RESEARCH ARTICLE

Mortality from Congenital Heart Disease in

Mexico: A Problem on the Rise
José Luis Torres-Cosme1, Constanza Rolón-Porras1, Mónica Aguinaga-Ríos2, Pedro
Manuel Acosta-Granado3, Enrique Reyes-Muñoz4, Teresa Murguía-Peniche1,5*
1 Research Division, Community Interventions, Instituto Nacional de Perinatología ‘Isidro Espinosa de los
Reyes’, Mexico City, Mexico, 2 Genetics Department, Research Division, Instituto Nacional de Perinatología
‘Isidro Espinosa de los Reyes’, Mexico City, Mexico, 3 Neurosciences Department, Instituto Nacional de
Perinatología ‘Isidro Espinosa de los Reyes’, Mexico City, Mexico, 4 Endocrinology Department, Instituto
Nacional de Perinatología ‘Isidro Espinosa de los Reyes’, Mexico City, Mexico, 5 Faculty of Health Sciences,
Universidad Panamericana, Mexico City, Mexico

* [email protected]

Abstract
Background and Objectives
Temporal trends in mortality from congenital heart disease (CHD) vary among regions. It is
therefore necessary to study this problem in each country. In Mexico, congenital anomalies
were responsible for 24% of infant mortality in 2013 and CHD represented 55% of total
deaths from congenital anomalies among children under 1 year of age. The objectives of
OPEN ACCESS
this study were to analyze the trends in infant mortality from CHD in Mexico (1998 to 2013),
Citation: Torres-Cosme JL, Rolón-Porras C, its specific causes, age at death and associated socio-demographic factors.
Aguinaga-Ríos M, Acosta-Granado PM, Reyes-
Muñoz E, Murguía-Peniche T (2016) Mortality from
Congenital Heart Disease in Mexico: A Problem on Methods
the Rise. PLoS ONE 11(3): e0150422. doi:10.1371/ Population-based study which calculated the compounded annual growth rate of death rom
journal.pone.0150422
CHD between 1998 and 2013. Specific causes, age at which death from CHD occurred and
Editor: Fatima Crispi, University of Barcelona, SPAIN risk factors associated with mortality were analyzed for the year 2013.
Received: July 14, 2015

Accepted: February 12, 2016 Results

Published: March 3, 2016 Infant mortality from CHD increased 24.8% from 1998 to 2013 (114.4 to 146.4/ 100,000 live
births). A total of 3,593 CHD deaths occurred in 2013; the main causes were CHD with left-
Copyright: © 2016 Torres-Cosme et al. This is an
open access article distributed under the terms of the to-right shunt (n = 487; 19.8/100,000 live births) and cyanotic heart disease (n = 410; 16.7/
Creative Commons Attribution License, which permits 100,000). A total of 1,049 (29.2%) deaths from CHD occurred during the first week of life.
unrestricted use, distribution, and reproduction in any Risk factors associated with mortality from CHD were, in order of magnitude: non-institu-
medium, provided the original author and source are
credited.
tional birth, rural area, birth in a public hospital and male sex.

Data Availability Statement: Data are available

from: http://www.sinais.salud.gob.mx/basesdedatos/
Conclusions
std_defunciones.html. Mortality from CHD has increased in Mexico. The main causes were CHD with left-to-right
Funding: The authors have no support or funding to shunt, which are not necessarily fatal if treated promptly. Populations vulnerable to death
report. from CHD were identified. Approximately one-third of the CHD occurred during the first
Competing Interests: The authors have declared week of life. It is important to promote early diagnosis, especially for non-institutional births.
that no competing interests exist.

PLOS ONE | DOI:10.1371/journal.pone.0150422 March 3, 2016 1 / 16

 Mortality from Congenital Heart Disease in Mexico

Introduction
Worldwide, mortality of children under 5 years of age decreased 47% from 1990 to 2015, from
90.6 to 42.5 deaths per 1,000 live births [1]. This was achieved primarily by decreasing diarrheal
and respiratory infections as well as diseases that can be prevented by vaccinations. Meanwhile,
congenital anomalies have become a more significant cause of death among children under 5
years of age, increasing from 5% of deaths in 2000 to 7% in 2013; and this increase was even
larger in the Americas, from 15 to 21% during this same period [2]. Furthermore, congenital
heart disease accounts for nearly one-third of congenital anomalies [3].
Temporal trends in congenital heart disease (CHD) vary among countries. Mortality from
CHD decreased over recent years in Canada and in 16 European countries, particularly for
children under 1 year of age [4–6]. A decrease in mortality from CHD was also observed in the
United States (U.S.) between 1970 and 1997, contributing to a 59% reduction in infant mortal-
ity during this period [7]. In contrast, the overall mortality rate from CHD in China increased
62% between 2003 and 2010 [8].
Mortality from CHD has increased in some countries, which may actually reflect a higher
number of CHD patients diagnosed and registered at birth and higher necropsy rates for peri-
natal deaths, rather than an increase in the prevalence of the disease. This deserves further
research.
In Mexico, significant efforts have been undertaken over recent years to decrease infant
mortality among children under 1 year of age. Thus, between 1990 and 2010 the infant mortal-
ity rate decreased from 24.0 to 11.8 per 1000 live births [9]. Nevertheless, mortality from CHD
was found to increase in both absolute and relative terms. In the year 2000, CHD caused the
death of 2,596 children under 1 year of age, representing 6.7% of total deaths in this age group.
By 2008, mortality from CHD increased to 2,848, representing 9.6% of the total [10].
Current and detailed information is required to identify national mortality trends related to
all CHD in the infant population, and needs to be based on the ICD-10 in order to make inter-
national comparisons. Information is also needed that is useful to those responsible for making
decisions about children’s public health in Mexico as well as to international health agencies,
about a problem which seems to be on the rise in developing nations and which is a global
health concern.
The main objective of the present study is to analyze trends in mortality from CHD in
Mexico between 1998 and 2013. Secondary objectives are identifying the principal causes of
death from CHD, determining the age at death from CHD and analyzing socio-demographic
factors associated with death from CHD.

Methods
Study period and variables included
This is a population-based study conducted in Mexico which analyzes trends in infant mortal-
ity from CHD from 1998—the year when this country began to use the 10th revision of the
ICD (ICD-10) [11]—until 2013. This made it possible to obtain a more detailed and specific
description of CHD. For the purpose of international comparisons, all deaths from CHD corre-
sponding to ICD-10 codes Q20-Q28 were included [12]. The study also included the following
variables contained in the death certificates: sex, age, primary place of residence, state, size of
locality and place where the death occurred. These data did not allow for estimating the pro-
portion of preterm patients or determining diagnoses associated with chromosomal abnormal-
ities. Fetal deaths were not included in the analysis because Mexico does not have a national
surveillance program to collect all diagnoses of fetal cardiac disease.

PLOS ONE | DOI:10.1371/journal.pone.0150422 March 3, 2016 2 / 16

 Mortality from Congenital Heart Disease in Mexico

The number of surgeons, cardiologists and pediatricians registered in Mexico in 2003 and
2012 were included in the analysis; these are the years with known information and were used
for comparison purposes.

Data sources
The information was obtained from all death certificates. These are routinely coded by the
General Department of Health Information (DGIS, Spanish acronym) which produces elec-
tronic databases available to the public through the National System for Health Information
(SINAIS, Spanish acronym), at the DGIS webpage [10]. The data for number of live registered
births (LB) and socio-demographic information were obtained from the webpage provided by
the National Institute of Statistics and Geography (INEGI, Spanish acronym) [13].
The number of surgeons, cardiologists and pediatricians registered in Mexico corresponds
to the job positions reported by public hospitals and the Social Security Institute, according to
the information from the DGIS [14]. This information was used to calculate rates of specialists
per 1,000 LB.

Operational definitions
Rates were calculated based on the number of deaths from CHD analyzed by this study for
each specific year divided by the number of LB for each corresponding year, multiplied by
100,000.
The Functional Classification of CHD proposed by Reller et al [15] was adapted for this spe-
cific analysis, which divides heart defects into four subgroups: left-to-right shunt, cyanotic
heart disease, obstructive defects of the left heart and obstructive defects of the right heart. All
cases of patent ductus arteriosus were included since gestational age was not available.
The classification proposed by the 2012 Mexican National Health and Nutrition Survey [16]
was used to define the regional geographic distribution of the 32 states in the country, as fol-
lows: North (Baja California, Baja California Sur, Coahuila, Chihuahua, Durango, Nuevo León,
Sinaloa, Sonora, Tamaulipas and Zacatecas); Central (Aguascalientes, Colima, Guanajuato,
Hidalgo, Jalisco, Mexico State, Michoacán, Nayarit, Querétaro, San Luis Potosí and Tlaxcala);
Mexico City (Federal District); and South (Campeche, Chiapas, Guerrero, Morelos, Oaxaca,
Puebla, Quintana Roo, Tabasco, Veracruz and Yucatán). Rural areas were defined as residential
localities with 2,500 or less inhabitants and urban areas as regions with a larger population [9].

Statistical analysis
1. The trend in CHD mortality rates per 100,000 LB was determined for the years 1998 to
2013. The CHD mortality rate and the compounded annual rate of growth (CARG) were
calculated to determine the accumulated adjusted change in the rate for the period 1998–
2013, according to the following formula: [final value/beginning value][1/number of years]
2. In 2013, the total deaths from CHD that occurred during the first year of life were calculated
by postnatal period, classified by our data source as follows: first 24 hours of life, 1 to 6 days
of life, 7 to 27 days of life and 28 days to 11 months of life. The number of deaths from CHD
per day was estimated for each period.
3. For the year 2013, a univariate analysis was performed [(odds ratio (OR); 95% confidence
interval (95%CI)] and the Mantel-Haenszel X2 was used to identify the factors associated
with deaths from CHD. A value of p<0.05 was considered statistically significant.

PLOS ONE | DOI:10.1371/journal.pone.0150422 March 3, 2016 3 / 16

 Mortality from Congenital Heart Disease in Mexico

4. IBM SPSS Statistics 19 statistical package was used for the statistical analysis [17]. OpenEpi
version 3.03 was used to calculate the odds ratio and confidence intervals [18].

Results
Trends in mortality from CHD and descriptive analysis
A total of 41,717,421 births were registered from 1998 to 2013 and 50,759 deaths from CHD were
diagnosed, resulting in a mortality rate for the period of 121.7 per 100,000 LB. Fig 1 shows an
increasing trend in the CHD mortality rate over time for children under 1 year of age in Mexico—
from 114.4 per 100,000 LB in the year 1998 to 146.4 per 100,000 LB in the year 2013, an increase
of 24.8%. Fig 1 also shows a higher overall CHD mortality rate for males than for females. While
the mortality rate for both sexes increased from 1998 to 2013, the increase was greater for females.
Table 1 presents the change in the CHD infant mortality rate from 1998 to 2013 in the 32
states in Mexico. Durango and Nayarit registered the highest increases in mortality from CHD
(roughly 150 to 180%) while 11 states registered decreases in the CHD mortality rate. In the
year 2013, a total of 1,876 deaths from CHD (52%) occurred in Mexico City and five states
(State of Mexico, Jalisco, Puebla, Guanajuato and Veracruz).
Fig 2 presents the distribution of the CHD mortality rates in the study regions for the year
2013.
Table 2 presents some of the socio-demographic characteristics of infants who died from
CHD in 2013. Mexico City reported the highest rate (188.5 per 100,000 LB) and the North
reported the lowest (126 per 100,000 LB) for that year.

Fig 1. Infant mortality rate from congenital heart disease by sex. Deaths per 100,000 live registered births, Mexico 1998–2013.
doi:10.1371/journal.pone.0150422.g001

PLOS ONE | DOI:10.1371/journal.pone.0150422 March 3, 2016 4 / 16

 Mortality from Congenital Heart Disease in Mexico

Table 1. Mortality rate from congenital heart disease by state, Mexico 1998 and 2013.

State 1998 2013 CARG†

Rate per 100,000 LB* (n) Rate per 100,000 LB* (n) 1998–2013
1 Aguascalientes 138.5 35 106.6 28 -26.0
2 Baja California 141.5 82 132.0 81 -6.9
3 Baja California Sur 110.0 11 104.9 13 -4.8
4 Campeche 90.7 17 195.7 37 78.9
5 Coahuila 88.2 52 104.2 65 16.8
6 Colima 229.9 29 60.5 8 -127.7
7 Chiapas 63.1 84 87.5 143 33.0
8 Chihuahua 90.5 72 131.8 87 38.1
9 Mexico City 162.3 293 188.5 272 15.0
10 Durango 32.7 15 132.4 52 146.5
11 Guanajuato 161.4 216 145.6 172 -10.3
12 Guerrero 39.9 39 100.0 102 94.7
13 Hidalgo 81.0 54 127.8 75 46.3
14 Jalisco 140.7 231 163.9 260 15.3
15 State of Mexico 155.9 517 215.4 681 32.7
16 Michoacán 74.1 91 111.7 117 41.6
17 Morelos 161.8 55 131.8 49 -20.4
18 Nayarit 27.8 7 149.5 36 178.1
19 Nuevo León 143.4 130 137.9 129 -3.9
20 Oaxaca 92.4 98 124.1 114 29.7
21 Puebla 121.6 186 202.1 280 51.7
22 Querétaro 176.2 68 151.7 63 -14.9
23 Quintana Roo 135.3 30 136.3 38 0.7
24 San Luis Potosí 127.2 84 118.0 64 -7.4
25 Sinaloa 46.4 34 115.6 65 94.1
26 Sonora 113.1 63 127.7 67 12.2
27 Tabasco 117.2 58 148.1 77 23.6
28 Tamaulipas 92.9 59 134.0 80 37.1
29 Tlaxcala 125.0 35 122.8 33 -1.8
30 Veracruz 83.5 181 132.2 211 46.7
31 Yucatán 150.8 57 143.5 55 -4.9
32 Zacatecas 93.0 36 111.9 39 18.5
National 114.4 3019 146.4 3593 24.8

*LB = Live registered births

†CARG = Compounded annual rate of growth
Source: INEGI/SSA

doi:10.1371/journal.pone.0150422.t001

The majority of births as well as deaths from CHD occurred in the urban area. In 2013,
there were roughly four times more deaths from CHD in this area than in the rural area. Never-
theless, the mortality rates from CHD were higher in the rural area.
In the year 2013, a total of 69.8% of infant deaths from CHD occurred in public hospitals
and 6.7% in private hospitals. The latter had the lowest CHD mortality rate while home births
had the highest rate—22.3% of total deaths from CHD. Home and other non-institutional
births represented 6.8% of the total births.

PLOS ONE | DOI:10.1371/journal.pone.0150422 March 3, 2016 5 / 16

 Mortality from Congenital Heart Disease in Mexico

Fig 2. Infant mortality rate from congenital heart disease by region. Deaths per 100,000 live registered births, Mexico, 2013. Reprinted from Instituto de
Geografía de la Universidad Nacional Autonóma de México under the Creative Commons Attribution License CC BY 3.0, with permission from Stephan
André Couturier, original copyright 2015.
doi:10.1371/journal.pone.0150422.g002

In the year 2003, the rate of pediatricians, surgeons and cardiologists per 1,000 LB was 2.4,
1.9 and 0.3, respectively. These rates increased to 4.0, 3.6 and 0.6 by the year 2012, respectively.

Specific causes of death from CHD

Fig 3 presents the temporal analysis of specific and individual causes of death from CHD for
the period 1998–2013. There was an increase in infant deaths from CHD with left-to-right
shunt—specifically, patent ductus arteriosus and ventricular septal defect—which were the
main causes of death in 2013, followed by discordant ventriculo-arterial connections and
obstruction of the left heart.
According to an analysis of causes of death from CHD for the year 2013, based on the adap-
tation of the functional classification proposed by Reller et al [15], CHD with left-to-right
shunt were the main causes of death in the first year of life. Topping the list were patent ductus
arteriosus, ventricular septal defect and auricular septal defect (Table 3). Cyanotic heart dis-
eases, as a group, took second place, most notably discordant ventriculo-arterial connection
and Tetralogy of Fallot.

CHD deaths by age at death

Table 4 presents the distribution of the specific causes of death from CHD by age at death for
the year 2013. Patent ductus arteriosus, obstructive defects of the left heart, discordant ventri-
culo-arterial connections represented the majority of the known causes of early neonatal death

PLOS ONE | DOI:10.1371/journal.pone.0150422 March 3, 2016 6 / 16

 Mortality from Congenital Heart Disease in Mexico

Table 2. Social and demographic characteristics of infant deaths from congenital heart disease. Mexico 2013.

Variable CHD deaths*(n) Births (n) Rate†

Total 3,593 2,454,929 146.4
Sex
Female 1,638 1,212,556 135.1
Male 1,946 1,242,232 156.7
NS¶ 9 141 -
Regions
North 678 538,212 126.0
Central 1,537 942,608 163.1
Mexico City 272 144,278 188.5
South 1,106 829,831 133.3
Area
Urban 2,813 2,047,323 137.4
Rural 703 218,429 321.8
NS¶ 77 189,177 -
Birth site
Public hospital 2,508 1,829,008 137.1
Private hospital 241 317,113 76.0
Home and others 802 167,104 479.9
NS¶ 42 141,703

* CHD = Congenital heart disease

† Rate per 100,000 live registered births
¶NS = Not-specified
Source: INEGI/SSA

doi:10.1371/journal.pone.0150422.t002

(under 7 days of life). Patent ductus arteriosus was the main cause of death during the neonatal
period as well as during the first year of life. Ventricular septal defect, patent ductus arteriosus
and coarctation of the aorta were the main causes during the post-neonatal period. Over 60%
of the death certificates did not contain a specific diagnosis of the CHD, and this information
was missing on 90% of deaths occurring on the first day of life.
In 2013, approximately 10 children under 1 year of age died from CHD per day, on average.
Most of the deaths per day occurred on the day of birth—351 deaths per day, representing approx-
imately 10% of total deaths—followed by the period corresponding to day 1 through 6 of life with
an average of 116 deaths per day from CHD, representing 19.4% of total deaths from CHD.

Factors associated with mortality from CHD

Factors (OR; 95% CI) associated with mortality from CHD were, in order of magnitude: non-
institutional birth (6.34;5.49–7.32) as compared to private hospital, rural area (2.35;2.16–2.55)
as compared to urban area, birth in public hospital (2.35;2.16–2.55) as compared to private
hospital, and male sex (1.16;1.09–1.24) as compared to females.

Discussion
Trends in mortality
In 2013 in Mexico, congenital anomalies were responsible for 24% of infant mortality and
CHD represented 55% of total deaths from congenital anomalies among children under 1 year

PLOS ONE | DOI:10.1371/journal.pone.0150422 March 3, 2016 7 / 16

 Mortality from Congenital Heart Disease in Mexico

Fig 3. Main causes of infant mortality from congenital heart disease, Mexico 1998–2013.
doi:10.1371/journal.pone.0150422.g003

of age [10]. From 1998 to 2013, a significant increase (24.8%) in mortality from CHD among
children under 1 year of age was detected. And by the year 2013, the CHD mortality rate for
children under 1 year of age was 146.4 per 100,000 LB—3.5 times higher than that reported for
the U.S. from 1999 to 2006 (41.5 per 100,000 LB) [19]. Two interesting studies that have inves-
tigated infant mortality from CHD in Mexico also show that it is on the rise. Nevertheless, one
[20] of these only included Mexican municipalities presenting high mortality from congenital
anomalies and the other [21] did not include CHD corresponding to International Classifica-
tion of Diseases (ICD-10) code Q25 (congenital anomalies of great arteries), which constitutes
a significant proportion (11.5%) of deaths from CHD among children under 1 year of age.
These studies also did not include updated information from the past decade in Mexico, nor
did they include a detailed analysis of the specific causes of CHD or the age at death. Thus, the
information provided by the work herein is very important because it gives a current and com-
plete national overview of this public health problem.
Also notable is that mortality from CHD increased roughly 150 to 180% in some of the
states. This information is important to improve planning related to health services in order to
address the demand created by this serious public health problem. Furthermore, it is important

PLOS ONE | DOI:10.1371/journal.pone.0150422 March 3, 2016 8 / 16

 Mortality from Congenital Heart Disease in Mexico

Table 3. Functional classification of deaths from congenital heart disease in children under 1 year of age. Rates and frequencies, Mexico 2013.

ICD- Cause Male Female NS* Total Male Female Total

10 n n n n rate† rate† rate†

I LEFT TO RIGHT SHUNT 250 237 0 487 20.1 19.5 19.8
Q25.0 Patent ductus arteriosus 107 128 0 235 8.6 10.6 9.6
Q21.0 Ventricular septal defect 78 64 0 142 6.3 5.3 5.8
Q21.1 Atrial septal defect 38 22 0 60 3.1 1.8 2.4
Q21.2 Atrioventricular septal defect 25 17 0 42 2.0 1.4 1.7
Q21.9 Congenital malformations of cardiac 0 5 0 5 0.0 0.4 0.2
septum, unspecified
Q21.4 Aortopulmonary septal defect 1 1 0 2 0.1 0.1 0.1
Q21.8 Other congenital malformations of 1 0 0 1 0.1 0.0 0.0
cardiac septum
II CYANOTIC CONGENITAL 239 171 0 410 19.2 14.1 16.7
HEART DEFECTS
Q20.3 Discordant ventriculo-arterial 68 28 0 96 5.5 2.3 3.9
connections
Q21.3 Tetralogy of Fallot 41 40 0 81 3.3 3.3 3.3
Q20.4 Double inlet ventricle 35 20 0 55 2.8 1.6 2.2
Q26.4 Anomalous pulmonary venous 22 19 0 41 1.8 1.6 1.7
connection, unspecified
Q20.0 Common arterial trunk 18 13 0 31 1.4 1.1 1.3
Q26.2 Total anomalous pulmonary venous 17 14 0 31 1.4 1.2 1.3
connection
Q22.5 Ebstein anomaly 10 18 0 28 0.8 1.5 1.1
Q22.4 Congenital tricuspid stenosis 16 7 0 23 1.3 0.6 0.9
Q20.9 Congenital malformations of cardiac 5 5 0 10 0.4 0.4 0.4
chambers and connections,
unspecified
Q20.8 Other congenital malformations of 3 3 0 6 0.2 0.2 0.2
cardiac chambers and connections
Q20.5 Discordant auriculoventricular 3 1 0 4 0.2 0.1 0.2
connection
Q20.1 Transposition of the great vessels in 1 3 0 4 0.1 0.2 0.2
the right ventricle
III LEFT HEART OBSTRUCTIVE DEFECTS 131 96 0 227 10.5 7.9 9.2
Q23.4 Hypoplastic left heart 58 34 0 92 4.7 2.8 3.7
syndrome
Q25.1 Coarctation of aorta 41 38 0 79 3.3 3.1 3.2
Q25.4 Other congenital malformations of 12 10 0 22 1.0 0.8 0.9
aorta
Q23.0 Congenital stenosis of aortic valve 10 8 0 18 0.8 0.7 0.7
Q23.2 Congenital mitral stenosis 4 2 0 6 0.3 0.2 0.2
Q23.3 Congenital mitral insufficiency 3 2 0 5 0.2 0.2 0.2
Q23.9 Congenital malformations of aortic 1 2 0 3 0.1 0.2 0.1
and mitral valves, unspecified
Q23.1 Congenital insufficiency of aortic 2 0 0 2 0.2 0.0 0.1
valve
IV RIGHT HEART OBSTRUCTIVE 60 59 0 119 4.8 4.9 4.8
DEFECTS
(Continued)

PLOS ONE | DOI:10.1371/journal.pone.0150422 March 3, 2016 9 / 16

 Mortality from Congenital Heart Disease in Mexico

Table 3. (Continued)

ICD- Cause Male Female NS* Total Male Female Total

10 n n n n rate† rate† rate†

Q25.5 Atresia of pulmonary artery 32 32 0 64 2.6 2.6 2.6
Q22.8 Other congenital malformations of 10 3 0 13 0.8 0.2 0.5
tricuspide valves
Q22.6 Hypoplastic right heart syndrome 6 7 0 13 0.5 0.6 0.5
Q22.0 Pulmonary valve atresia 6 6 0 12 0.5 0.5 0.5
Q25.6 Stenosis of pulmonary artery 4 4 0 8 0.3 0.3 0.3
Q22.1 Congenital pulmonary valve stenosis 1 3 0 4 0.1 0.2 0.2
Q25.7 Other congenital malformations of 1 2 0 3 0.1 0.2 0.1
pulmonary artery
Q22.3 Other congenital malformations of pulmonary valve 0 2 0 2 0.0 0.2 0.1
NON CLASSIFICABLE 60 35 0 95 4.8 2.9 3.9
Q24.8 Other congenital heart malformations 51 32 0 83 4.1 2.6 3.4
Q24.6 Congenital heart block 4 0 0 4 0.3 0.0 0.2
Q24.5 Malformation of coronary vessels 3 0 0 3 0.2 0.0 0.1
Q26.9 Congenital malformations of great veins, 2 3 0 5 0.2 0.2 0.2
unspecified
Q27 Other congenital malformations of 3 4 0 7 0.2 0.3 0.3
peripheral vascular system
Q28 Other congenital malformations of 2 3 0 5 0.2 0.2 0.2
circulatory system
Q24.9 Congenital malformations of heart, 1201 1033 9 2243 96.7 85.2 91.4
unspecified
Total 1946 1638 9 3593 156.7 135.1 146.4

*NS = Not specified

†Rate per 100,000 live registered births
Source: INEGI/SSA

doi:10.1371/journal.pone.0150422.t003

to perform epidemiological studies to understand the high risk of CHD observed in certain
geographic areas. In 2013, 52% percent of the deaths from CHD occurred in Mexico City and
five states. It is of interest that lower-income states (Guerrero, Oaxaca Chiapas and others) had
lower CHD mortality rates than states with a higher gross domestic product per capita (Mexico
City, Mexico State, Campeche) [9]. This could be related to either the presence of high risk
genetic or environmental factors in the latter states or, alternatively, it might represent under-
reporting of deaths in states with fewer resources. A study by Hernandez et al [22] detected
that 22.6% (95% confidence interval (CI):12.3–36.2) of deaths among children under 5 years of
age were underreported in Mexican municipalities with a very low human development index.
Although we do not have a clear explanation for the observed trend in CHD mortality, risk
factors associated with CHD are on the rise in Mexico. For example, diabetes and substance
abuse—both known to be associated with CHD [16, 23–25]—are more prevalent now in
women of reproductive age than 10 years ago. [23–25] Also, the number of specialists per cap-
ita increased over recent years; thus, it is possible that CHD deaths that previously went unde-
tected are now diagnosed and registered as causes of infant mortality. This deserves further
research. Meanwhile, factors that reduce the risk of CHD [26] need to be promoted. These
include: perinatal consumption of folic acid; avoidance of smoking, alcohol and drugs by the

PLOS ONE | DOI:10.1371/journal.pone.0150422 March 3, 2016 10 / 16

 Mortality from Congenital Heart Disease in Mexico

Table 4. Main causes of infant mortality from congenital heart disease by age at death (frequencies and proportions). Mexico 2013.

ICD-10 Cause Less than 24 1–6 days 7–27 days 28 days-11 Less than 1
hs months year

n (%) n (%) n (%) n (%) N (%)

1 Q25.0 Patent ductus 4 1.1 45 6.4 93 11.9 93 5.3 235 6.5
arteriosus
2 Q21.0 Ventricular septal 1 0.3 9 1.3 27 3.5 105 6.0 142 4.0
defect
3 Q20.3 Discordant ventriculo- 3 0.9 19 2.7 30 3.8 44 2.5 96 2.7
arterial connections
4 Q23.4 Hypoplastic left heart syndrome 5 1.4 35 5.0 28 3.6 24 1.4 92 2.6
5 Q21.3 Coarctation of aorta 2 0.6 9 1.3 14 1.8 56 3.2 81 2.3
6 Q25.1 Tetralogy of Fallot 1 0.3 15 2.1 31 4.0 32 1.8 79 2.2
7 Q25.5 Atresia of pulmonary artery 3 0.9 12 1.7 17 2.2 32 1.8 64 1.8
8 Q21.1 Atrial septal defect 1 0.3 8 1.1 19 2.4 32 1.8 60 1.7
9 Q20.4 Double inlet ventricle 0 0.0 15 2.1 7 0.9 33 1.9 55 1.5
10 Q21.2 Atrioventricular septal defect 0 0.0 4 0.6 8 1.0 30 1.7 42 1.2
11 Q26.4 Common arterial trunk 0 0.0 5 0.7 7 0.9 29 1.6 41 1.1
12 Q20.0 Anomalous pulmonary, 0 0.0 0 0.0 10 1.3 21 1.2 31 0.9
venous connection
unspecified
13 Q26.2 Congenital tricuspid 1 0.3 3 0.4 5 0.6 22 1.2 31 0.9
Stenosis
14 Q22.5 Ebstein anomaly 2 0.6 6 0.9 10 1.3 10 0.6 28 0.8
15 Q22.4 Congenital stenosis of aortic valve 0 0.0 1 0.1 6 0.8 16 0.9 23 0.6
Subtotal 23 6.6 186 26.6 312 39.9 579 32.9 1100 30.6
Q24.9 Congenital 310 88.3 471 67.5 408 52.2 1054 59.8 2243 62.4
malformation of heart, unspecified
The rest of Q20-Q28 18 5.1 41 5.9 62 7.9 129 7.3 250 7.0
Total 351 100 698 100 782 100 1762 100 3593 100

Source: INEGI/SSA

doi:10.1371/journal.pone.0150422.t004

mother; control of maternal obesity and diabetes; and avoidance of taking medications begin-
ning early in pregnancy (certain types of analgesics and medications proven to be teratogenic).

Causes of death
The main causes of death from CHD are due to their prevalence and lethality. In Mexico, CHD
with left-to-right shunt were the main causes of mortality among children under 1 year of age
in 2013. Although CHD with left-to-right shunt are not more severe than others types—such
as critical and complex CHD—they are more prevalent [3], which may explain our findings.
Patent ductus arteriosus was the leading cause of mortality from CHD. Although it is
known that patent ductus arteriosus could be a physiological manifestation of extreme prema-
turity, when a death certificate indicates this as a cause of death it is clearly due to a pathologi-
cal problem that would have received treatment [27]. Therefore, this study included patent
ductus arteriosus among the types of CHD responsible for infant mortality, although with the
limitation that information about the gestational age of patients with specific CHD was not
available for the study population. As has been identified in different populations, the

PLOS ONE | DOI:10.1371/journal.pone.0150422 March 3, 2016 11 / 16

 Mortality from Congenital Heart Disease in Mexico

persistence of patent ductus arteriosus is a public health problem. Prevalence studies at the
global level and in other countries have reported patent ductus arteriosus to be among the
three main causes of congenital heart disease [3]. In a population study performed by Gilboa
et al [19] in the U.S., patent ductus arteriosus was the fourth cause of infant mortality from
CHD (rate of 1.3/100,000 LB) between 1999 and 2006. This indicates the importance of patent
ductus arteriosus during the first year of life even in developed countries. In a report on Mexico
by Cervantes-Salazar et al [28], patent ductus arteriosus and ventricular septal defects were the
CHD that most frequently required a surgical intervention, according to the Mexican Registry
of Pediatric Heart Surgery.
Ventricular septal defects became an increasingly more important cause of mortality from
CHD over time, representing the second highest rate (5.8/100,000 per LB) by the year 2013.
Given its mortality rate of 1.1/100,000 LB in the U.S. [19], this problem should also be
addressed globally, since mortality should not be high when detected early and managed
appropriately.
Cyanotic CHD, such as discordant ventriculo-arterial connections, and obstructive defects
of the left heart, such as left ventricular hypoplasia and coarctation of the aorta, were among
the six main causes of mortality from CHD. The latter are complex CHD which require special-
ized care and could be associated with worse outcomes. This is especially true for left ventricu-
lar hypoplasia, which is the main cause of mortality from CHD among children under 1 year of
age in the U.S., with a mortality rate of 8.6/100,000 LB [19].

Age at death
In the year 2013, a total of 3,593 deaths from CHD occurred in Mexico, approximately 10% of
which occurred on the first day of life. The analysis of the causes of death during the first 24
hours found that roughly 90% did not have a specific diagnosis, which is a cause for concern.
For those that did have a diagnosis, left ventricular hypoplasia, patent ductus arteriosus, discor-
dant ventriculo-arterial connection and pulmonary valve atresia were the leading causes. The
lack of a specific diagnosis is a significant weakness in our hospital information system and,
therefore, prenatal and early postnatal diagnosis in the first hours of life should be encouraged.
This should include a complete physical exam and universal screening for CHD using pulse
oximetry [29], a non-invasive procedure which is increasingly used for the timely diagnosis of
CHD. Major CHD can be diagnosed during the prenatal period, which provides the opportu-
nity to adequately plan postnatal treatment for these patients. In a systematic review of the lit-
erature performed by Holland et al [30], newborns with a prenatal diagnosis of critical
congenital heart disease were significantly less likely to die prior to planned cardiac surgery
than were those with a comparable postnatal diagnosis (OR, 0.26; 95% CI: 0.08–0.84). Also, a
recent paper published by Eckersley et al [31] about a population study in New Zealand showed
that early (antenatal or pre-discharge) diagnosis of critical CHD had better outcomes than late
diagnosis (post-discharge). This study reported a higher mortality for late diagnosis than for
early diagnosis (27% vs 16%, respectively; p<0.04).
Patent ductus arteriosus was the main cause of death from CHD during the neonatal period.
The main causes of CHD during the post-neonatal period were ventricular septal defects, patent
ductus arteriosus and coarctation of the aorta. A specific diagnosis of the cause of death during
these periods was lacking in more than half of the cases. This may indicate a failure to adequately
fill out the death certificate or a diagnostic failure by the echocardiography. These problems need
to be improved in order to understand the epidemiology of mortality from CHD in Mexico. The
absence of this data has also been reported in developed countries. For instance, in the U.S. a spe-
cific diagnosis was found to be lacking in as many as 34% of the CHD [19].

PLOS ONE | DOI:10.1371/journal.pone.0150422 March 3, 2016 12 / 16

 Mortality from Congenital Heart Disease in Mexico

The number of deaths from CHD per day reported for different ages indicates the impor-
tance of the first week of life as a critical period, when the majority of deaths from CHD
occurred, and particularly the first day of life. This situation should enable specialized health
teams to design and plan strategies to determine diagnoses during prenatal and early neonatal
periods and to provide effective and timely care [32]. To improve the life expectancy of patients
with CHD, modern intensive care units are needed as well as health personnel trained to man-
age them. Finally, the implementation of a good hospital referral system is important.

Factors associated with mortality

Factors associated with mortality from CHD were identified which indicate the social determi-
nants of health that need to be addressed to improve infant survival. While 6.8% of births
occurred at home or at other non-institutional sites, roughly 25% of the children under 1 year
of age who died from a CHD in 2013 were born in this setting (data not shown), and the odds
of dying from CHD were lower for births at private hospitals than for those at home and in
public institutions. This fact leads us to emphasize the importance of implementing the WHO
recommendation [33] to examine babies born at home as soon as possible during the first 24
hours of life, and to have three more check-ups thereafter. It is also important to promote insti-
tutionalized births and to understand the gap between outcomes at public and private institu-
tions. Better outcomes at private institutions could be explained by better socioeconomic
conditions and a better health status of the population using private centers, or these hospitals
may use better practices to diagnose and treat CHD. In addition, living in a rural area was asso-
ciated with 2.35 times more likelihood of dying from CHD compared to children living in
urban areas. Meanwhile, a different result was observed in China, where a higher probability of
dying (OR 1.20; 95%CI: 1.04–1.05) was associated with urban areas for children under 5 years
of age [8]. These differences demonstrate the importance of conducting regional analyses. And
finally, being a male—a non-modifiable biological risk factor—was associated with higher mor-
tality, as has been previously reported [7].
National strategies need to focus on these vulnerable populations in the country and provide
more investment and infrastructure for public hospitals, as well as eliminate the social barriers
to accessing health services. Studies should also be conducted to determine whether rural areas
present more exposure to factors that can increase the risk of CHD.
The strengths of this study include its population-based design and covering a long period
of time, as well as the use of the ICD-10 on death certificates in Mexico which enabled making
international comparisons. Another strength is the functional sub-classification presented. The
inclusion of the age at death from CHD and the geographic analysis of these deaths are also
very important since they contribute to establishing public health strategies to improve the
prognosis of patients with CHD.
The weaknesses of the study include a lack of connection between birth certificates and
death certificates in Mexico, not knowing the gestational age of CHD-specific deaths and not
having identified deaths from CHD associated with chromosomal abnormalities. Another
weakness is the high proportion of unspecified diagnoses, particularly on the first day of life,
and the lack of fetal data which certainly underestimates the magnitude of this problem.

Conclusions
In Mexico, CHD presents a serious public health problem and has been increasing over recent
years, as is occurring in several developing countries. Research is needed to understand this
finding. The persistence of ductus arteriosus and other CHD with left-to-right shunt are the

PLOS ONE | DOI:10.1371/journal.pone.0150422 March 3, 2016 13 / 16

 Mortality from Congenital Heart Disease in Mexico

main causes of death among children under 1 year of age who have CHD. The neonatal period
is the stage when most infant deaths from CHD occur.
Improving the early detection and epidemiological monitoring of CHD is also needed in
Mexico. And linking the health information system to data from death and birth certificates is
indispensable.
Finally, it is important for Mexico’s health system to promote comprehensive postnatal
medical visits as recommended by the WHO.

Acknowledgments
The authors are grateful to Gerardo Sierra-Murguía for his help creating figures and data anal-
ysis and Gabriela Gómez Rodríguez for creating Fig 2.

Author Contributions
Conceived and designed the experiments: JLTC TMP. Performed the experiments: JLTC TMP.
Analyzed the data: JLTC TMP. Contributed reagents/materials/analysis tools: JLTC TMP.
Wrote the paper: JLTC CRP MAR PMAG ERM TMP.

References
1. You D, Hug L, Ejdemyr S, Idele P, Hogan D, Mathers C, et al. Global, regional, and national levels and
trends in under-5 mortality between 1990 and 2015, with scenario-based projections to 2030: a system-
atic analysis by the UN Inter-agency Group for Child Mortality Estimation. Lancet. 2015. Epub 2015/09/
13. doi: 10.1016/s0140-6736(15)00120-8 PMID: 26361942.
2. World Health Organization: World Health Statistics 2015. 2015 ed. Luxembourg: WHO Press; 2015.
Available: http://www.who.int/gho/publications/world_health_statistics/2015/en/. Accessed: 18 Sep-
tember 2015.
3. van der Linde D, Konings EE, Slager MA, Witsenburg M, Helbing WA, Takkenberg JJ, et al. Birth preva-
lence of congenital heart disease worldwide: a systematic review and meta-analysis. J Am Coll Cardiol.
2011; 58(21):2241–7. Epub 2011/11/15. doi: 10.1016/j.jacc.2011.08.025 PMID: 22078432.
4. Ionescu-Ittu R, Marelli AJ, Mackie AS, Pilote L. Prevalence of severe congenital heart disease after
folic acid fortification of grain products: time trend analysis in Quebec, Canada. BMJ. 2009; 338:b1673.
Epub 2009/05/14. doi: 10.1136/bmj.b1673 PMID: 19436079; PubMed Central PMCID:
PMCPmc2682153.
5. Khoshnood B, De Vigan C, Vodovar V, Goujard J, Lhomme A, Bonnet D, et al. Trends in prenatal diag-
nosis, pregnancy termination, and perinatal mortality of newborns with congenital heart disease in
France, 1983–2000: a population-based evaluation. Pediatrics. 2005; 115(1):95–101. Epub 2005/01/
05. doi: 10.1542/peds.2004-0516 PMID: 15629987.
6. Khoshnood B, Loane M, Garne E, Addor MC, Arriola L, Bakker M, et al. Recent decrease in the preva-
lence of congenital heart defects in Europe. J Pediatr. 2013; 162(1):108–13.e2. Epub 2012/07/28. doi:
10.1016/j.jpeds.2012.06.035 PMID: 22835879.
7. Lee K, Khoshnood B, Chen L, Wall SN, Cromie WJ, Mittendorf RL. Infant mortality from congenital mal-
formations in the United States, 1970–1997. Obstet Gynecol. 2001; 98(4):620–7. Epub 2001/09/29.
PMID: 11576578.
8. Hu Z, Yuan X, Rao K, Zheng Z, Hu S. National trend in congenital heart disease mortality in China dur-
ing 2003 to 2010: a population-based study. J Thorac Cardiovasc Surg. 2014; 148(2):596–602.e1.
Epub 2013/11/26. doi: 10.1016/j.jtcvs.2013.08.067 PMID: 24268955.
9. Instituto Nacional de Geografía e Informática (INEGI). Página oficial con información geográfica y
demográfica de México. Instituto Nacional de Geografía e Informática.Bases de datos de: Economía
[Internet]. Available: http://www3.inegi.org.mx/sistemas/mexicocifras. Accessed: 10 September 2015.
10. Secretaría de Salud. Dirección General de Informacion en Salud (DGIS). Sistema Nacional de Informa-
ción en Salud. Bases de datos: Defunciones [Internet]. Available: http://www.dgis.salud.gob.mm/
contenidos/basesdedatos/estandar.html. Accessed: 1 September 2015.
11. Organización Panamericana de la Salud-Organización Mundial de la Salud. Clasificación Estadística
Internacional de Enfermedades y Problemas relacionados con la Salud. 10° Revisión. 2003 ed. Wash-
ington (DC) US: WHO Press; 1995.

PLOS ONE | DOI:10.1371/journal.pone.0150422 March 3, 2016 14 / 16

 Mortality from Congenital Heart Disease in Mexico

12. Yeh SJ, Chen HC, Lu CW, Wang JK, Huang LM, Huang SC, et al. Prevalence, mortality, and the dis-
ease burden of pediatric congenital heart disease in Taiwan. Pediatr Neonatol. 2013; 54(2):113–8.
Epub 2013/04/18. doi: 10.1016/j.pedneo.2012.11.010 PMID: 23590956.
13. Instituto Nacional de Geografía e Informática (INEGI). Página oficial con información geográfica y
demográfica de México. Consulta interactiva de datos. Base de datos de: Nacimientos [Internet]. Avail-
able: http://www.inegi.org.mx/est/lista_cubos/consulta.aspx?p=adm&c=5. Accessed: 1 September
2015.
14. Secretaría de Salud. Dirección General en información en Salud (DGIS). Base de datos: Recursos
Humanos [Internet]. Available: http://www.dgis.salud.gob.mx/contenidos/sinais/estadisticas.html.
Accessed: 3 March 2014.
15. Reller MD, Strickland MJ, Riehle-Colarusso T, Mahle WT, Correa A. Prevalence of congenital heart
defects in metropolitan Atlanta, 1998–2005. J Pediatr. 2008; 153(6):807–13. Epub 2008/07/29. doi: 10.
1016/j.jpeds.2008.05.059 PMID: 18657826; PubMed Central PMCID: PMCPmc2613036.
16. Gutiérrez JP R-DJ, Shamah-Levy T, Villalpando-Hernández S, Franco A, Cuevas-Nasu L, Romero-
Martínez M, et al. Encuesta Nacional de Salud y Nutrición 2012. Resultados Nacionales. 1st ed. Cuer-
navaca, Morelos, México: Instituto Nacional de Salud Pública; 2012.
17. SPSS Inc., and IBM Compay. Manual del usuario del sistema básico de IBM SPSS Statistics, version
19. New York, USA; 2010.
18. Dean AG SK, Soe MM. OpenEpi: Open Source Epidemiologic Statistics for Public Health. 2015
[updated 2015/05/042015/10/03]. Versión 3.03a: Available: www.openepi.com/Menu/OE_Menu.htm.
Accessed: 3 September 2015.
19. Gilboa SM, Salemi JL, Nembhard WN, Fixler DE, Correa A. Mortality resulting from congenital heart
disease among children and adults in the United States, 1999 to 2006. Circulation. 2010; 122
(22):2254–63. Epub 2010/11/26. doi: 10.1161/circulationaha.110.947002 PMID: 21098447.
20. Valdes-Hernandez J, Canun-Serrano S, Reyes-Pablo AE, Navarrete-Hernandez E. [Birth defects mor-
tality in five-year-old minors of age, Mexico, 1998–2006]. Salud Pública Mex. 2009; 51(5):381–9. Epub
2009/11/26. PMID: 19936551.
21. Gomez-Alcala AV, Rascon-Pacheco RA. [Infant mortality from congenital malformations in Mexico: an
issue of opportunity and access to treatment]. Rev Panam Salud Pública. 2008; 24(5):297–303. Epub
2009/01/15. PMID: 19141171.
22. Hernandez B, Ramirez-Villalobos D, Duarte MB, Corcho A, Villarreal G, Jimenez A, et al. [Underreport-
ing of deaths in children and birth certification in a representative sample of the 101 municipalities with
lowest human development index in Mexico]. Salud Publica Mex. 2012; 54(4):393–400. Epub 2012/07/
27. PMID: 22832831.
23. Instituto Nacional de Psiquiatiría Ramón de la Fuente Muñiz; Instituto Nacional de Salud Pública;
Secretaría de Salud. Encuesta Nacional de Adicciones 2011: Reporte de Drogas. Villatoro-Velázquez
JA, Medina-Mora ME, Fleiz-Bautista C et al. México D.F. México: INPRFM; 2012. Available: www.inprf.
gob.mx. Accessed: 3 September 2015.
24. Simeone RM, Devine OJ, Marcinkevage JA, Gilboa SM, Razzaghi H, Bardenheier BH, et al. Diabetes
and congenital heart defects: a systematic review, meta-analysis, and modeling project. Am J Prev
Med. 2015; 48(2):195–204. Epub 2014/10/19. doi: 10.1016/j.amepre.2014.09.002 PMID: 25326416;
PubMed Central PMCID: PMCPmc4455032.
25. Aryana A, Williams MA. Marijuana as a trigger of cardiovascular events: speculation or scientific cer-
tainty? Int J Cardiol. 2007; 118(2):141–4. Epub 2006/09/29. doi: 10.1016/j.ijcard.2006.08.001 PMID:
17005273.
26. Oster ME, Riehle-Colarusso T, Simeone RM, Gurvitz M, Kaltman JR, McConnell M, et al. Public health
science agenda for congenital heart defects: report from a Centers for Disease Control and Prevention
experts meeting. J Am Heart Assoc. 2013; 2(5):e000256. Epub 2013/08/30. doi: 10.1161/jaha.113.
000256 PMID: 23985376; PubMed Central PMCID: PMCPmc3835228.
27. Hammerman C, Bin-Nun A, Kaplan M. Managing the patent ductus arteriosus in the premature neo-
nate: a new look at what we thought we knew. Seminars in perinatology. 2012; 36(2):130–8. Epub
2012/03/15. doi: 10.1053/j.semperi.2011.09.023 PMID: 22414884.
28. Cervantes-Salazar J, Calderon-Colmenero J, Ramirez-Marroquin S, Palacios-Macedo A, Bolio-Cerdan
A, Vizcaino Alarcon A, et al. [Pediatric cardiovascular surgical data base registry in Mexico. First report].
Revista de investigacion clinica; organo del Hospital de Enfermedades de la Nutricion. 2013; 65
(6):476–82. Epub 2014/04/02. PMID: 24687354.
29. Good RJ, Canale SK, Goodman RL, Yeager SB. Identification of Critical Congenital Heart Disease in
Vermont: The Role of Universal Pulse Oximetry Screening in a Rural State. Clinical pediatrics. 2015; 54
(6):570–4. Epub 2014/11/16. doi: 10.1177/0009922814558247 PMID: 25398625.

PLOS ONE | DOI:10.1371/journal.pone.0150422 March 3, 2016 15 / 16

 Mortality from Congenital Heart Disease in Mexico

30. Holland BJ, Myers JA, Woods CR Jr. Prenatal diagnosis of critical congenital heart disease reduces
risk of death from cardiovascular compromise prior to planned neonatal cardiac surgery: a meta-analy-
sis. Ultrasound in obstetrics & gynecology: the official journal of the International Society of Ultrasound
in Obstetrics and Gynecology. 2015; 45(6):631–8. Epub 2015/04/24. doi: 10.1002/uog.14882 PMID:
25904437.
31. Eckersley L, Sadler L, Parry E, Finucane K, Gentles TL. Timing of diagnosis affects mortality in critical
congenital heart disease. Archives of disease in childhood. 2015. Epub 2015/07/02. doi: 10.1136/
archdischild-2014-307691 PMID: 26130379.
32. Amsbaugh S, Scott SD, Foss K. Pulse Oximetry Screening for Critical Congenital Heart Disease: Bring-
ing Evidence Into Practice. Journal of pediatric nursing. 2015; 30(4):591–7. Epub 2014/12/03. doi: 10.
1016/j.pedn.2014.10.013 PMID: 25458107.
33. World Health Organization. WHO Recommendations on Posnatal care of the mother and newborn,
2013. Geneva, Switzerland: WHO Press; 2014. Available: http://apps.who.int/iris/bitstream/10665/
97603/1/9789241506649_eng.pdf. Accessed: 3 Jun 2015.

PLOS ONE | DOI:10.1371/journal.pone.0150422 March 3, 2016 16 / 16