Superior results following the Ross procedure in patients with congenital heart disease.

Abstract

The Ross procedure is a versatile operation that can be applied for aortic valve replacement (AVR) in patients with congenital heart disease (CHD), including small infants and those with complex left ventricular outflow tract (LVOT) obstruction. Herein, the clinical outcome is reported following the Ross procedure in patients with CHD at the authors' institution.
The medical records of patients who underwent the Ross procedure for CHD between 1991 and 2007 were reviewed. A competing-risks methodology was used to determine the time-related prevalence and associated factors for three mutually exclusive end states after the Ross procedure, namely death prior to subsequent cardiac reoperation, cardiac reoperation, and survival without subsequent reoperation.
A total of 151 patients (98 males, 53 females) was identified. The median age at the time of surgery was 8.6 years (range: 4 days to 33 years). Previously, 103 patients (68%) had undergone cardiac interventions, and 43 (28%) required LVOT enlargement (modified Ross-Konno procedure). A competing-risk analysis showed that, at 10 years after the Ross procedure, 8% of patients had died without subsequent reoperation, 26% underwent cardiac reoperation, and 66% remained alive without further reoperation. The 10-year freedom from autograft and homograft reoperation was 95% and 71%, respectively. Factors associated with early risk of mortality were age < 1 year and no prior surgical/percutaneous intervention at the time of the Ross procedure. Surgical factors associated with cardiac reoperation were concurrent cardiac surgery and the use of fresh homografts. There were no bleeding or thromboembolic complications, and the 15-year freedom from endocarditis was 95%. Ultimately, 99% of the survivors were in NYHA class I or II.
The Ross procedure remains the authors' procedure of choice for AVR in patients with CHD. Outcomes in infants aged < 1 year may improve with better patient selection and palliative surgical/percutaneous interventions prior to valve replacement. The late survival was excellent and valve-related complications were minimal. The high autograft longevity led to few patients requiring late reoperation for graft replacement.

Full text

Page 1

Superior Results Following the Ross Procedure in
Patients with Congenital Heart Disease
Bahaaldin Alsoufi1, Zohair Al-Halees1, Cedric Manlhiot2, Brian W. McCrindle2,
Mohammad Kandeel1, Mansoor Al-Joufan1, Avedis Kalloghlian1, Bahaa Fadel1,
Charles C. Canver1
1King Faisal Heart Institute, King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia, 2Labatt Family
Heart Center, the Hospital for Sick Children and the University of Toronto, Toronto, Ontario, Canada
Aortic valve disease is common in children, and fre-
quently requires intervention. Congenital aortic valve
disease is the most common pathology in North
America, and is encountered in approximately 5% of
children with congenital heart disease (CHD) (1-3).
Several developments in interventional cardiology
have allowed for the early treatment of congenital aor-
tic stenosis, while recent experience with aortic valve
repair techniques have demonstrated encouraging
short- and mid-term results in the treatment of con-
genital aortic valve disease (4-7). Nonetheless, in
patients with significant valve destruction, or follow-
ing repair or intervention failure, aortic valve replace-
ment (AVR) is required (2-4,8).
The Ross procedure is considered by many as the
procedure of choice for AVR in children and young
adults, as it is associated with excellent hemodynamics
and cardiac recovery, it offers growth potential, and it
does not require long-term anticoagulation (2,3,8-19).
However, the procedure is technically demanding, and
Presented at the Fifth Biennial Meeting of the Society for Heart Valve
Disease, 27th-30th June 2009, Berlin, Germany
Address for correspondence:
Bahaaldin Alsoufi MD, King Faisal Heart Institute (MBC 16), King
Faisal Specialist Hospital and Research Center, P. O. Box 3354,
Riyadh, Saudi Arabia 11211
e-mail: balsoufi@hotmail.com
© Copyright by ICR Publishers 2010
Background and aim of the study: The Ross procedure
is a versatile operation that can be applied for aortic
valve replacement (AVR) in patients with congenital
heart disease (CHD), including small infants and
those with complex left ventricular outflow tract
(LVOT) obstruction. Herein, the clinical outcome is
reported following the Ross procedure in patients
with CHD at the authors’ institution.
Methods: The medical records of patients who under-
went the Ross procedure for CHD between 1991 and
2007 were reviewed. A competing-risks methodology
was used to determine the time-related prevalence
and associated factors for three mutually exclusive
end states after the Ross procedure, namely death
prior to subsequent cardiac reoperation, cardiac reop-
eration, and survival without subsequent reopera-
tion.
Results: A total of 151 patients (98 males, 53 females)
was identified. The median age at the time of surgery
was 8.6 years (range: 4 days to 33 years). Previously,
103 patients (68%) had undergone cardiac interven-
tions, and 43 (28%) required LVOT enlargement
(modified Ross-Konno procedure).
A competing-risk analysis showed that, at 10 years
after the Ross procedure, 8% of patients had died
without subsequent reoperation, 26% underwent car-
diac reoperation, and 66% remained alive without
further reoperation. The 10-year freedom from auto-
graft and homograft reoperation was 95% and 71%,
respectively. Factors associated with early risk of
mortality were age <1 year and no prior surgi-
cal/percutaneous intervention at the time of the Ross
procedure. Surgical factors associated with cardiac
reoperation were concurrent cardiac surgery and the
use of fresh homografts. There were no bleeding or
thromboembolic complications, and the 15-year free-
dom from endocarditis was 95%. Ultimately, 99% of
the survivors were in NYHA class I or II.
Conclusion: The Ross procedure remains the authors’
procedure of choice for AVR in patients with CHD.
Outcomes in infants aged <1 year may improve with
better patient selection and palliative surgi-
cal/percutaneous interventions prior to valve replace-
ment. The late survival was excellent and
valve-related complications were minimal. The high
autograft longevity led to few patients requiring late
reoperation for graft replacement.
The Journal of Heart Valve Disease 2010;19:269-278

Page 2

late autograft failure, aortic root dilatation and homo-
graft reoperations have led to cause for concern (20-
24). Recent reports have suggested a higher early risk
but an improved autograft longevity and diminished
cardiac reoperations in children with underlying con-
genital aortic valve pathology, as compared to other
pathologies (especially rheumatic valve disease) (8).
The study aim was to review a large single-institu-
tion experience with the Ross procedure in children
and young adults with CHD, and to examine the
mutually exclusive time-related events of death and
cardiac reoperation in this group of patients.
Clinical material and methods
Patients
Between 1991 and 2007, a total of 151 consecutive
patients (98 males, 53 females) with CHD underwent
AVR, utilizing the Ross procedure, at the King Faisal
Specialist Hospital and Research Center in Riyadh,
Saudi Arabia. The patients were identified using the
hospital surgical database, while clinical, operative,
and outcome data were abstracted from their medical
records.
Approval to conduct the study was obtained from
the Research Ethics Board at the authors’ institution;
all requirements for patients to provide their individ-
ual consent to participate were waived for this obser-
vational study.
Operative details
Although details of the surgical steps will not be pro-
vided here, some important information is briefly list-
ed (8,9). A midline sternotomy and standard
cardiopulmonary bypass and myocardial protection
techniques were used in all cases. The pulmonary
autograft was implanted as a full root, with coronary
transfer in all cases. The autograft muscle cuff was
trimmed with sutures placed almost directly at the
base of the cusps; the proximal suture line was
installed using running polypropylene sutures but was
not reinforced in the patients, so as not to limit growth.
In 43 patients (28%) with a narrow left ventricular out-
flow tract (LVOT), a modified Ross-Konno technique
was used. In these patients, the fibrous annulus of the
aortic valve was often divided, with the cut being par-
tially taken down to the septum. Part of the septum
was cored out to completely open up the LVOT with-
out creating a large ventricular septal defect (VSD).
Right ventricle-to-pulmonary artery (RV-PA) conti-
nuity was established with a homograft in 146
patients, with a Contegra bovine jugular valved con-
duit in four patients, and with a non-valved pericar-
dial conduit in one patient. The median homograft size
used to reconstruct the right ventricular outflow tract
was 21 mm (range: 11 mm to 27 mm). Fresh homo-
grafts were used in 53% of cases, and cryopreserved
homografts in 47%. Pulmonary homografts were used
in 80% of cases, and aortic homografts in 20%. The
homografts were not generally matched to the blood
group of the patients.
In total, 24 patients (16%) required 25 additional con-
comitant cardiac surgeries, including mitral valve
repair (n = 13), mitral valve replacement (n = 1), aortic
arch reconstruction (n = 4), VSD closure (n = 2), coro-
nary artery bypass grafting (n = 2), pulmonary artery
augmentation (n = 2), and resection of an ascending
aortic aneurysm (n = 1).
The immediate postoperative results were assessed
in all patients in the operating room, using trans-
esophageal echocardiography.
Follow up
The patients were evaluated clinically, and by means
of detailed echocardiography upon discharge, at six
weeks and six months after surgery, and yearly there-
after.
Late outcomes were determined from recent office
visits to the King Faisal Specialist Hospital and
Research Center, or from direct correspondence with
the patients’ families. The follow up was complete in
92% of the patients and partial in 4%; six patients (4%)
who lived outside Saudi Arabia were lost to follow up.
The median duration of follow up was 5.9 years (max-
imum 17.4 years).
Statistical analysis
All data were presented as mean ± SD, medians with
minimum and maximum and frequencies, as appropri-
ate. Time-dependent outcomes (death and reoperation)
after the Ross procedure were parametrically modeled.
Parametric probability estimates for time-dependent
outcomes used models based on multiple, overlapping,
phases of risk (available for use with the SAS system at
http://www.clevelandclinic.org/heartcenter/hazard).
The HAZARD procedure used maximum likelihood
estimates to resolve risk distribution of time to event in
up to three phases of risk (early, constant, and late). A
competing risk analysis (CRA) was performed to
model the probability over time of each of two mutual-
ly exclusive endpoints, namely aortic valve reoperation
and death without a second AVR; the remainder of the
patients were alive and free from second AVR. In addi-
tion, a CRA was performed to model the probability
over time of each of two additional mutually exclusive
endpoints, namely cardiac reoperation and death with-
out cardiac reoperation; the remainder of the patients
were alive and free from cardiac reoperation. Variables
that might potentially influence the likelihood of out-
comes in the competing risk models were sought from
270 Ross procedure in congenital heart disease
B. Alsoufi et al
J Heart Valve Dis
Vol. 19. No. 3
May 2010

Page 3

demographic, anatomic and surgical variables via a
bootstrap bagging algorithm (500 samples). Variables
appearing in ≥50% of samples were retained in the final
model, which used a stepwise regression algorithm (p
<0.05 to enter) to obtain the final model for each phase
of risk. Effects of covariates on the probability of out-
comes in competing risk models were given as param-
eter estimates and percentage reliability in the
bootstrap algorithm. The clinical relevance of identified
covariates on the likelihood of selected outcomes was
established by solving the regression equations for
multiple ‘typical’ test patients. All statistical analyses
were performed using SAS statistical software V9.1
(The SAS Institute, Cary, NC, USA).
Results
Patient characteristics
The median age at the Ross procedure was 8.6 years
(range: 4 days to 33 years). The patient group included
seven neonates (aged ≤1 month), 13 infants (aged >1
month to ≤1 year), 118 children (aged >1 year to ≤18
years), and 13 adults (age range: 18 to 33 years). The
hemodynamic aortic valve dysfunction was primarily
stenosis in 51 patients (34%), primarily regurgitation in
13 patients (9%), and mixed in 87 patients (58%).
The majority of patients had bicuspid aortic valve (n
= 139), either in isolation or association with subaortic
membrane, VSD, coarctation/arch hypoplasia. In
addition, two patients had atrioventricular septal
defects, two had undergone an arterial switch opera-
tion, two had interrupted aortic arch, and six had
Shone’s complex. Three patients had endocarditis at
the time of surgery.
Before the Ross procedure, 103 patients (68%) had
undergone a prior cardiac intervention, including per-
cutaneous balloon aortic valvuloplasty in 43 (28%) and
prior cardiac surgery in 73 (48%); 13 patients had
undergone both percutaneous and surgical interven-
tions. Details of the preoperative cardiac interventions
in all patients are listed in Table I.
CRA for subsequent AVR or death without aortic
valve reoperation after the Ross procedure
Following the 151 initial Ross procedures, four
patients (3%) had their autografts replaced and 12 (8%)
died without a further AVR; the majority of these
deaths (n = 11) were operative mortalities prior to hos-
pital discharge. Freedom from autograft reoperation
was 100% at one year, 99% at five years, and 95% at 10
years.
The hazard function for time-related transition to
aortic valve reoperation was characterized by the pres-
ence of a late hazard phase that increased as years
since surgery progressed (Fig. 1). The hazard function
for time-related transition to death without aortic
valve reoperation was characterized by an early haz-
ard phase in the immediate postoperative period, but
stable survival thereafter (Fig. 2). The competing risks
for the two events showed that, at 10 years after the
Ross procedure, approximately 8% of patients had
died without subsequent aortic valve reoperation, 5%
Ross procedure in congenital heart disease
B. Alsoufi et al
271
J Heart Valve Dis
Vol. 19. No. 3
May 2010
Table I: Cardiac interventions prior to the Ross procedure.
Procedure No. of
patients
No prior intervention
Previous percutaneous intervention
Balloon valvuloplasty
Coarctation dilatation
Previous surgical intervention
Aortic valve commissurotomy
Sub-aortic membrane resection/myectomy
Repair of aortic coarctation/hypoplastic arch
Ventricular septal defect closure
Mitral valve repair
Aortic valve repair
Arterial switch operation
Complete atrioventricular septal defect repair
Partial atrioventricular septal defect repair
RV-PA conduit
Tricuspid valve repair
Vascular ring
48 (32)
43 (28)
41
3
73 (48)
44
20
15
10
4
4
2
1
1
1
1
1
Values in parentheses are percentages.
RV-PA: Right ventricle-to-pulmonary artery.
Figure 1: Parametric model for survival to aortic valve
reoperation is composed of a late hazard phase that steadily
increases as years since surgery progressed. The solid line
represents parametric point estimates; the dashed lines
enclose the 95% confidence interval.

Page 4

had undergone aortic valve reoperation, and 87% were
alive and free from aortic valve reoperation.
CRA for death or subsequent cardiac reoperation
after the Ross procedure
Following the 151 initial Ross procedures, 25 patients
(17%) underwent cardiac reoperation, and 11 (7%) died
without any further reoperation (all operative deaths).
In addition, one patient died shortly after cardiac reop-
eration.
The hazard function for time-related transition to
cardiac reoperation was characterized by the absence
of an early hazard phase of risk, but the presence of
constant and late hazard phases with reoperation rate
steadily increased as the years since surgery pro-
gressed (Fig. 3). The hazard function for time-related
transition to death without cardiac reoperation was
characterized by an early hazard phase in the immedi-
ate postoperative period, but stable survival thereafter.
The competing risks for the two events showed that, at
five years after the Ross procedure, approximately 8%
of the patients had died without subsequent cardiac
reoperation, 12% had undergone cardiac reoperation,
and 80% were alive and free from cardiac reoperation.
At 10 years, 26% had undergone cardiac reoperation
and 66% were alive and free from cardiac reoperation
(Fig. 4).
Factors associated with mortality
There were 11 (7%) operative deaths prior to hospital
discharge. One patient died from an embolic stroke,
one had uncontrolled hemorrhage, one patient had
clear evidence of coronary ischemia, two patients died
from sepsis/multi-organ failure, and six had persistent
refractory low cardiac output. The operative mortality
was 57% (4/7) in neonates, 15% (2/13) in infants aged
>1 month, 4% (5/118) in children aged >1 year, and 0%
(0/13) in adults aged >18 years.
Mortality in patients who required concomitant arch
reconstruction was 25%, and similar in patients with
and without annular enlargement (modified Ross-
Konno). The need for annular reduction was not iden-
tified as a risk factor for mortality (6.9% in patients
with annular enlargement versus 7.4% without
enlargement). When concomitant surgery was
required, the operative mortality was 21% compared
to 4.7% when an isolated Ross procedure was per-
formed.
In addition, only one late death occurred following
redo arch reconstruction and subsequent RV-PA homo-
graft replacement one year after the Ross procedure.
To date during the study period, there have been no
additional late deaths beyond that stage. The overall
survival at both one and 10 years was 92%.
Factors associated with a higher risk of death after the
Ross procedure were sought; the results are listed in Table
II. The significant factors for early-phase mortality were
surgery in children aged <1 year (parameter estimate
(PE): 2.2 ± 0.6, p = 0.0002) and primary Ross with no pre-
vious percutaneous/surgical intervention (PE: 2.3 ± 0.8, p
= 0.004). The higher risk of early-phase mortality in chil-
dren aged <1 year is depicted graphically in Figure 5.
272 Ross procedure in congenital heart disease
B. Alsoufi et al
J Heart Valve Dis
Vol. 19. No. 3
May 2010
Figure 2: Parametric model for overall survival is
composed of early phase of risk in the immediate
postoperative period and a stable survival thereafter. The
solid line represents parametric point estimates; the dashed
lines enclose the 95% confidence interval.
Figure 3: Parametric model for survival to cardiac
reoperation is composed of constant and late hazard phases
with a reoperation rate that steadily increases as years
since surgery progressed. The solid line represents
parametric point estimates; the dashed lines enclose the
95% confidence interval.

Page 5

Factors associated with aortic valve reoperation
During the follow up period there were four auto-
graft reoperations (3%) to replace the aortic valve at
1.4, 5.5, 6.3, and 6.4 years following the Ross proce-
dure, respectively. All reoperations were carried out to
treat the development of significant aortic valve regur-
gitation (one patient also developed supra-aortic
stenosis). The factors associated with a higher risk of
subsequent aortic valve reoperation were sought, but
no significant factors were identified (see Table II).
Factors associated with cardiac reoperation
In total, 25 patients (17%) underwent 27 cardiac
reoperations. At 10 years after AVR, approximately
67% of the patients had undergone additional cardiac
reoperations, and all patients but one had survived
reoperative surgery. Cardiac reoperations included
RV-PA homograft replacement (n = 17; 13 surgical, four
percutaneous pulmonary valve implantations in the
most recent experience), AVR (n = 4), mitral valve
replacement (n = 4), mitral valve repair (n = 2), tricus-
pid valve repair (n = 3), pulmonary artery reconstruc-
tion (n = 3), permanent pacemaker implantation (n =
2), and redo arch reconstruction (n = 1). Freedom from
RV-PA homograft replacement at 10 years was 71%.
Factors associated with a higher risk of cardiac reoper-
ation following surgery were sought and are listed in
Table II.
Significant factors associated with cardiac reopera-
tion were additional simultaneous cardiac surgery at
the time of the Ross procedure (PE: 1.1 ± 0.5, p = 0.03
for the constant hazard phase and PE: 1.5 ± 0.5, p =
0.004 for the late hazard phase), and the use of fresh
versus cryopreserved homografts (PE: 1.0 ± 0.5, p =
0.04 for constant hazard phase only).
The higher risk of undergoing cardiac reoperation in
patients with concurrent surgery at the time of the
Ross procedure is depicted graphically in Figure 6.
Functional status and additional late morbidity
At the most recent follow up, the majority of sur-
vivors (96%) had a normal clinical exercise ability and
were in NYHA functional class I (only one patient was
in NYHA class III).
There were no complications of bleeding, thrombosis
or thromboembolism in any of the patients during the
study period. One patient developed endocarditis of
an 11-year-old stenotic homograft that required
replacement. The freedom from endocarditis was 100%
at 10 years, and 95% at 15 years.
Discussion
The present study described the experience of a sin-
gle institution with AVR, using the Ross procedure in
151 patients with CHD. A CRA approach was chosen
because these patients were simultaneously at risk for
two mutually exclusive events, namely death and
reoperation. Conventional time-related analyses con-
sider individual events such as death or reoperation
Ross procedure in congenital heart disease
B. Alsoufi et al
273
J Heart Valve Dis
Vol. 19. No. 3
May 2010
Figure 4: Competing risk depiction of outcomes following the
Ross procedure in patients with CHD. The competing risks
for the two events (cardiac reoperation and death without
cardiac reoperation) showed that, at five years after the Ross
procedure, approximately 8% of patients had died without
cardiac reoperation, 12% had undergone cardiac reoperation,
and 80% were alive and free from cardiac reoperation. At 10
years, 26% had undergone cardiac reoperation and 66% were
alive and free from cardiac reoperation.
Figure 5: Parametric model for survival in patients
younger and older than 1 year. The solid lines represent
parametric point estimates; the dashed lines enclose the
95% confidence intervals.

Page 6

either in isolation or as a combined end point and,
although useful, they do not address the question of
how often an event may occur in the absence of other
events for which a patient is at simultaneous risk.
Mortality risk
Despite its surgical complexity, the Ross procedure
has been shown to be safe in experienced hands (2,3,8-
19). At the present authors’ center, the overall opera-
tive mortality was 2.3%, and similar to that reported
from The International Registry for the Ross Procedure
(2.5%).
In previous reports, it has been identified that AVR
in children with CHD is associated with a higher oper-
ative mortality as compared to that in children with
other underlying etiologies (8). For example, early
mortality in children following the Ross procedure for
rheumatic heart disease (RHD) at the present authors’
institution was less than 1%, compared to 7% in a cur-
rent report in patients with CHD (8).
Patients with congenital aortic valve pathology are
more likely to require AVR at a very young age.
Indeed, age at the time of surgery (especially infants
aged <1 year) has been identified as a significant factor
for early mortality. Mortality in this age group was
30% (57% in neonates, 15% in older infants), but only
3.8% in patients aged >1 year. These results paralleled
those of other reports which indicated a higher early
mortality in infants. For example, when Kadner et al.
(19) reported outcomes following the Ross procedure
in 52 children (including 15 infants), there were five
operative mortalities, all in infants (33%). Likewise,
reports from other experienced centers have described
that the majority of operative mortalities had occurred
in infants (11-19).
Children requiring AVR during infancy may have
complex LVOT obstruction, in addition to hypoplasia
of the left ventricle, mitral valve, and aorta (4), such
that associated surgery such as arch reconstruction and
mitral valve repair may be required. In the present
series, four patients required concurrent arch recon-
struction, with one early and one late mortality. In
addition, mitral valve repair was required in five of 20
infants, of which four died (80% mortality). The under-
lying diagnosis in four of these 20 infants was Shone’s
abnormality, and none of the four survived. On the
other hand, 16 of the infants had bicuspid aortic valve
(associated with interrupted aortic arch in two case),
and 14 (88%) of these survived.
In light of the present results, infants with multiple
obstructive lesions in the left ventricle who require
AVR at a very early age - and especially those requir-
ing simultaneous mitral and arch interventions - may
be better served with a single-ventricle, multi-stage
palliation strategy. However, in infants with isolated
LVOT obstruction the Ross procedure would be advis-
able. In those infants with complex LVOT obstruction,
the present authors use the modified Ross-Konno tech-
nique for annular enlargement without the creation of
a VSD, and therefore without the need for a separate
VSD patch placement. The mortality in patients who
required annular enlargement in the present current
series was 7% (3/43) for the entire cohort (22% for
infants), and similar to those who did not require
LVOT enlargement.
An adequate assessment of those left heart structures
suitable for biventricular repair is complex and beyond
the scope of the present discussion (4,25). Nonetheless,
274 Ross procedure in congenital heart disease
B. Alsoufi et al
J Heart Valve Dis
Vol. 19. No. 3
May 2010
Table II: Incremental risk factors for time-related transition from the Ross procedure to either death or reoperation.
Factors associated with:Variable Estimated PE* p-value
A higher risk of death
(early phase)
No percutaneous intervention/
surgery prior to Ross procedure
2.27 ± 0.780.004
Age <1 year2.17 ± 0.59 0.0002
A higher risk of
aortic valve reoperation
(late phase)
None
A higher risk of
cardiac reoperation
(constant phase)
Concurrent surgery at
time of Ross procedure
1.12 ± 0.50 0.03
Fresh homograft0.99 ± 0.480.04
A higher risk of cardiac
reoperation (late phase)
Concurrent surgery at
time of Ross procedure
1.47 ± 0.500.004
*Values are mean ± SE.
PE: Parameter estimate.

Page 7

an improved selection of biventricular candidates may
help to reduce early mortality among young infants
who require the Ross procedure for non-repairable
aortic valve pathology (4,25).
In the present series, those children who had under-
gone a prior aortic valve intervention (either percuta-
neous or surgical) had an improved early-phase
survival. The present data support an attempted aortic
valve repair to relieve the LVOT obstruction as a pal-
liative intervention in selected patients with isolated
LVOT pathology, with or without arch obstruction,
and adequate left heart structures. However, the opti-
mal choice of intervention remains controversial, and
is beyond the scope of the present discussion (4,5).
Despite a higher operative mortality in young chil-
dren, the Ross procedure remains the best replacement
choice currently available (2,3,8). At the authors’ institu-
tion, the operative mortality for patients aged <5 years
who had undergone AVR with a mechanical prosthesis
was 50% (8). In the present series, mortality after the
Ross procedure in this similar age group was 15%.
Most importantly, survival was stable following the
Ross procedure, with no late cardiac mortality during
the study period. This compared very favorably with
the results of others following AVR with mechanical
valves in children and young adults (2,3,8,26). In a pre-
vious report from the authors’ institution, which com-
pared long-term outcomes in 346 children who had
undergone AVR (131 mechanical, 215 Ross), there was
no late cardiac mortality among the Ross group com-
pared to 19 late cardiac-related deaths in the mechani-
cal group (8). Notably, many of these deaths were
sudden in nature, highlighting the delicate problem of
compliance with an anticoagulation regimen and its
implication on the patients’ lifestyle that is especially
difficult to control in children and young adults.
Reoperation risk
Cardiac reoperation following the Ross procedure
remains a major concern (2,3,8-24), with several recent
reports having provided evidence of progressive
dilatation of the sinotubular junction (STJ), aortic root
and worsening aortic regurgitation requiring autograft
reoperation and AVR (20-24).
In a previous report from the present authors, sever-
al risk factors were identified which included rheu-
matic fever, especially among those patients with pure
preoperative aortic regurgitation and dilated aortic
annulus as significant risk factors for aortic valve reop-
eration (8,29). On the other hand, it has been noted that
autograft reoperations occurred infrequently among
patients who required the Ross procedure for CHD (8).
In the present series, the 10-year freedom from AVR
was favorable (95%).
Dilatation of the weak right ventricular muscle
beneath the implanted autograft may be prone to
dilatation, due to an exposure to systemic pressure
with localized rupture and eventual aneurysm forma-
tion and subsequent development of regurgitation (27-
29). Whilst earlier in the present study, a greater
amount of the right ventricular muscle was left
attached to the autograft to compensate for the size dif-
ference between a smaller pulmonary autograft and a
dilated aortic root, the subsequent procedure - in
response to such observations - was to resect most of
the right ventricular muscle beneath the valve, with
the sutures passing directly to the base of the cusps
(27-29).
Notably, concerns of pulmonary autograft dilatation
have stimulated recent developments in surgical
strategies to address late root dilatation. Most recently,
inspired by a technique that was previously used in
mitral valve replacement with a pulmonary autograft,
Slater and colleagues (30) described a modified pul-
monary autograft implantation procedure in which the
pulmonary autograft was encased in a Dacron tube so
as to prevent future dilatation. These authors suggest-
ed that such a procedure could be applied to fully
grown children and young adults, and reported excel-
lent short-term results with respect to autograft valve
function and lack of dilatation of the annulus or STJ
(30). Moreover, valve-preserving root replacement
techniques in cases associated with root dilatation and
autograft regurgitation have been described which
may increase valve longevity, despite aortic wall rein-
tervention (31).
Currently, replacement of the RV-PA homograft
remains the most common reoperation after the Ross
Ross procedure in congenital heart disease
B. Alsoufi et al
275
J Heart Valve Dis
Vol. 19. No. 3
May 2010
Figure 6: Parametric model for survival to cardiac
reoperation in patients who received concomitant cardiac
surgery and those who did not. The solid lines represent
parametric point estimates; the dashed lines enclose the
95% confidence intervals.

Page 8

procedure (2,3,8-19). In particular, the mortality risk
during homograft replacement is low, while recent
advances with percutaneous pulmonary valve replace-
ment have allowed cardiologists to address this prob-
lem without the need for surgery. To date, the
immediate and short-term results of percutaneous pul-
monary valve implantation have been very encourag-
ing, and these procedures have recently been
conducted at the present authors’ institution in several
patients following the Ross procedure, with excellent
immediate results (32).
Finally, the results of the present series showed that
the risk of cardiac reoperation, other than that for the
autograft or homograft, would be increased in those
patients who required concomitant surgery at the time
of the Ross procedure. While the risk of reoperation is
smallest in children with isolated aortic valve patholo-
gy, survival is not affected by concurrent surgery
requirement in older children with adequate left heart
structures. Moreover, the use of the Ross procedure is
supported as the best replacement alternative, espe-
cially in small children.
In conclusion, the Ross procedure is a versatile opera-
tion that can be applied for AVR in patients with vari-
ous forms of CHD, including small infants and those
with complex LVOT obstruction. Neonates and small
infants, and especially those with additional
mitral/arch pathologies, have the highest operative
mortality and may be better served with a single-ven-
tricle palliation strategy. In those infants deemed suit-
able for biventricular repair, aortic valvuloplasty
(either percutaneous or surgical) offers an initial palli-
ation and allows the Ross procedure to be performed,
if needed, at an older age; this is associated with an
improved early survival. Late mortality was shown to
be minimal in all age groups, and the autograft
longevity to be high. Although, currently, homograft
replacement is the most common cause for reopera-
tion, the evolution of percutaneous pulmonary valve
replacement techniques may, in time, permit a non-
operative management of this problem.
References
1. Salomon NW, Stinson EB, Oyer P, Copeland JG,
Shumway NE. Operative treatment of congenital
aortic stenosis. Ann Thorac Surg 1978;26:452-460
2. Brown JW, Ruzmetov M, Vijay P, Rodefeld MD,
Turrentine MW. Surgery for aortic stenosis in chil-
dren: A 40-year experience. Ann Thorac Surg
2003;76:1398-1411
3. Karamlou T, Jang K, Williams WG, et al. Outcomes
and associated risk factors for aortic valve replace-
ment in 160 children: A competing-risks analysis.
Circulation 2005;112:3462-3469
4. Alsoufi B, Karamlou T, McCrindle BW, Caldarone
CA. Management options in neonates and infants
with critical left ventricular outflow tract obstruc-
tion. Eur J Cardiothorac Surg 2007;31:1013-1021
5. McCrindle BW, Blackstone EH, Williams WG, et al.
Are outcomes of surgical versus transcatheter bal-
loon valvotomy equivalent in neonatal critical aor-
tic stenosis? Circulation 2001;104:I152-I158
6. Tweddell JS, Pelech AN, Frommelt PC, et al.
Complex aortic valve repair as a durable and effec-
tive alternative to valve replacement in children
with aortic valve disease. J Thorac Cardiovasc Surg
2005;129:551-558
7. Alsoufi B, Karamlou T, Bradley T, et al. Short and
midterm results of aortic valve cusp extension in
the treatment of children with congenital aortic
valve disease. Ann Thorac Surg 2006;82:1292-1299
8. Alsoufi B, Al-Halees Z, Manlhiot C, et al.
Mechanical valves versus the Ross procedure for
aortic valve replacement in children: Propensity-
adjusted comparison of long-term outcomes. J
Thorac Cardiovasc Surg 2009;137:362-370
9. Al Halees Z, Pieters F, Qadoura F, Shahid M, Al-
Amri M, Al-Fadley F. The Ross procedure is the
procedure of choice for congenital aortic valve dis-
ease. J Thorac Cardiovasc Surg 2002;123:437-441
10. Turrentine MW, Ruzmetov M, Vijay P, Bills RG,
Brown JW. Biological versus mechanical aortic
valve replacement in children. Ann Thorac Surg
2001;71(Suppl.):S356-S360
11. Starnes VA, Luciani GB, Wells WJ, Allen RB, Lewis
AB. Aortic root replacement with the pulmonary
autograft in children with complex left heart
obstruction. Ann Thorac Surg 1996;62:442-448
12. Elkins RC, Lane MM, McCue C. Ross operation in
children: Late results. J Heart Valve Dis 2001;10:736-
741
13. Hraska V, Krajci M, Haun Ch, et al. Ross and Ross-
Konno procedure in children and adolescents: Mid-
term results. Eur
2004;25:742-747
14. Marino BS, Wernovsky G, Rychik J, Bockoven JR,
Godinez RI, Spray TI. Early results of the Ross pro-
cedure in simple and complex left heart disease.
Circulation 1999;100(Suppl.2):162-166
15. Khwaja S, Nigro JJ, Starnes VA. The Ross procedure
is an ideal aortic valve replacement operation for
the teen patient. Semin Thorac Cardiovasc Surg
Pediatr Card Surg Annu 2005:173-175
16. Ohye RG, Gomez CA, Ohye BJ, Goldberg CS, Bove
AL. The Ross/Konno procedure in neonates and
infants: Intermediate-term survival and autograft
function. Ann Thorac Surg 2001;72:823-830
17. Piccardo A, Ghez O, Gariboldi V, et al. Ross and
Ross-Konno procedures in infants, children and
J Cardiothorac Surg
276 Ross procedure in congenital heart disease
B. Alsoufi et al
J Heart Valve Dis
Vol. 19. No. 3
May 2010

Page 9

adolescents: A 13-year experience. J Heart Valve Dis
2009;18:76-82
18. Hazekamp MG, Grotenhuis HB, Schoof PH,
Rijlaarsdam ME, Ottenkamp J, Dion RA. Results of
the Ross operation in a pediatric population. Eur J
Cardiothorac Surg 2005;27:975-979
19. Kadner A, Raisky O, Degandt A, et al. The Ross
procedure in infants and young children. Ann
Thorac Surg 2008;85:803-808
20. Laudito A, Brook MM, Suleman S, et al. The Ross
procedure in children and young adults: A word of
caution. J Thorac Cardiovasc Surg 2001;122:147-153
21. David TE, Omran A, Ivanov J, et al. Dilation of the
pulmonary autograft after the Ross procedure. J
Thorac Cardiovasc Surg 2000;119:210-220
22. Solymar L, Sudow G, Holmgren D. Increase in size
of the pulmonary autograft after the Ross operation
in children: Growth or dilation? J Thorac
Cardiovasc Surg 2000;119:4-9
23. Pasquali SK, Cohen MS, Shera D, Wernovsky G,
Spray TL, Marino BS. The relationship between
neo-aortic root dilation, insufficiency, and reinter-
vention following the Ross procedure in infants,
children and young adults. J Am Coll Cardiol
2007;49:1806-1812
24. Hanke T, Stierle U, Boehm JO, et al. Autograft
regurgitation and aortic root dimensions after the
Ross procedure: The German Ross Registry experi-
ence. Circulation 2007;116(Suppl.):I251-I258
25. Hickey EJ, Caldarone CA, Blackstone EH, et al.
Critical left ventricular outflow tract obstruction:
The disproportionate impact of biventricular repair
in borderline cases. J Thorac Cardiovasc Surg
2007;134:1429-1436
26. Alexiou C, McDonald A, Langley SM, Dalrymple-
Hay MJR, Haw MP, Monro JL. Aortic valve replace-
ment in children: Are mechanical prostheses a good
option? Eur J Cardiothorac Surg 2000;17:125-133
27. Shahid MS, Al-Halees Z, Khan SM, Pieters FA.
Aneurysms complicating pulmonary autograft pro-
cedure for aortic valve replacement. Ann Thorac
Surg 1999;68:1842-1843
28. Metras D, Kreitmann B, Habib G, Riberi A, Yao JG,
Ross DN. False aneurysm: A rare and potentially
severe complication after pulmonary autograft
operation. J Heart Valve Dis 1996;5:414-416
29. Pieters FA, Al-Halees Z, Hatle L, Shahid MS, Al-
Amri M. Results of the Ross operation in rheumat-
ic versus non-rheumatic aortic valve disease. J
Heart Valve Dis 2000;9:38-44
30. Slater M, Shen I, Welke K, Komanapalli C,
Ungerleider R. Modification to the Ross procedure
to prevent autograft dilatation. Semin Thorac
Cardiovasc Surg Pediatr Card Surg Annu 2005:181-
184
31. Ishizaka T, Devaney EJ, Ramsburgh SR, Suzuki T,
Ohye RG, Bove EL. Valve sparing aortic root
replacement for dilatation of the pulmonary auto-
graft and aortic regurgitation after the Ross proce-
dure. Ann Thorac Surg 2003;75:1518-1522
32. Nordmeyer J, Coats L, Bonhoeffer P. Current expe-
rience with percutaneous pulmonary valve implan-
tation. Semin Thorac
2006;18:122-125
Cardiovasc Surg
Ross procedure in congenital heart disease
B. Alsoufi et al
277
J Heart Valve Dis
Vol. 19. No. 3
May 2010

Page 10

Meeting discussion
DR. J. J. M. TAKKENBERG (Rotterdam, The
Netherlands): I find it confusing that ‘no previous
intervention’ is seen as a risk factor. I would leave that
out of my analysis, because it obviously represents
those patients younger than 1 year or neonates, most
likely with critical aortic stenosis, who undergo a
Ross-Konno procedure. Is that correct?
DR. BAHAALDIN ALSOUFI (Riyadh, Saudi Arabia):
No prior intervention was identified as a risk factor for
early mortality in our Current series. It was in fact a
significant risk factor for all age groups, more so for
neonates and infants. Our data suggests that surgical
or catheter palliation is beneficial in all patients. It
allows them to grow to an older age when they can tol-
erate the procedure better. It’s an interesting finding
that the Ross procedure outcomes at our institution are
much inferior in neonates and infants than in older
patients. This is not related to deficiency in the post-
operative care of this age group at our part of the
world. For example, mortality following the Norwood
procedure at our institution is about 15%, and follow-
ing the arterial switch procedure about 2%; and yet
still, the mortality following the Ross procedure in
neonates is far inferior than those comparably complex
procedures in the same age group..
DR. TAKKENBERG: But from a statistical point of
view, did you look at the correlation between previous
procedures and the different age groups?
DR. ALSOUFI: Yes, and that was shown in one of my
slides. Prior intervention was associated with lower
early mortality in all age groups, more so in neonates
and infants. For example, survival for patients older
than 1 year who’ve undergone prior intervention
approached 100%, compared to about 90% in those
older than 1 year of age without prior intervention. On
the other hand, survival for those younger than 1 year
of age who’ve undergone a prior intervention was 90%
as compared to less than 50% in those with no prior
intervention. So obviously, the role of palliative inter-
vention is more critical in neonates and infants.
DR. TAKKENBERG: Yes, it’s all early mortality, but
did you determine a correlation coefficient? Are you
not looking at ‘no previous intervention’ being a proxy
for being in a particular age group?
DR. ALSOUFI: While no prior intervention is more
prevalent in neonates and infants, our finding is not as
a result that no prior intervention is a proxy for
younger patient. It was shown as an independent fac-
tor for early mortality on multi-variable analysis.
DR. TAKKENBERG: So that remains a nice
point for your discussion
SIR MAGDI YACOUB (London, UK): The application
of the Ross procedure in infants, as well as Konno,
valvotomy and repair of the aortic root in the infant
age group, gives excellent results and can keep these
children alive until they are aged 20, or even more. I
am slightly worried as to why there was such a high
incidence Ross operations in children aged less than
one year, and also of the Konno procedure which, in
my experience, is very, very rarely required.
DR. ALSOUFI: I’ll answer the second part first, what
we call a‘modified Konno’ is when an aggressive sep-
tal myectomy is performed - we cut through the annu-
lus slightly towards the conal septum, but we don’t
need to place a VSD patch. In other words, the ‘modi-
fied Konno’ is aggressive left ventricular outflow tract
enlargement. This was not shown to be a significant
risk factor for early death or for reoperation in our
series. At our institution, initial palliation, either inter-
ventional or surgical, is attempted to delay the require-
ment for aortic valve replacement to a later age. In our
series, only 32% of patient didn’t have a prior catheter
or surgical intervention. We believe that aortic valvu-
loplasty will improve the hemodynamic manifesta-
tion, will stabilize left ventricular size and allows the
patients to grow for many years prior to the require-
ment of aortic valve replacement and that‘s shown in
our series where the majority of procedure were per-
formed late in childhood. Infant / neonatal Ross pro-
cedure was required in only 20 patients in our series.
The outcomes were good in those with isolated aortic
valve disease whereas in those infants who required
additional arch surgery or mitral valve repair, the out-
comes were poor suggesting that those patients may
be better managed with single ventricle palliation.
- thank you.
278 Ross procedure in congenital heart disease
B. Alsoufi et al
J Heart Valve Dis
Vol. 19. No. 3
May 2010