The challenge of congenital heart disease worldwide: epidemiologic and demographic facts.

Abstract

Congenital heart disease (CHD) afflicts a large number of children every year. The incidence of CHD is generally considered to be 8 per 1,000 live births. However, this estimate is perhaps inaccurate and does not take into consideration regional differences. A large review of the literature was performed to establish the true incidence of CHD and geographical variations. Data on the incidence of specific lesions and their geographical variation, as well as on mortality from CHD, was also reviewed. Taking into consideration the available data on incidence, mortality, and access to care, the global challenge that CHD represents was analyzed. Insight into how to confront this challenge is given.

Full text

Page 1

The Challenge of Congenital Heart Disease
Worldwide: Epidemiologic and Demographic Facts
Pierre-Luc Bernier, Ada Stefanescu, Gordan Samoukovic, and Christo I. Tchervenkov
Congenital heart disease (CHD) afflicts a large number of children every year. The inci-
dence of CHD is generally considered to be 8 per 1,000 live births. However, this estimate
is perhaps inaccurate and does not take into consideration regional differences. A large
review of the literature was performed to establish the true incidence of CHD and geo-
graphical variations. Data on the incidence of specific lesions and their geographical
variation, as well as on mortality from CHD, was also reviewed. Taking into consideration
the available data on incidence, mortality, and access to care, the global challenge that
CHD represents was analyzed. Insight into how to confront this challenge is given.
Semin Thorac Cardiovasc Surg Pediatr Card Surg Ann 13:26-34 © 2010 Elsevier Inc. All
rights reserved.
Introduction
C
maincauseofdeathinthefirstyearoflife.Inthelast50years,
the remarkable advances in cardiac surgery have offered real
hope of a normal life for many of these children.
It is generally stated that the incidence of CHD (ie, the
number of new cases occurring in a defined population dur-
ing a specified period of time) is 8 per 1,000 live births.1
However,thisfrequentlyquotednumberisanestimateandis
possibly quite inaccurate. Of the approximately 130 million
babies born in the world every year, there are more than one
million born with CHD. Little is known about the true vari-
ation in incidence around the world. Many studies seem to
reveal that there could be important geographical variations
in the incidence of CHD, as well as of specific heart malfor-
mations.
AlthoughthereareseveralCHDandbirthdefectsregistries
around the world, there have been few comprehensive anal-
yses of the differences between regional incidences world-
wide. Additionally, many landmark studies looking at the
incidence of CHD were performed before standardization of
diagnostic modalities, such as Doppler echocardiography.
Therefore, some contradictions are seen between those re-
sults and newer studies.
ongenital heart disease (CHD), the most common con-
genital birth defect, has long been known as one of the
Furthermore, the lack of systematic large studies is com-
pounded by the presence of different definitions of CHD and
diagnostic methods. The recent work of the International
Society for Nomenclature of Paediatric and Congenital Heart
Diseaseisleadingthenecessaryprocessforstandardizationof
the diagnostic coding methodology across institutions and
geographical borders with the establishment of the Interna-
tional Paediatric and Congenital Cardiac Code.2
It is safe to assume that it is important to study differences
in incidence in population groups because those variations
can be at the basis for identification of pathogenesis and risk
factors, both genetic and environmental. Indeed, even
though it is estimated that only 8% to 12% of CHD can be
accounted for by environmental factors,3it still represents a
significant burden of disease that could be potentially
avoided by controlling the causative agents. Such modifiable
risk factors include maternal alcohol consumption, rubella,
hydantoin, thalidomide, Accutane, poorly controlled insu-
lin-dependent diabetes, phenylketonuria.3
The challenges of CHD worldwide are numerous. First,
there is a need to properly identify the extent of this global
health problem by establishing its true incidence. Second,
inadequacies with access to diagnosis and uniformity of di-
agnostic modalities must be resolved. Finally, the care of
individuals with CHD is complex and very resource-inten-
sive. Access to care is not equal for all, depending on the
countries of origin and regions of the world. Efforts are being
made to guarantee some level of care to all children born
with CHD, but, needless to say, a tremendous amount of
work remains to be done. The recent establishment of the
World Society for Pediatric and Congenital Heart Surgery
Division of Cardiovascular Surgery, The Montreal Children’s Hospital of the
McGill University Health Centre, Montreal, Quebec, Canada.
AddresscorrespondencetoChristoI.Tchervenkov,MDCM,FRCSC,FACS,
Division of Cardiovascular Surgery, The Montreal Children’s Hospital of
the McGill University Health Centre, Montreal, Quebec, Canada; e-mail:
christo.tchervenkov@muhc.mcgill.ca
26
1092-9126/10/$-see front matter © 2010 Elsevier Inc. All rights reserved.
doi:10.1053/j.pcsu.2010.02.005

Page 2

(WSPCHS) will hopefully act as a catalyst for the global im-
provement of care for children and adults with CHD.
In this article, we will focus on the global challenge of
CHD. We will review the literature on the incidence of CHD
and, based on this analysis, try to establish regional differ-
ences. A careful evaluation of these studies will be done to
understand the reason for differences in incidence. From the
analysis of the incidence of CHD, we will crudely quantify its
global burden. Finally, we will provide some data on the
access to care for children and adults with CHD across the
world.
Methodology
The first step in identifying and understanding the problem
of CHD worldwide is to gather some objective data on the
epidemiology and demographic profile of this disease.
Many regional studies have been published to describe the
incidence of CHD in specific areas of the world. We con-
ducted a literature search of all articles indexed in PubMed,
Embase, and Ovid. The search terms used to conduct this
process were combinations of: “congenital”; “heart” or “car-
diac”; “disease” or “defect”; and “incidence” or “mortality.”
From the large amount of papers identified, we included 115
that presented data on CHD in the general population, either
at birth or on admission to a pediatric hospital, published up
toJuly2009andwithatleastanabstractinEnglishorFrench
(Fig.1).Studieslookingonlyatspecificgroupsofindividuals
or case series of studies of rare defect were excluded from the
analysis. We also identified institutional and governmental
surgical databases and registries detailing the nature and vol-
ume of cases operated on in various areas of the world. We
extracted data from those studies on the population, timing,
methods of detection, defects reported, incidence, and prev-
alence measured or observed.
In an effort to classify data in a geographical manner, we
indexed countries as grouped by the United Nations. Using
this classification, countries were grouped by continent.4
Also, using data published by the World Bank, countries
wereclassifiedbasedontheirincomegroup.TheWorldBank
divides economies according to their 2008 gross national
income per capita, calculated using the World Bank Atlas
method. The groups are: low income, $975 or less; lower
middle income, $976 to $3,855; upper middle income,
$3,856 to $11,905; and high income, $11,906 or more.5
To establish the global resources available in the treatment
of CHD, the WSPCHS carried out a worldwide survey in
2007. The aim of this survey was to identify all congenital
heart surgeons distributed around the globe and get an un-
derstanding of their type of practice.
Incidence of CHD and
Geographical Variations
To precisely identify the incidence of CHD, we performed a
literature search that yielded 115 studies for analysis. Of
those, 35 presented statistics about incidence in the general
population. The rest report diagnoses made in patients re-
ferred to tertiary care centers, thereby introducing the bias of
clinicalpresentationandaccesstocare(Fig.1).However,the
different study methods have a specific relevance: screening
newborns with echocardiography overestimates the number
of CHD that will be clinically significant, but is sensitive
enough to detect potential risk factors. On the other hand,
reportsofnumbersofcasespresentingtospecialistsoverlook
patientswithsubclinicaldiseaseornoaccesstoproperhealth
care resources, but gives useful information on current mor-
bidity and mortality of CHD.
The largest population studies stem from national and re-
gional databases of CHD. The largest of those is EUROCAT,
the European Registry of Congenital Anomalies. It combines
data about congenital defects at birth from 43 registries in 20
countries, with a study population of over 18 million births.
Data about birth incidence has been collected since 1973,
and data about access to cardiac surgery was added in 2005.
Raw data is available online, and a recent publication com-
piles results about CHD from the period 2000-2005.6The
average incidence of CHD found in live infants 1 week after
birth was 6.4 per 1,000. Three-quarters of that number are
ventricular septal defect (VSD), atrial septal defect (ASD),
and pulmonary vein stenosis (PVS), while the incidence of
Figure 1 Flow chart representing the selection of studies during the
literature review.
The Challenge of Congenital Heart Disease Worldwide27

Page 3

severe CHD in infants who survive their first week of life is
1.3 per 1,000, with 1.04 per 1,000 requiring surgery.
Another large registry is the ECLAMC, the Latin American
Study of Congenital Malformations. Founded in Buenos
Aires in 1967, it has now expanded to all South American
countries. Participation to the database is on a volunteer ba-
sis; member centers report congenital anomalies at birth
along with the next born baby of the same gender as a con-
trol. Whereas it has a large and varied study population, it
only reports malformations diagnosed at birth, which ex-
cludesmanycongenitalheartdefects.Thereportedincidence
of CHD on analysis stemming from their database is 1.2 per
1,000, significantly lower than in other studies.7While there
is no North American database looking primarily at the inci-
dence of CHD, some data can be obtained from surgical
databases such as the Society of Thoracic Surgeons (STS)
database, regional initiatives, and other surgical registries.
The Canadian Congenital Anomalies Surveillance System,
founded in 1966 after the thalidomide scandal, combines
data from provincial registries and direct analysis of hospital
discharge diagnoses entered into the government health
records.3Although the heterogeneity of the provincial re-
porting methods reduce its precision, it covers about
345,000 births annually and is thus a precious source of
information about rare defects. There are several regional
initiatives in the United States, supervised by the Centers for
DiseaseControl:theMetropolitanAtlantaCongenitalDefects
Program (MACDP), established in 1967; the National Birth
Defects Prevention Network; and 15 state-based registries.8
The MACDP data from 1998 to 2005 was published in 2008;
an average incidence of CHD of 8.14 per 1,000 births was
found.9The International Clearinghouse for Birth Defects
Surveillance and Research (ICBDSR) is an association
founded in 1974 with the mission to promote the collabora-
tionbetweenregionalregistries.10Theynowmonitorover40
defects,andatotalof3.5millionbirthseachyear.Whiletheir
main focus is the search for unknown teratogens, they can be
a useful resource in compiling data from multiple countries.
The ICBDSR publishes annual reports with the data from
individual monitoring agencies; they report the incidence of
transposition of the great arteries, tetralogy of Fallot (TOF),
hypoplastic left heart syndrome, and coarctation of the
aorta.11
In addition to data on the incidence of CHD in the general
population, a precious resource is the database of the Inter-
national Congenital Heart Surgery Nomenclature and Data-
base Project, a collaboration between the European Associa-
tion for Cardio-Thoracic Surgery and the Society of Thoracic
Surgeons (North America). The two Societies have estab-
lished a common nomenclature and coding scheme to pro-
cess data on the presentation, severity, and outcome of pa-
tients undergoing congenital cardiac surgery.2While these
databases do not specifically look at the incidence of CHD,
they capture substantial data on patients who underwent
surgery and therefore give an estimate of the access to care.
They also give a rough estimation of the incidence of some
lesions, which are obligatorily treated surgically, if we make
the assumption that access to care is universal.
Areviewofthestudiesextractedfromtheliteraturereveals
a reported incidence of CHD varying between 1.2 and 17 per
1,000 (Figs. 2 and 3, Tables 1 and 2) The study methods and
populations of the reports of incidence of CHD and the pro-
portions of different defects are variable enough to make it
difficult to draw conclusions about risk factors. The studies
found in the literature used different methods (including
clinical, echocardiographic, and pathologic), populations
(newbornsversusschool-agechildren,cohortsborninahos-
Incidence of CHD
0.00
Austria (42)
2.00
4.00
6.00
8.00
10.00
12.00
14.00
16.00
18.00
20.00
Belgium (28)
Brazil (43)
Brazil (44)
Canada (3)
Colombia (7)
Czech Republic (27)
EUROCAT (6)
France (21)
France (37)
Hong Kong (33)
Iceland (22)
India (30)
India (31)
India (32)
Italy (41)
Nepal (24)
Oman (35)
Spain (40)
Sweden (37)
Sweden (38)
Taiwan (34)
Turkey (36)
UK (23)
USA (45)
USA (17)
USA (37)
USA (46)
Incidence/ 1000
Figure 2 Incidence of CHD worldwide.
Distribution of Incidence of CHD across
29 studies
0
1
2
3
4
5
6
7
8
<33 to 55 to 7 7 to 99 to 11>11
Incidence of CHD/1000
Frequency
Figure 3 Distribution of incidence of CHD worldwide.
28P.-L. Bernier et al

Page 4

pital versus cases referred to a cardiologist or surgeon), and
definitions. Therefore, conclusions are difficult to draw on
potential causal factors or a trend in the worldwide inci-
dence. There are a few smaller studies, however, that draw
attention to a potentially significant higher risk for CHD in
certain populations. For instance, in 1995, Sadiq et al12stud-
ied the incidence of CHD in a tertiary care center in Birming-
ham, UK. In a sample of 1,111 patients, they noticed the
proportion of children with CHD of Asian descent was larger
than expected (17 versus 9 per 1,000 in the population).12In
addition,Asianchildrenhaveahigherproportionofcomplex
CHD. These findings have not been replicated, but would
warrant more careful research. Although challenging, in
some situations well-designed retrospective studies have
brought to light interesting conclusions. For example, an
analysis of the government health records in the Province of
Quebec, Canada around the time of addition of folic acid to
flour has shown a significant decrease in incidence of severe
CHD.13It is interesting to note that studies of populations
from low, lower-middle, and higher-middle income coun-
tries (based on the World Bank classification) seem to ob-
serve, except for some exceptions, a lower incidence of CHD
than the accepted value if 8 per 1,000 (Table 2). Although
many factors could explain this, one has to consider that a
lack of resources leading to difficult access to medical care,
and therefore diagnosis, could be central to this observation.
Also, it seems that if we look at the distribution of the inci-
dence of CHD in all the studies analyzed, the median clearly
Table 2 Incidence of CHD Worldwide (Classified by World
Bank Income Groups)
Country Incidence Per 1,000
Low Income
Dharan, Nepal24
Lower-Middle Income
New Delhi, India30
Kanpur, India31
New Delhi, India32
Higher-Middle Income
Anatolia, Turkey36
Columbia7
Londrina, Brazil43
Brazil44
High Income
Hong Kong33
Oman35
EUROCAT6
Bohemia, Czech Republic27
Sweden37
Gothenburg, Sweden38
United Kingdom39
Iceland22
Navarra, Spain40
Florence, Italy41
Central/Eastern France37
Tyrol, Austria42
Belgium28
Indre-Loire, France21
California, USA37
USA17
Atlanta, GA, USA45
USA46
Canada3
5.8
3.9
10.47
4.2
7.7
1.2
5.94
9.58
10.3
7.1
6.6
6.4
2.5
6.4
6.4
17.00
8.96
12.3
2.89
8.2
8.3
9.8
3.16
3.1
6.2
8.8
10.4
NOTE. Country groups obtained from the World Bank Data & Statistics
(http://web.worldbank.org/WBSITE/EXTERNAL/DATASTATISTICS).
Table 1 Incidence of CHD Worldwide (broken down by world
regions, as defined by the United Nations)
Country
Sample
SizeYears
Incidence
Per 1,000
South-Central Asia
New Delhi, India30
Kanpur, India31
New Delhi, India32
Dharan, Nepal24
Eastern Asia
Hong Kong33
Taiwan34
Western Asia
Oman35
Anatolia, Turkey36
Europe
EUROCAT6
Eastern Europe
Bohemia, Czech
Republic27
Northern Europe
Sweden37
Gothenburg,
Sweden38
United Kingdom39
Iceland22
Southern Europe
Navarra, Spain40
Florence, Italy41
Western Europe
Central/Eastern
France37
Tyrol, Austria42
Belgium28
Indre-Loire,
France21
South America
Columbia7
Londrina, Brazil43
Brazil44
Northern America
California, USA37
USA17
Atlanta, GA,
USA45
USA46
Canada3
10,964
10,641
11,833
14,461
1994 3.9
10.47
4.2
5.82006
20,928 1987-1989
45,725
10.3
32.5†
139,707 1994-1996
219,870 1995-2002
7.1
7.7
3,300,000 20096.6
91,8231980 6.4
1,270,000 1981-1992
58,105 1941-1950
2.5
6.4
690,215 1985-2004
44,013 1990-1999
6.4
17.00
47,783 1989-1998
46,896 1975-1984
8.96
12.3
950,000 1983-19922.89
41,725 1979-1983
111,225
26,082 1991-1994
8.2
8.3
9.8
2002
44,985 2001-2005
80,262 1989-1998
29,770 1990-2003
1.2
5.94
9.58
2,200,000 1985-1992
40,600,000 1979-1997
937,195 1968-1997
3.16
3.1
6.2
19,044
325,000
1978
1999
8.8
10.4
Point prevalence estimate obtained from (birth incidence*birth rate).
The populations were obtained from: Population Division of the
Department of Economic and Social Affairs of the United Nations
Secretariat, World Population Prospects: The 2008 Revision,
http://esa.un.org/unpp, Monday, October 05, 2009.
†Includes arrhythmias, myocarditis, Marfan’s syndrome.
The Challenge of Congenital Heart Disease Worldwide 29

Page 5

sits around 8 per 1,000 live births. This corresponds to the
value accepted by most over the years (Fig. 3).
Several studies have shown an increase in the incidence of
CHD, particularly VSDs, in the last 30 years, and it is a hotly
debatedsubjectwhetherthatrepresentsatrueincreaseinthe
prevalence of the condition or if it is due to better screening
tools.3,11,14,15Ontheotherhand,prenataldetectionofcardiac
malformations was found to be as high as 25% in some Eu-
ropean centers, and the proportion of aborted fetuses with
cardiac disease in hospital-based registries has been increas-
ing.16However, there has clearly been a decrease in mortality
fromCHDdocumentedinseveralstudies.Forexample,there
were half as many deaths due to CHD in children less than 5
years old in the period 1995-1997 when compared with
1979-1981 in the United States.17Mortality decreased across
all defects, although the change was modest for hypoplastic
left heart syndrome.18
Future directions of study include the maintenance of
large, homogeneous databases (such as the International
Heart Surgery Database and Nomenclature Project or the
International Clearinghouse) that will continue to monitor
for new risks factors for CHD. Another promising database is
the National Children’s Study, an ongoing study of 100,000
children from the United States who will be followed from
birth up to age 21.19The long duration of follow-up, as well
as the large cohort, will likely make it one of the richest
sources of information on the incidence of clinically signifi-
cant CHD.
Incidence and
Variation of Specific Lesions
The largest number of studies available comment on the
overall incidence of CHD and of the three most common
defects, VSD, ASD, and TOF (Tables 3-5; Figs. 4 and 5). We
excluded patent ductus arteriosus from the analysis because
its pathologic significance is highly dependent on gestational
age.Themostcommoncongenitalheartlesioninallpediatric
populations studied was VSD, with an incidence ranging
from 0.3 to 7.7 per 1,000. It represented 30% to 50% of all
defects, being two to three times more common than ASD,
which, with an incidence of 0.3 to 4.2 per 1,000, was the
second most frequent. The incidence of TOF varied between
0.16 and 0.46 per 1,000, but represented between 5% and
10% of pediatric cardiology patients in the majority of stud-
ies.10,20-22The incidence of other more rare defects, such as
atrio-ventricular septal defect (0.4 per 1,000), hypoplastic
left heart syndrome (0.2 per 1,000), PVS (0.34-1.28 per
1,000), total anomalous pulmonary venous return (0.04-
0.21 per 1,000), transposition of the great arteries (0.32 per
1,000), is presented in Table 5.
Table 3 Incidence of VSD
CountrySample Year
1995-1996
1994
1995-1997
1989-1998
Incidence Per 1,000
0.364
1.4
1.66
1.68
1.93
2
2.01
2.51
2.65
3.13
3.56
4
5.62
6.36
7.68
% of CHD
36%
35%
30%
28%
46%
28%
31%
33%
33%
Egypt (Referrals to a cardiologist)51
India30
Atlanta, GA, USA45
Brazil43
India32
Oman35
Bohemia, Czech Republic27
Turkey36
EUROCAT6
Taiwan34
Canada3
Nepal (patients in tertiary hospital)24
India (patients admitted to cardiology)31
France21
Iceland22
Saudi Arabia (children referred to hospital)48
Cote d’Ivoire49
Nigeria50
10,964
937,195
80,262
11,833
139,707
91,823
219,870
3,300,300
45,725
325,000
14,461
10,641
26,082
44,013
1994-1996
1980
1995-2002
2008
1999
2006
34%
70%
54%
64%
45%
32.5%
38.6%
27%-46%
1991-1994
1990-1999
1994-1996
1969-1976
1965-1970
604
259
260
Table 4 Incidence of ASD
Country
Nepal24
USA45
Brazil43
France21
Czech
Republic27
India32
Oman35
Taiwan34
Belgium28
EUROCAT6
Iceland22
India (pts on
Cardiology
wards)31
Canada3
Sample
Size
14,461
937,195 1968-1997
80,262 1989-1998
26,082 1991-1994
91,823
Year
2006
Incidence
ASD (per
1,000)
0.3
0.46
0.475
0.5
0.73
ASD/
CHD
5%
8%
8.10%
5%
11%1980
11,833
139,707 1994-1996
45,725
111,225
3,300,000
44,013 1990-1999
10,461
0.76
0.99
1.25
1.46
1.6
2.03
4.99
18%
14%
4%
16%
20%
12%
2002
325,00019994.23 41%
30P.-L. Bernier et al