 died within 30 days of VSR closure.
Our data suggest that multiorgan failure dictates a very high
mortality rate irrespective of closure attempt. Recently,
aggressive ventricular mechanical support using biventricu-
lar assist device or extracorporeal membrane oxygenators
have been proposed to bridge patient presenting with post-
AMI VSR to closure or cardiac transplantation in case of end-
stage ventricular failure.14–16 Recent data from the Society
of Thoracic Surgeons database reported on >2800 surgical
post-MI VSR closures performed in North America between
 64  Circ Cardiovasc Interv  February 2013

Table 2.  Myocardial Infarction Localization, Hemodynamics, and Coronary Artery Disease Extent of the Study Population
Overall Cohort Primary Ventricular Residual Ventricular P
(n=30) Septal Rupture (n=12) Septal Rupture (n=18) Value
STEMI localization*, n (%) 0.4
 Anterior 15 (50) 4 (33) 11 (61)
 Antero-lateral 1 (4) 0 (0) 1 (6)
 Inferior 11 (39) 6 (50) 5 (27)
  Inferior plus RV 2 (7) 1 (7) 1 (6)
CI, L/(min m2) 2.4±0.7 2.5±0.7 2.3±0.7 0.5
LVEDP, mm Hg 21±6.8 19±1.7 22±1.7 0.2
Qp/Qs† 2.25±1 2.5±0.4 2.1±0.2 0.2
Mean PA pressure 29±6.7 29±7.5 29±6.4 0.9
VSR diameter by angiography, mm 10 (12–16) 18 (11–20) 10 (10–12) 0.004
Coronary artery disease 0.5
  Single-vessel CAD, n (%)‡ 11 (42) 5 (45) 6 (40)
  Two-vessel CAD, n (%)‡ 10 (38) 5 (45) 5 (33)
  Multivessel CAD, n (%)‡ 5 (19) 1 (10) 4 (27)
Culprit lesion revascularization, n (%) 21 (70) 5 (42) 16 (89) 0.01
CABG before VSR closure, n (%) 16 (53) 2 (17) 14 (78) 0.001
PCI before VSR closure, n (%) 5 (17) 3 (25) 2 (11) 0.7
Data are presented as mean±SD, median (interquartile range), or frequency (%).
CABG indicates coronary artery bypass grafting; CAD, coronary artery disease; CI, cardiac index; LVEDP, left ventricular end-diastolic pressure; PA, pulmonary artery;
PCI, percutaneous coronary intervention; RV, right ventricle; STEMI, ST-elevation myocardial infarction; and VSR, ventricular septal rupture.
*Localization of the infarcted area is not reported for the patient who presented non-ST elevation myocardial infarction-related VSR.
†This refers to the ratio between pulmonary and systemic blood flow at the time of percutaneous closure.
‡The percent refers to the 26 patients with angiogram available for review.
Downloaded from http://ahajournals.org by on June 28, 2022

1999 and 2010.17 Total 30-day mortality (including intraop- creatinine and need for preoperative dialysis were powerful
erative mortality) was 42.9%. Univariate and multivariate predictors of cardiovascular death. Use of MELD-XI score
logistic regression analysis confirmed that increased serum may improve our ability to identify this high-risk group of
patients earlier in whom aggressive mechanical ventricular
Table 3.  Technical Details of Transcatheter VSR Closure support might improve overall outcome as a bridge to trans-
plant or as adjunctive therapy in conjunction with surgical or
Patient
percutaneous closure.
Vascular access, n (%) Eighteen patients were referred to our institution after
  FA, FV 16 (53) failure to achieve adequate surgical closure of VSR. In this
  FA, FV, LSV 3 (10) subgroup of patients, our results were most promising, with
  FA, FV, RIJV 10 (33) excellent technical outcome (immediate effective closure in
  FA, RIJV 1 (4) every patient) and relatively low mortality at 30 days. One of
the concerns that prevent early surgical post-AMI VSR clo-
Trans-septal puncture, n (%) 24 (80)
sure is that the margins of the defect do not offer sufficient
Loop, n (%)
strength for suturing the patch, which could result in higher
 AV 9 (30) rate of recurrence after early surgery. However, waiting for
 VV 21 (70) complete healing of the VSR margins has significant risk.5
Venous exteriorization of guide wire loop, n (%) Our data suggest that a hybrid approach, with early surgery
 RFV 22 (73) and potential transcatheter closure of residual or recurrent
 RIJV 7 (23) defects, is feasible. This strategy may prevent the detrimen-
 LSV 1 (4) tal effects of prolonged poor systemic perfusion and left heart
overload, and seems particularly reasonable in patients with
VSR localization, n (%)
large defects in which a higher rate of device dislodgment and
  Anterior ventricular septum 5 (17)
technical failure may be anticipated.10
  Posterior ventricular septum 10 (33)
  Mid-muscular ventricular septum 14 (46) Post-AMI VSR: Technical and Procedural
  Apical ventricular septum 1 (4) Considerations
A-V indicates arterio-venous; FA, femoral artery; FV, femoral vein; LSV, left Our current practice is to perform VSR closure, as possible,
subclavian vein; RIJV, right internal jugular vein; VSR, ventricular septal defect; within 24 hours of diagnosis. We tend to favor a dual venous
and V-V, veno-venous. access for establishing the wire loop with a trans-septal
 Egidy Assenza et al   Post-AMI VSR Transcatheter Closure   65

Table 4.  Device Characteristics and Periprocedural Adverse Events

Variable
Balloon sizing performed, n (%) 12 (40)
TEE guidance utilized, n (%) 19 (63)
Device type, n (%) Overall Cohort Primary Ventricular Septal Residual Ventricular Septal
(Total Device n=40) Rupture (Total Device n=15) Rupture (Total Device n=25)
 Clamshell 13 (32.5) 4 (33) 9 (36)
 CardioSEAL 16 (45) 4 (33) 12 (48)
 STARFlex 11 (22.5) 7 (54) 4 (14)
Device diameter, n (%)*
  23 mm 4 (10)
  28 mm 7 (17.5)
  33 mm 16 (40)
  38 mm 3 (7.5)
  40 mm 5 (12.5)
  43 mm (STARFlex) 5 (12.5)
Need for additional devices placement, n (%) 8 (20)
Stroke, n (%) 0 (0)
Device embolization, n (%) 2 (7)
Pulseless arrhythmias requiring emergent 1 (3.5)
measure, n (%)
Periprocedural complete heart block requiring 2 (7)
pacing, n (%)
Pacemaker implantation, n (%) 0 (0)
Periprocedural death, n (%) 1 (3.5)
Subsequent surgical closure, n (%) 3 (10)
Downloaded from http://ahajournals.org by on June 28, 2022

Data presented as frequency (%).

TEE indicates transesophageal echocardiogram.
*Percentages are calculated considering the total number of devices implanted (n=40).

approach because it lessens hemodynamic instability related to
the retrograde crossing of aortic valve. In addition, the trans-
septal course of the delivery catheter allows a more favorable
intracardiac manipulation of the device because it permits the
opening of the distal device arms in the inflow portion of the
right ventricle rather than in the right ventricular apex.
VSR are classically serpiginous ruptures rather than
smoothly defined passages (Figure 3, Video II in the online-
only Data Supplement). Often the LV entry and the right
ventricular exit of the defect can be distant and posteriorly
located. Particular attention to anatomic details and proximity
to adjacent structures are required. Localized angiography
(Video III in the online-only Data Supplement) and multipla-
nar imaging seem to be useful in this setting.
We usually defer balloon sizing the defect, given that necro-
sis and myocardial remodeling likely change defect size and
shape, and that balloon manipulation may worsen tissue dehis-
cence or lead to rupture extension of the free (ventricular) wall.
It is not uncommon to require additional device implantation
either at the time of initial closure or within 3 to 6 days of the
initial procedure, and patients and families should be coun-
seled regarding this potential. The risk of subacute recurrence

Table 5.  Predictors of Mortality at 30 Days After Transcatheter Post-AMI VSR Closure
Variable Unadjusted OR (95% CI) P Value
Cardiogenic shock 6.5 (0.7–63) 0.06
Time to closure (per one unit difference)* 0.88 (0.7–1.0) 0.049
Residual VSR 0.2 (0.02–1.1) 0.053
Qp/Qs (per 1-unit difference)† 3.0 (1.2–10.0) 0.001
43 mm, 6-arm Starflex device 9.0 (1.3–61.0) 0.02
MELD-XI score (per one unit difference) 1.6 (1.1–2.2) < 0.001
AMI indicate acute myocardial infarction; CI, confidence interval; MELD-XI, Model for End-stage Liver Disease Excluding International normalized ratio; OR, odds ratio;
and VSR, ventricular septal rupture.
*Time to closure OR was analyzed only for the primary VSR group.
†This refers to the ratio between pulmonary and systemic blood flow at the time of percutaneous closure.
 66  Circ Cardiovasc Interv  February 2013
Figure 3.  A, Transgastric short-axis echocardiographic view (intraprocedural transesophageal echocardiography) of a posterior basal
ventricular septal rupture after an inferior ST elevation myocardial infarction (white arrow). B, Left anterior oblique angiographic view of
the same patient at the time of percutaneous ventricular septal rupture closure. Black arrow and the asterisk identify the irregular margins
of the serpiginous rupture.
of shunt after transcatheter closure of post-AMI VSR should
prompt scheduled echocardiographic surveillance to improve
early recognition and to guide therapeutic decisions.
Post-AMI VSR: Device-Specific Considerations
In our series, there was 1 intraprocedural death and major non-
fatal complications in 4 patients. The majority of the reports
Downloaded from http://ahajournals.org by on June 28, 2022
on transcatheter closure of post-MI VSR have been limited
to a single case or small series of patients.18–23 Larger series
reporting on the use of Amplatzer occluder devices present a
30-day mortality rate between 28% and 42%.8–10 In those stud-
ies, the intraprocedural death rate ranged from 0% to 17%.
Interestingly, the majority of intraprocedural deaths were
related to myocardial rupture during intracardiac manipula-
tion of the Amplatzer device.9,10 No direct comparison can be
made between our results and the other series that reported on
use of alternative devices because of the nonrandomized study
design, the highly selected groups of patients, and different
protocols. However, multiple concerns have been raised on the
use of Amplatzer device in this setting, including rigid struc-
ture of the device, which may increase the risk of myocardial
rupture; limited available sizes of the Amplatzer muscular
ventricular septal defect occluders; and a suboptimal profile of
the Amplatzer atrial septal defect occluder, with its relatively
short waist that can lead to device deformation after deploy-
ment (cobra effect).9 In our experience, we did not observe
any major mechanical complications related to device manip-
ulation and deployment. We speculate that the lower and more
flexible profile of CardioSEAL and STARFlex devices may
reduce the risk of myocardial rupture during manipulation of
the device in the infarcted ventricle during deployment phase,
as well as during ventricular remodeling, as compared with
the Amplatzer device.
Limitations
This is a retrospective study reporting on patients who under-
went device closure of post-AMI VSR during a 20-year period.
As such, it is certain that there are a number of limitations,
including changes in management and unmeasured variables
that might introduce bias. The limited number of patients in
this series precluded any meaningful multivariable analysis to
identify independent predictors of mortality.
We did not use 3-dimensional transesophageal echocar-
diography in this study. This advanced noninvasive imag-
ing modality is increasingly recognized as a powerful
tool to visualize complex anatomic structures, improving
procedural outcome during cardiac surgery and complex
catheter-based interventions, including post-AMI VSR per-
cutaneous closure.24
Conclusions
Post-AMI VSR is a lethal complication of acute coronary syn-
dromes. The detrimental interplay between cardiogenic shock,
high-volume intracardiac shunting, and pulmonary overcircu-
lation frequently results in severe multiorgan failure and leads
to high risk of mortality. Transcatheter reduction or closure
of post-AMI VSR using double-umbrella, CardioSEAL, or
STARFlex devices is feasible either as a primary therapy or
in a hybrid approach including transcatheter closure of poten-
tial residual VSR after surgical repair. The MELD-XI score,
a marker of multiorgan dysfunction, seems to be a promising
risk stratifier in this population of patients.
Acknowledgments
The authors acknowledge the clinical and administrative staff of the
Boston Adult Congenital Heart (BACH) and Pulmonary Hypertension
Program, of Boston Children’s Hospital and Brigham and Women’s
Hospital, for their invaluable care and support for these patients. The
authors thank Emily Harris for the support with the art work.
Sources of Funding
This study has been supported, in part, by the Dunlevie Fund of the
Boston Adult Congenital Heart (BACH) and Pulmonary Hypertension
Program, Boston Children’s Hospital and Brigham and Women’s
Hospital Boston.
 Egidy Assenza et al   Post-AMI VSR Transcatheter Closure   67
Disclosures
Dr Landzberg has received grant and research support from NMT
Medical and was a consultant and a member of the advisory board
of NMT Medical. Dr Lock has received device royalties from NMT
Medical. Clamshell, CardioSEAL, and STARFlex devices are not ap-
proved by the Food and Drug Administration to treat post-AMI VSR.
References
1. Birnbaum Y, Fishbein MC, Blanche C, Siegel RJ. Ventricular septal rupture
after acute myocardial infarction. N Engl J Med. 2002;347:1426–1432.
2. Crenshaw BS, Granger CB, Birnbaum Y, Pieper KS, Morris DC, Kleiman
NS, Vahanian A, Califf RM, Topol EJ. Risk factors, angiographic patterns,
and outcomes in patients with ventricular septal defect complicating acute
myocardial infarction. GUSTO-I (Global Utilization of Streptokinase and
TPA for Occluded Coronary Arteries) Trial Investigators. Circulation.
2000;101:27–32.
3. Lee WY, Cardon L, Slodki SJ. Perforation of infarcted interventricular
septum. Report of a case with prolonged survival, diagnosed ante mortem
by cardiac catheterization, and review of the literature. Arch Intern Med.
1962;109:731–741.
4. Cooley DA, Belmonte BA, Zeis LB, Schnur S. Surgical repair of ruptured
interventricular septum following acute myocardial infarction. Surgery.
1957;41:930–937.
5. Deja MA, Szostek J, Widenka K, Szafron B, Spyt TJ, Hickey MS, Sos-
nowski AW. Post infarction ventricular septal defect - can we do better?
Eur J Cardiothorac Surg. 2000;18:194–201.
6. Papadopoulos N, Moritz A, Dzemali O, Zierer A, Rouhollapour A, Acker-
mann H, Bakhtiary F. Long-term results after surgical repair of postinfarc-
tion ventricular septal rupture by infarct exclusion technique. Ann Thorac
Surg. 2009;87:1421–1425.
7. Antman EM, Anbe DT, Armstrong PW, Bates ER, Green LA, Hand M, Ho-
chman JS, Krumholz HM, Kushner FG, Lamas GA, Mullany CJ, Ornato JP,
Pearle DL, Sloan MA, Smith SC Jr, Alpert JS, Anderson JL, Faxon DP, Fus-
ter V, Gibbons RJ, Gregoratos G, Halperin JL, Hiratzka LF, Hunt SA, Jacobs
AK, Ornato JP. ACC/AHA guidelines for the management of patients with
ST-elevation myocardial infarction; A report of the American College of
Downloaded from http://ahajournals.org by on June 28, 2022
Cardiology/American Heart Association Task Force on Practice Guidelines
(Committee to Revise the 1999 Guidelines for the Management of patients
with acute myocardial infarction). J Am Coll Cardiol. 2004;44:E1–E211.
8. Holzer R, Balzer D, Amin Z, Ruiz CE, Feinstein J, Bass J, Vance M, Cao
QL, Hijazi ZM. Transcatheter closure of postinfarction ventricular septal
defects using the new Amplatzer muscular VSD occluder: Results of a
U.S. Registry. Catheter Cardiovasc Interv. 2004;61:196–201.
9. Thiele H, Kaulfersch C, Daehnert I, Schoenauer M, Eitel I, Borger M,
Schuler G. Immediate primary transcatheter closure of postinfarction ven-
tricular septal defects. Eur Heart J. 2009;30:81–88.
10. Maltais S, Ibrahim R, Basmadjian AJ, Carrier M, Bouchard D, Cartier R,
Demers P, Ladouceur M, Pellerin M, Perrault LP. Postinfarction ventricu-
lar septal defects: towards a new treatment algorithm? Ann Thorac Surg.
2009;87:687–692.
11. Matthews JC, Pagani FD, Haft JW, Koelling TM, Naftel DC, Aaronson
KD. Model for end-stage liver disease score predicts left ventricular as-
sist device operative transfusion requirements, morbidity, and mortality.
Circulation. 2010;121:214–220.
12. Ailawadi G, Lapar DJ, Swenson BR, Siefert SA, Lau C, Kern JA,
Peeler BB, Littlewood KE, Kron IL. Model for end-stage liver dis-
ease predicts mortality for tricuspid valve surgery. Ann Thorac Surg.
2009;87:1460–1467.
13. Heuman DM, Mihas AA, Habib A, Gilles HS, Stravitz RT, Sanyal AJ,
Fisher RA. MELD-XI: a rational approach to “sickest first” liver trans-
plantation in cirrhotic patients requiring anticoagulant therapy. Liver
Transpl. 2007;13:30–37.
14. Tsai MT, Wu HY, Chan SH, Luo CY. Extracorporeal membrane oxygen-
ation as a bridge to definite surgery in recurrent postinfarction ventricular
septal defect. ASAIO J. 2012;58:88–89.
15. Rohn V, Spacek M, Belohlavek J, Tosovsky J. Cardiogenic shock in pa-
tient with posterior postinfarction septal rupture–successful treatment
with extracorporeal membrane oxygenation (ECMO) as a ventricular as-
sist device. J Card Surg. 2009;24:435–436.
16. Gregoric ID, Bieniarz MC, Arora H, Frazier OH, Kar B, Loyalka P. Percu-
taneous ventricular assist device support in a patient with a postinfarction
ventricular septal defect. Tex Heart Inst J. 2008;35:46–49.
17. Arnaoutakis GJ, Zhao Y, George TJ, Sciortino CM, McCarthy PM, Conte
JV. Surgical repair of ventricular septal defect after myocardial infarction:
outcomes from the Society of Thoracic Surgeons National Database. Ann
Thorac Surg. 2012;94:436–443.
18. Mullasari AS, Umesan CV, Krishnan U, Srinivasan S, Ravikumar M,
Raghuraman H. Transcatheter closure of post-myocardial infarction ven-
tricular septal defect with Amplatzer septal occluder. Catheter Cardiovasc
Interv. 2001;54:484–487.
19. Lee EM, Roberts DH, Walsh KP. Transcatheter closure of a residual post-
myocardial infarction ventricular septal defect with the Amplatzer septal
occluder. Heart. 1998;80:522–524.
20. Lowe HC, Jang IK, Yoerger DM, MacGillivray TE, de Moor M, Palacios
IF. Compassionate use of the amplatzer ASD closure device for residual
postinfarction ventricular septal rupture following surgical repair. Cath-
eter Cardiovasc Interv. 2003;59:230–234.
21. Goldstein JA, Casserly IP, Balzer DT, Lee R, Lasala JM. Transcatheter
closure of recurrent postmyocardial infarction ventricular septal defects
utilizing the Amplatzer postinfarction VSD device: a case series. Catheter
Cardiovasc Interv. 2003;59:238–243.
22. Martinez MW, Singh CP, Mookadam F. Transcatheter closure of a postin-
farction ventricular septal rupture. J Am Soc Echocardiogr. 2006;19:1401.
e5–1401.e7.
23. Eshtehardi P, Garachemani A, Meier B. Percutaneous closure of a postin-
farction ventricular septal defect and an iatrogenic left ventricular free-
wall perforation using two Amplatzer muscular VSD occluders. Catheter
Cardiovasc Interv. 2009;74:243–246.
24. Scandura S, Ussia GP, Caggegi A, Cammalleri V, Millan G, Tamburino
C. Postinfarction ventricular septal defect: the role of three-dimensional
echocardiography. Echocardiography. 2012;29:E107–E109.