International Journal of Cardiology 176 (2014) 956961

Contents lists available at ScienceDirect

International Journal of Cardiology

journal homepage: www.elsevier.com/locate/ijcard

Air pollution associated hypertension and increased blood pressure may

be reduced by breastfeeding in Chinese children: The Seven
Northeastern Cities Chinese Children's Study
Guang-Hui Dong a,, Zhengmin (Min) Qian b, Edwin Trevathan b, Xiao-Wen Zeng a, Michael G. Vaughn c,
Jing Wang d, Yang Zhao e, Yu-Qin Liu e, Wan-Hui Ren f, Xiao-Di Qin a
a

Department of Preventive Medicine, School of Public Health, Sun Yat-Sen University, Guangzhou, Guangdong Province 510080, PR China
Department of Epidemiology, College for Public Health and Social Justice, Saint Louis University, Saint Louis, MO 63104, USA
c
School of Social Work, College for Public Health and Social Justice, Saint Louis University, Saint Louis, MO 63104, USA
d
Department of Biostatistics, College for Public Health and Social Justice, Saint Louis University, Saint Louis, MO 63104, USA
e
Department of Biostatistics and Epidemiology, School of Public Health, China Medical University, Shenyang, China
f
Department of Ambient Air Pollution Monitor, Shenyang Environmental Monitoring Center, Shenyang, China
b

a r t i c l e

i n f o

Article history:
Received 20 June 2014
Accepted 17 August 2014
Available online 26 August 2014
Keywords:
Breastfeeding
Air pollution
Hypertension
Blood pressure
Interaction effect

a b s t r a c t
Background: Little is known about the association between air pollution and hypertension among children, and
no studies report whether breastfeeding modies this association in children.
Methods: Nine thousand three hundred fty-four Chinese children, ages 517 years old, from 24 elementary schools
and 24 middle schools in the Seven Northeastern Cities during 20122013 were evaluated. The weight, height, and
BP were measured. Four-year average concentrations of particles with an aerodynamic diameter of 10 m (PM10),
sulfur dioxide (SO2), nitrogen dioxides (NO2), ozone (O3), and carbon monoxide (CO) were calculated from monitoring stations. Two-level regression analysis was used to examine the effects, controlling for covariates.
Results: The results showed that associations existed between hypertension and pollutants. The odds ratios for hypertension ranged from 1.12 per 46.3 g/m3 increase for O3 (95% condence interval [CI], 1.101.13) to 1.68 per
30.6 g/m3 increase for PM10 (95% CI, 1.531.86). The increases in mean diastolic BP ranged from 0.58 mm Hg
per 46.3 g/m3 increase for O3 (95% CI, 0.520.63 mm Hg) to 2.89 mm Hg per 563.4 g/m3 increase for CO
(95% CI: 2.533.24 mm Hg). The increase in systolic BP ranged from 0.50 mm Hg per 46.3 g/m3 increase for O3
(95% CI: 0.430.57 mm Hg) to 2.10 mm Hg per 30.6 g/m3 increase for PM10 (95% CI, 1.732.47 mm Hg). Compared
with children who had been breastfed, non-breastfed children exhibited consistently stronger effects.
Conclusion: Study ndings indicate that high levels of PM10, SO2, NO2, O3, and CO are associated with increased arterial BP and hypertension among the children. Breastfeeding may reduce the risk.
2014 Elsevier Ireland Ltd. All rights reserved.

1. Introduction
Almost one billion people worldwide have hypertension, a major
risk factor for cardiovascular disease. Blood pressure (BP), on average,
has slightly decreased worldwide between 1980 and 2008. However,
many populations in developing countries have experienced an
Abbreviations: PM10, Particulate matter with an aerodynamic diameter 10 m; SO2,
Sulfur dioxide; NO2, Nitrogen dioxide; O3, Ozone; ORs, Odds ratios; 95% CI, 95%
Condence interval; BP, Blood pressure; SNECCS, Seven Northeastern Cities Chinese
Children's Study.
Competing nancial interests declaration: The authors report no relationships that
could be construed as a conict of interest.
Corresponding author at: Department of Preventive Medicine, School of Public Health,
Sun Yat-sen University, 74 Zhongshan 2nd Road, Yuexiu District, Guangzhou 510080, PR
China. Tel.: +86 20 87333409; fax: +86 20 87330446.
E-mail addresses: donggh5@mail.sysu.edu.cn, donggh512@hotmail.com (G.-H. Dong).

http://dx.doi.org/10.1016/j.ijcard.2014.08.099
0167-5273/ 2014 Elsevier Ireland Ltd. All rights reserved.

increase in BP, and an increased prevalence of hypertension [1]. Kearney

et al. estimate that by 2025 approximately 1.17 billion people in developing countries will have hypertension, nearly three-fourths of the
worldwide population with hypertension [2]. While genetic factors,
changing lifestyles, and dietary shifts are contributing factors, environmental factors such as ambient air pollution may increase the risk for
hypertension [36].
Evidence shows that ambient air pollution [46] and breastfeeding
[711] are associated with hypertension among child'ren. Breast milk
contains long-chain polyunsaturated fatty acids (LCPUFAs), which are
important structural components of tissue membrane systems, including
the vascular endothelium, which is responsible for the reduction of vasoconstriction and platelet aggregation [12]. In particular, breastfeeding
can inuence the programming of neonatal sodium restriction, degrade
insulin resistance, and promote lipid metabolism that persists into
adolescence and adulthood [1315]. Previous studies suggest that

G.-H. Dong et al. / International Journal of Cardiology 176 (2014) 956961

breastfeeding is associated with decreased effects of other environmental toxicants on children's health [16]. Our own study also reported that
breastfeeding is a modier of the respiratory effects of air pollution in
Chinese children [17]. To test the hypothesis that breastfeeding decreases the adverse BP response to ambient air pollution exposure
among children may provide insight into the effects of co-exposure to
ambient air pollution and breastfeeding on hypertension and blood pressure in children, which in turn may provide the benet of predicting and
controlling BP later in life.
Using data from the Seven Northeastern Cities Chinese Children's
Study, we investigated two hypotheses: (1) populations with longterm exposure to ambient air pollution experience higher BP and a
higher prevalence of hypertension among children, and (2) breastfeeding
decreases the adverse BP response to ambient air pollution exposure
among children. We tested these two hypotheses by analyzing data
from the Seven Northeastern Cities Chinese Children's Study. This
cross-sectional study evaluated air pollution and child health in a large,
well-characterized population-based sample of children residing in
northeast China, where there are wide differences in inter- and intracity pollution ranges and high ambient pollutant levels, thus offering an
opportunity to assess the associations between exposure and response.

957

cigarettes daily in the home. Breastfeeding exposure, based on parental report, was
dened as mainly breastfed for 3 months or more. In present study, new mothers had a
3-month pregnancy leave after delivery. After 3 months, most of the new mothers had
to return to work and many of them stopped breastfeeding at that time. Therefore, we designed the question to collect only 3-month breastfeeding information. The early diet of
children not breastfed includes animal milk, juice, and soups made from egg, rice, chicken,
pork, beef, sh, or vegetables [17].
2.4. Ambient air pollution

2. Methods

In each of the selected study districts, there was one municipal air monitoring station
located 1 km from the study participants' homes. Measurements of PM10, SO2, NO2, and O3
concentrations from 2009 to 2012 were obtained at the stations, using uniform methods
and quality assurance. These stations were separated by specied distances from major
roads, industrial sources, buildings, or residential sources of emissions from the combustion of coal, waste, or oil, assuring that the air pollution measurements were more likely
to reect the background air pollution levels. These measurements were used to estimate
long-term exposures for the participants. Measurements strictly followed the methodological standards set by the State Environmental Protection Administration of China [19].
Concentrations of each pollutant were assessed continuously and reported hourly:
PM10 by beta-attenuation, SO2 by ultraviolet uorescence, NO2 by chemiluminescence,
O3 by ultraviolet photometry, and CO by non-dispersive infrared spectrometry [20]. We
calculated average daily concentrations of PM10, SO2, NO2, O3, and CO (averaged over
eight hours). These data were based on days for which at least 75% of the 1-h values
were available, after excluding abnormal values in the hourly data collected from each
monitoring station. Exposure parameters in the present study were 4-year averages
(20092012) for concentrations calculated from the 24-h PM10, SO2, NO2, and CO concentrations, and 10:00 AM to 6:00 PM 8-h O3 concentrations in each district.

2.1. Study cities selection and subject recruitment

2.5. Statistical analysis

The Seven Northeastern Cities Chinese Children's Study (SNECCS) is a cross-sectional

study of children's health outcomes based on exposure to ambient air pollutants. This region encompasses more than 20 million people residing in 14 cities in Liaoning province in
Northeastern China. To maximize the inter- and intra-city gradients of the pollutants of interest and also to minimize the correlation between district-specic ambient pollutants, in
April 2012 the seven cities of Shenyang, Dalian, Anshan, Fushun, Benxi, Liaoyang, and
Dandong in Liaoning province were selected as study sites, based on air pollution measurements taken between 2009 and 2011. In each of the seven cities, we selected all
urban districts for the study. There are ve districts in Shenyang, four districts in Dalian
and Fushun, three districts in Anshan, Benxi, and Dandong, and two districts in Liaoyang.
In each of the 24 study districts, one elementary school and one middle school within one
mile of air monitoring sites were randomly selected, and then one or two classrooms were
randomly selected from each grade of the selected schools. In any classroom targeted for
participation, all children who had lived in the district for at least 2 years before the
start of study were included in the study. The design and conduct of this investigation
was in accordance with the World Medical Association Declaration of Helsinki-Ethical
Principles for Medical Research Involving Human Subjects and ethical approval was obtained from Human Studies Committee of China Medical University. Before data collection,
a written informed consent form was obtained from each participant and their parents.

Data were tested for homogeneity (using Bartlett's unequal variances test) and normality (using ShapiroWilks W-test). For each group, the values of mean standard deviation were calculated for continuous variables. Relative frequencies were calculated for
categorical variables. Contingency tables and 2-tests were used to calculate the relationship associations between categorical variables. We assessed the association of ambient air
pollutants with BP using generalized additive models. To investigate the relationship between hypertension and ambient air pollution, we considered a two-level binary logistic
regression model with participants being the rst-level units and districts being the
second-level units. At the child level, we predicted the logit of the hypertension's prevalence rate by breastfeeding (BF) and k covariates (X1, , Xk) as follows:

2.2. BP measurements
All investigators and staff of the study were required to successfully complete a training program based on procedures formulated by the American Academy of Pediatrics [18],
designed to facilitate the administration of the study questionnaire and a standardized
protocol to measure BP. At the end of the training program, each trainee was required to
take a qualifying examination and those who passed were given a BP observer certicate.
Participants were advised not to smoke, drink alcohol, coffee, or tea, and to abstain from
exercise for at least 30 min before having their BP measured.
Measurements were performed in a quiet and temperate room. The BP measurements
were performed by a team of four carefully trained nurses whose adherence to the measurement protocol was assessed regularly. After 5 min of rest, sitting BP was measured
three times by using standardized mercuric-column sphygmomanometer, with the use
of an appropriate cuff size adapted to arm circumference according to standardized procedural guidelines. The time interval between successive pairs of BP measurements was 2
min. The average of the three BP measurements was used to identify hypertension. The
hypertension status of the participants was assessed based on the fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents [18].
Hypertension in children was dened as average SBP and DBP that is 95th percentile for
gender, age and height. This denition of the BP threshold has gained worldwide support
and has been used in the vast majority of studies on pediatric hypertension [18].
2.3. Anthropometric measurements and data collection
Children were measured without shoes in light garments. Weight and height were
measured with the use of electronic scales (at 0.1 kg) and xed stadiometers (at
0.1 cm). The body mass index (BMI) was calculated as weight divided by height squared
(kg/m2). Parental education was dened as the highest education level completed by either parent. Passive smoking exposure was dened as living with someone who smokes

logit Psymptomij  j jBFij 1X1ij : kX1ij eij

where the subscript j is for districts (j = 1,, 24), the subscript i is for children (i = 1, nj),
j is the intercept at the district level, j is the regression coefcient for breastfeeding,
1, , k are the regression coefcients of covariates, and eij is the random error, assumed
to have means of zero and constant variance. The j and j are random coefcients as they
are assumed to vary across districts. In general, a district with a high j is predicted to have
higher prevalence rates than a district with a low j. Similarly, differences in j indicate that
the relationship between breastfeeding and prevalence rates is not the same in all districts.
In districts with a high (low) j, breastfeeding has a large (small) effect on prevalence rates
or the difference between breastfed children and non-breastfed children is relatively large
(small).
At the district level, we regressed the district-specic intercepts j and coefcients j
on the district-specic pollutant level (Zj) to explain the variations of j and j, as follows:

j 1 Zj u1j

j 2 Zj u2 j:

Eq. (2) predicts the prevalence rates in a district by Zj. If 1 is positive, then, adjusting
for covariates, the districts with higher pollutant levels have a higher prevalence rate. Conversely, if 1 is negative, then, adjusting for covariates, the prevalence rates are lower in
districts with a higher pollutant level. Eq. (3) states that, adjusting for covariates, the relationship between prevalence rates and breastfeeding in a district depends on the district's
pollutant level Zj. If 2 is positive, then, adjusting for covariates, the breastfeeding effect on
prevalence rates is larger with a higher pollutant level. Conversely, if 2 is negative, then
adjusting for covariates, the breastfeeding effect on prevalence rates is smaller with a
higher pollutant level. The u-terms u1j and u2j are random errors at the district level, assumed to be independent and have mean of zero and constant variance. These random errors characterize the variation between districts and are assumed to be independent from
eij at the child level. Note that , , 1 ,, k, 1 and 2 are not assumed to vary across
districts. Therefore, they have no subscript j to indicate to which district they belong;
they are referred to as xed effects as they apply to all districts.

958

G.-H. Dong et al. / International Journal of Cardiology 176 (2014) 956961

The above models can be written as a single regression equation by substituting

Eqs. (2) and (3) into Eq. (1):

logitPsymptomj 1 Zj BFij 1X1ij kXkij 2Zj BFij 4

u2 j BFij u1 j eij:
The terms in the rst and second parentheses in Eq. (4) are often respectively called
the xed (or deterministic) and random (or stochastic) parts of the model. The product
term ZjBFij is a cross-level interaction between the child-level variable BFij and the
district-level variable Zj. The random error u2j BFij is different for different children, a situation that in ordinary multiple regression analysis is called heteroscedasticity.
We conducted all analyses using the GLIMMIX procedure in SAS 9.2. Statistical significance was assessed using at a 5% level of signicance using a two-sided test.

3. Results
There were a total of 10,428 children randomly selected from 24 districts of the seven cities, of whom 9567 completed the survey and examination resulting in an overall response rate of 91.7%. We excluded 213
children from further analyses (2.2%) because they had resided in their
current district for less than 2 years. Among the 9354 children analyzed,
the average age was 10.9 years (SD = 2.5 years; range from 5 to 17),
and 4771 (51.0%) were males. For all participants, the prevalence rate
of hypertension was 13.78%. The characteristics of the survey participants in this study, stratied by gender, are shown in Table 1.
Table 2 summarizes the distribution of the annual mean air pollutant
concentration in the 24 monitoring stations from 2009 to 2012. There
are wide variations between district gradients for PM10 (50.0
132.5 g/m3), SO2 (14.879.0 g/m3), NO2 (10.553.3 g/m3), O3
(15.0574.0 g/m3) and CO (572.33095.7 g/m3). Comparing them to
the WHO guidelines and Chinese National Ambient Air Quality Standards (NAAQS), the table shows that PM10 levels exceed WHO guidelines in all districts, and 95.8% of all districts exceed WHO guidelines
for SO2.
For all participants, using the GLIMMIX model, we observed signicant associations between prevalence rate of hypertension and PM10
(OR = 1.68; 95% CI: 1.531.86), SO2 (OR = 1.33; 95% CI: 1.211.47),
NO2 (OR = 1.33; 95% CI: 1.221.44), O3 (OR = 1.12; 95% CI: 1.10
1.13), and CO (OR = 1.64; 95% CI: 1.481.82). When stratied by gender, these relationships were obtained in both males and females

(Table 3). No signicant interactions were identied between air pollutants and gender on occurrence of hypertension.
We also found signicant associations between long-term ambient
air pollutant exposure with SBP and DBP (Table 4). For example, in
the single pollutant model, DBP and SBP increased by 1.93 mm Hg
(95% CI, 1.622.25 mm Hg) and 2.10 mm Hg (95% CI, 1.732.47 mm
Hg), respectively, for every 30.6 g/m3 increase in the 4-year mean
PM10. When stratied by gender we found that these relationships
were still signicant.
We further analyzed the interaction between air pollution and
breastfeeding on the hypertension and arterial BP among children
(Table 5). Among non-breastfed children, the ORs for hypertension
were consistently higher than those of breastfed children, with statistically signicant interactions between breastfeeding and PM10 (P = 0.016),
O3 (P b 0.001), CO (P = 0.055), and NO2 (P = 0.080). No signicant interactions were observed between breastfeeding and pollution on blood
pressure (Table 5).
4. Discussion
We found an association between long-term ambient air pollution
and elevated BP among 9354 children aged 517 years. The ndings
supported our hypothesis that long-term ambient air pollution is adversely associated with hypertension, as well as arterial SBP and DBP
in children. Furthermore, compared with breastfed children, nonbreastfed children exhibited stronger effects of air pollution on both hypertension and arterial BP.
To our knowledge, this study is the rst to report the effect of
breastfeeding on the association between air pollution and hypertension in children. The signicant modications of breastfeeding on the
associations identied from this study show coherence with the results
from our previous studies [17]. This study has important features: 1) the
study involved 24 districts in Northeast China; 2) there was a wide
cross-city gradient of long-term air pollution levels, and the pollution
levels differed substantially between the district; 3) the trend in pollution levels across the study district was relatively stable during the
study period; 4) residential populations were stable and response rate
was high gradient; and 5) centrally and rigorously trained interviewers
in this study used objectively uniform criteria and closed-ended

Table 1
Characteristics of the study population in the 24 districts of seven cities, China.
Characteristics

Male (n = 4771)

Female (n = 4583)

Total (n = 9354)

Age (years); mean (SD)

Height (cm); mean (SD)
Weight (kg); mean (SD)
Body mass index (kg/m2); mean (SD)
Low birth weight (%)

10.9 (2.62)
151.07 (16.58)
47.57 (18.15)
20.15 (4.64)
122 (2.56)
249 (5.22)
3193 (66.93)
2925 (61.31)

10.9 (2.57)
148.83 (14.15)
42.79 (14.09)
18.92 (3.85)
157 (3.43)
231 (5.04)
3279 (71.55)
2834 (61.84)

10.9 (2.59)
149.77 (15.49)
45.22 (16.46)
19.55 (4.32)
279 (2.98)
480 (5.13)
6472 (69.19)
5759 (61.57)

524 (10.98)
608 (12.74)
1653 (34.65)
1764 (36.97)
222 (4.65)

508 (11.08)
603 (13.16)
1577 (34.41)
1677 (36.59)
218 (4.76)

1032 (11.03)
1211 (12.95)
3230 (34.53)
3441 (36.79)
440 (4.70)

1441 (30.20)
414 (8.68)
2269 (47.56)
462 (9.68)
22.98 (10.79)
7.96 (7.86)
4344 (91.05)
1699 (35.61)
659 (13.81)

1446 (31.55)
388 (8.47)
2217 (48.37)
426 (9.30)
22.30 (9.56)
7.27 (7.62)
4227 (92.23)
1774 (38.71)
630 (13.75)

2887 (30.86)
802 (8.57)
4486 (47.96)
888 (9.49)
22.65 (10.21)
7.62 (7.75)
8571 (91.63)
3473 (37.13)
1289 (13.78)

Premature birth (%)

Breastfeeding (%)
Parental education less than higher school
Income per year
5000 RMB (%)
500010 000 RMB (%)
10,00030,000 RMB (%)
30,000100,000 RMB (%)
N100,000 RMB (%)
Passive smoking exposure source
Father (%)
Mother (%)
Anyone (%)
Home coal use (%)
Area of residence per person (m2); (SD)
Exercise per week (hour); mean (SD)
Parents as responders (%)
Family history of hypertension (%)
Hypertension (%)

Values are n (%). Signicant difference exists between male and female by t or 2 test, p b 0.05.

RMB, Chinese Yuan.

G.-H. Dong et al. / International Journal of Cardiology 176 (2014) 956961

959

Table 2
Air pollution levels in the 24 districts of seven cities, China, from 2009 to 2012 (g/m3).
Air pollution levels

PM10

SO2

NO2

O3

CO

Cities
Shenyang (mean SD)
Dalian (mean SD)
Fushun (mean SD)
Anshan (mean SD)
Benxi (mean SD)
Liaoyang (mean SD)
Dandong (mean SD)
Mean (SD)
Median
Min
Max
Interquartile range
NS
% of NNS
WHO guideline
% of NWHO

108.55 (8.15)
78.69 (26.47)
81.25 (11.13)
86.92 (9.04)
96.75 (16.15)
86.08 (42.21)
70.88 (3.01)
88.90 (21.31)
90.38
50.00
132.50
30.61
70
79.17
20
100

67.10 (7.08)
44.69 (1.16)
50.06 (8.61)
47.00 (17.39)
55.17 (22.05)
37.00 (21.54)
31.00 (8.13)
49.75 (16.03)
48.37
14.75
79.00
23.40
60
29.17
20
95.83

36.35 (5.90)
43.31 (10.25)
36.50 (8.49)
23.92 (18.48)
42.83 (9.86)
40.75 (9.96)
25.50 (2.83)
36.44 (11.10)
35.00
10.50
53.25
13.00
40
37.50
40
37.50

71.98 (17.62)
45.25 (34.33)
32.31 (6.09)
48.83 (2.90)
350.60 (277.63)
210.67 (314.66)
32.75 (4.60)
106.92 (165.80)
43.75
15.00
574.00
46.33
160
12.50
100
12.50

1515.40 (476.57)
604.56 (33.92)
1249.38 (154.39)
1368.17 (513.04)
2432.72 (628.51)
1391.92 (56.42)
1401.00 (103.24)
1390.55 (597.31)
1289.50
572.25
3095.66
563.35
4000
0

CO, carbon monoxide; NO2, nitrogen dioxide; NS, national standard; O3, ozone; PM10, particle with aerodynamic diameter of 10 m; SO2, sulfur dioxide; and WHO, World Health
Organization.
Interquartile range (IQR): range from 25th to 75th percentile of district-specic concentrations.

China national ambient air quality standard in 2012 [19].

WHO 2005 air quality guidelines [21].

questions to collect data, and both the interviewers and parents or

guardians were blinded to the greatest extent possible from knowing
the study objective and the participants' exposure status. Therefore,
we expected this study would provide a reasonable assessment of
health effects of ambient air pollution exposure.
This study has limitations. First, our cross-sectional study design cannot establish a cause-and-effect relationship between long-term ambient
air pollution and BP and hypertension. Second, possible misclassication
of the behavioral risk factors (passive smoking and exercising) could
exist due to the constrictive nature of the survey questions (yes or no).
Third, we developed a variable of 4-year average concentrations to
serve as surrogates for the long-term exposure to PM10, SO2, NO2, O3,
and CO. However, these measurements are more likely to reect the
background air pollution levels in a city, an advantage of our study compared to prior investigations. Trafc-related air pollution exposure data
are not available in this study. Therefore, this approach may lead to signicant uncertainty as to long-term exposure, since the concentration
of air pollutants may vary more widely over a longer span of time. Addressing spatial variations of trafc-related air pollution effects would
be one of the future directions our research could take. Fourth, it would
be ideal to adjust for noise when air pollution effects on SBP or hypertension are assessed because exposures to PM2.5 and NOx together with exposure to noise (especially trafc noise) have also been incriminated as a
causal factor for incident hypertension. Unfortunately, no noise data are

Table 3
Adjusted OR and 95% CI for the prevalence of hypertension of long-term exposure to the
air pollutant (n = 9354).
Pollutant, g/m3

PM10
SO2
NO2
O3
CO

Males

Females

OR

95% CI

OR

95% CI

1.55
1.32
1.28
1.12
1.70

1.351.77
1.151.52
1.141.43
1.091.14
1.471.97

1.85
1.35
1.39
1.11
1.58

1.612.13
1.181.55
1.231.56
1.091.14
1.361.82

Interaction p-value

0.202
0.867
0.506
0.583
0.579

CI, condence interval; CO, carbon monoxide; NO2, nitrogen dioxide; O3, ozone; PM10, particle with aerodynamic diameter of 10 m; SO2, sulfur dioxide.
Adjusted for age, sex, BMI, parental education, low birth weight, premature birth,
breast, income, passive smoking exposure, home coal use, exercise time, area residence
per person, family history of hypertension, and district.

ORs were scaled to the interquartile range (IQR) for each pollutant (30.6 g/m3 for
PM10, 23.4 g/m3 for SO2, 13.0 g/m3 for NO2, 46.3 g/m3 for O3, and 563.4 g/m3for CO).

available to this study. Nevertheless, these factors are not likely to vary
signicantly across the seven studied cities, because these cities are all
densely populated cities in northeastern China, where city landscapes,
trafc density, trafc patterns, and time activity patterns are quite
similar.
Although the association between exposure to ambient air pollution
and elevated arterial BP in adults has been reported, there are little published data on the relationship between long-term exposure to ambient
pollutants and measured BP or hypertension in children. In a systematic
Medline search, only three previous studies have reported on the association between air pollution and BP in children, and no consistent results were reported. Sughis et al. reported that the BP of children from
the school in the highly polluted area was signicantly higher than
that of children attending the school in the less polluted area (115.9/
70.9 and 108.3/66.4 mm Hg, respectively) [6]. In the PIAMA birth cohort
study conducted in the Netherlands, Bilenko et al. indicated that an IQR
increase in NO2 (7.8 g/m3) was associated with estimated increase in
mean DBP of 0.83 mm Hg (95% CI, 0.061.61 mm Hg), but not with

Table 4
Estimated absolute increase in arterial blood pressure (mm Hg) with 95% CI per interquartile range of long-term exposure to the air pollutant (n = 9354).
Pollutant
(g/m3)

Males
Estimate

Females

Total

95% CI

Estimate

95% CI

Estimate

95% CI

Diastolic blood pressure

1.67
PM10
SO2
0.57
NO2
0.83
O3
0.57
CO
3.02

1.232.12
0.121.02
0.461.21
0.480.65
2.513.52

2.21
0.99
1.43
0.58
2.74

1.782.64
0.561.41
1.081.79
0.500.66
2.253.23

1.93
0.76
1.15
0.58
2.89

1.622.25
0.451.07
0.891.41
0.520.63
2.533.24

Systolic blood pressure

PM10
1.58
SO2
1.00
NO2
0.73
O3
0.54
CO
2.26

1.042.12
0.461.54
0.281.18
0.430.64
1.652.87

2.62
1.46
1.64
0.46
1.83

2.123.12
0.971.95
1.232.06
0.370.56
1.262.41

2.10
1.25
1.15
0.50
2.07

1.732.47
0.881.61
0.841.46
0.430.57
1.642.49

CI, condence interval; CO, carbon monoxide; NO2, nitrogen dioxide; O3, ozone; PM10, particle with aerodynamic diameter of 10 m; SO2, sulfur dioxide.
Adjusted for age, sex, BMI, parental education, low birth weight, premature birth,
breast, income, passive smoking exposure, home coal use, exercise time, area residence
per person, family history of hypertension, and district.

Estimate was scaled to the interquartile range for each pollutant (30.6 g/m3 for PM10,
23.4 g/m3 for SO2, 13.0 g/m3 for NO2, 46.3 g/m3 for O3, and 563.4 g/m3for CO).

960

G.-H. Dong et al. / International Journal of Cardiology 176 (2014) 956961

Table 5
Adjusted OR of hypertension and estimated absolute increase in arterial BP (mm Hg) with
95% CI for the interaction between air pollution and breastfeeding among children aged 5
17 years.
Breastfeeding only

OR/estimate (95% CI)

No breastfeeding

Interaction

OR/estimate (95% CI)

p-value

Hypertension
PM10
1.57 (1.401.76)
SO2
1.29 (1.151.44)
NO2
1.27 (1.151.40)
O3
1.10 (1.081.12)
CO
1.55 (1.371.74)

2.06 (1.702.48)
1.47 (1.221.76)
1.49 (1.281.74)
1.17 (1.131.20)
1.92 (1.592.32)

0.016
0.237
0.080
b0.001
0.055

Diastolic blood pressure

PM10
1.83 (1.462.19)
SO2
0.66 (0.291.03)
NO2
1.07 (0.771.38)
O3
0.54 (0.480.62)
CO
2.85 (2.433.26)

2.15 (1.562.74)
0.96 (0.391.54)
1.25 (0.771.73)
0.63 (0.520.75)
2.98 (2.313.64)

0.470
0.388
0.583
0.305
0.833

Systolic blood pressure

PM10
2.13 (1.702.57)
SO2
1.17 (0.741.60)
NO2
1.27 (0.901.64)
O3
0.50 (0.410.58)
CO
1.97 (1.472.47)

2.01 (1.302.72)
1.39 (0.702.07)
0.87 (0.301.44)
0.50 (0.370.64)
2.27 (1.473.06)

0.531
0.666
0.214
0.965
0.602

CI, condence interval; CO, carbon monoxide; NO2, nitrogen dioxide; O3, ozone; PM10, particle with aerodynamic diameter of 10 m; SO2, sulfur dioxide.
Adjusted for age, sex, BMI, parental education, low birth weight, premature birth,
breast, income, passive smoking exposure, home coal use, exercise time, area residence
per person, family history of hypertension, and district.

OR and estimate were scaled to the interquartile range (IQR) for each pollutant
(30.6 g/m3 for PM10, 23.4 g/m3 for SO2, 13.0 g/m3 for NO2, 46.3 g/m3 for O3, and
563.4 g/m3for CO).

SBP. However, no signicant associations of PM with BP were found [4].

Liu et al. conducted secondary analyses to investigate changes in BP associated with changes in long-term exposure to ambient air pollutant
by applying land-use regression models among 2368 children age
10 years from two cities of Germany (1454 from Munich and 914
from Wesel) [5]. The results of that study showed no signicant relationship between air pollution and BP. Compared with these investigations, our study looked at not only the air pollutant PM but also other air
pollutants, such as SO2, NO2, O3, and CO. Ultimately, we observed a signicant relationship between these pollutants and the prevalence of hypertension. The ndings in the present investigation could be explained
by the differences in spatial and temporal variability of air pollutants
existing between the different regions. For example, in recent years, it
has been reported that lignite from a local Shenyang coal eld had the
highest concentrations of nickel (75 g/g) and chromium (79 g/g) in
the world, and some carcinogenic substances in coal were released
into the air during the combustion of lignites in Shenyang [22].
The benecial effects of breastfeeding on the levels of blood pressure
have been reported in many studies [711]. Singhal et al. reported that
breastfeeding was associated with a 2.7-mm Hg reduction in systolic
and a 3.2-mm Hg reduction in diastolic BP in adolescence [8]. In a prospective cohort study, the systolic and diastolic BP among breastfed children were 1.2 mm Hg (95% CI: 0.51.9) and 0.9 mm Hg (0.31.4),
respectively, lower than children who were never breastfed [9]. Furthermore, a recent cohort study reported that a lower duration of
breastfeeding was associated with a higher mean blood pressure
among children [7].
The mechanisms underlying the modication of breastfeeding on associations between ambient air pollution and both blood pressure and
hypertension are not well-established. One possibility is that systemic
inammation may be caused by ambient air pollution and reduced by
breast milk. Several biological pathways may be involved in the elicitation of oxidative stress, systemic pro-inammatory responses, and the
activation of pulmonary reexes in turn leading to arterial remodeling
[2328].

Children may be more susceptible than adults to the same environmental factors. Between 2009 and 2010, we conducted another investigation to evaluate the association between residential long-term
exposure to air pollution and blood pressure/hypertension among
24,845 Chinese adults in three northeastern cities [29]. Our ndings
showed that the odds ratio for hypertension ranged from 1.11 per
20.0 g/m3 increase for SO2 (95% CI, 1.041.18) to 1.13 per 22.0 g/m3
increase for O3 (95% CI, 1.061.20). The ndings from the current
study suggest that the implementation of strict environmental protections, specically for ambient air pollution, may be more relevant to preserve cardiovascular health not only in adults, but also in the pediatric
population. Prevention of air pollution exposure in early childhood is
supported by several factors. First, the efcacy of reducing air pollutant
concentrations in order to reduce cardiovascular mortality has been
convincingly demonstrated [30]. Second, elevations in BP during childhood consistently continue into adulthood [31]. Furthermore, air pollution exposure in children may not only compromise their long-term
cardiovascular health, but it may compromise their respiratory health,
as well: air pollution has been shown to affect lung function and therefore may constitute an important cause of childhood pulmonary
obstructive [32].
5. Conclusions
In this present study, long-term exposure to air pollution may increase arterial BP and be associated with hypertension in children,
and the negative impact of air pollution on BP may be reduced by
breastfeeding. Reducing the risk of childhood hypertension is an
added reason to reduce environmental air pollution exposure, and to
promote breastfeeding. Additional research is needed to verify these
ndings in other populations, to elucidate the etiologic mechanisms of
how air pollution inuences blood pressure, and to further examine
the public health implications of our ndings.
Acknowledgments
The authors acknowledge the cooperation of the seven cities,
school principals, teachers, and students and their parents. This
work was supported by grants from China Environmental Protection
Foundation (CEPF2008-123-1-5). The authors report no competing
nancial interest.
References
[1] Danaei G, Finucane MM, Lin JK, et al. National, regional, and global trends in systolic
blood pressure since 1980: systematic analysis of health examination surveys and
epidemiological studies with 786 country-years and 5 4 million participants. Lancet 2011;377:56877.
[2] Kearney PM, Whelton M, Reynolds K, Muntner P, Whelton PK, He J. Global burden of
hypertension: analysis of worldwide data. Lancet 2005;365:21723.
[3] Brook RD, Weder AB, Rajagopalan S. Environmental hypertensionology the effects of
environmental factors on blood pressure in clinical practice and research. J Clin
Hypertens (Greenwich) 2011;13:83642.
[4] Bilenko N, Rossem LV, Brunekreef B, et al. Trafc-related air pollution and noise and
children's blood pressure: results from the PIAMA birth cohort study. Eur J Prev
Cardiol 2013. http://dx.doi.org/10.1177/2047487313505821.
[5] Liu C, Fuertes E, Tiesler CM, et al. The associations between trafc-related air pollution and noise with blood pressure in children: results from the GINIplus and
LISAplus studies. Int J Hyg Environ Health 2014;217:499505.
[6] Sughis M, Nawrot TS, Ihsan-ul-Haque S, Amjad A, Nemery B. Blood pressure and particulate air pollution in schoolchildren of Lahore, Pakistan. BMC Public Health 2012;
12:378.
[7] Amorim Rde J, Coelho AF, Lira PI, Lima Mde C. Is breastfeeding protective for blood
pressure in schoolchildren? A cohort study in northeast Brazil. Breastfeed Med
2014;9:14956.
[8] Singhal A, Cole TJ, Lucas A. Early nutrition in preterm infants and later blood
pressure: two cohorts after randomised trials. Lancet 2001;357:4139.
[9] Martin RM, Ness AR, Gunnell D, Emmett P, Davey Smith G, ALSPAC Study Team.
Does breast-feeding in infancy lower blood pressure in childhood? The Avon Longitudinal Study of Parents and Children (ALSPAC). Circulation 2004;109:125966.
[10] Hosaka M, Asayama K, Staessen JA, et al. Breastfeeding leads to lower blood pressure
in 7-year-old Japanese children: Tohoku Study of Child Development. Hypertens Res
2013;36:11722.

G.-H. Dong et al. / International Journal of Cardiology 176 (2014) 956961

[11] Lawlor DA, Riddoch CJ, Page AS, et al. Infant feeding and components of the metabolic syndrome: ndings from the European Youth Heart Study. Arch Dis Child 2005;
90:5828.
[12] Lesson CPM, Kattenhorn M, Morley R, Lucas A, Deaneld JE. Impact of low birth
weight and cardiovascular risk factors on endothelial function in early adult life. Circulation 2001;103:12648.
[13] Martin RM, Gunnell D, Smith GD. Breastfeeding in infancy and blood pressure in
later life: systematic review and meta-analysis. Am J Epidemiol 2005;161:1526.
[14] Leon DA, Ronalds G. Breast-feeding inuences on later lifecardiovascular disease.
Adv Exp Med Biol 2009;639:15366.
[15] Owen CG, Whincup PH, Gilg JA, Cook DG. Effect of breast-feeding in infancy on blood
pressure in later life: systematic review and meta-analysis. BMJ 2003;327:118995.
[16] Baker RJ, Hertz-Picciotto I, Dostl M, et al. Coal home heating and environmental tobacco smoke in relation to lower respiratory illness in Czech children, from birth to
3 years of age. Environ Health Perspect 2006;114:112632.
[17] Dong GH, Qian Z, Liu MM, et al. Breastfeeding as a Modier of the Respiratory Effects
of Air Pollution in Children. Epidemiology 2013;24:38794.
[18] National High Blood Pressure Education Program Working Group on High Blood
Pressure in Children and Adolescents. The fourth report on the diagnosis, evaluation,
and treatment of high blood pressure in children and adolescents. Pediatrics 2004;
114:55576.
[19] State Environmental Protection Administration of China (SEPA). Standardized Environmental Monitoring and Analysis Methods. Beijing, China: SEPA; 2012.
[20] Qian Z, He Q, Lin HM, Kong L, Zhou D, Liao D. Part 2. Association of Daily Mortality
with Ambient Air Pollution, and Effect Modication by Extremely High Temperature
in Wuhan, China. Public Health and Air Pollution in Asian (PAPA): Coordinated

[21]
[22]

[23]

[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]

961

Studies of Short-Term Exposure to Air Pollution Daily Mortality in Four Cities, HEI
Research Report Boston, MA: Health Effects Institute; 2010.
World Health Organization. WHO air quality guidelines global update 2005. Report
on a Working group meeting. Bonn, Germany; October 1820 2005.
Ren DY, Xu DW, Zhao FH. A preliminary study on the enrichment mechanism and
occurrence of hazardous trace elements in the Tertiary lignite from the Shenbei coaleld, China. Int J Coal Geol 2004;57:18796.
Brook RD, Urch B, Dvonch JT, et al. Insights into the mechanisms and mediators of
the effects of air pollution exposure on blood pressure and vascular function in
healthy humans. Hypertension 2009;54:65967.
Brook RD. Why physicians who treat hypertension should know more about air pollution. J Clin Hypertens (Greenwich) 2007;9:62935.
Buescher ES. Anti-inammatory characteristics of human milk: how, where, why.
Adv Exp Med Biol 2001;501:20722.
Garofalo R. Cytokines in human milk. J Pediatr 2010;156:S3640.
Hanson LA. Breastfeeding provides passive and likely long-lasting active immunity.
Ann Allergy Asthma Immunol 1998;81:52333.
Mills NL, Donaldson K, Hadoke PW, et al. Adverse cardiovascular effects of air pollution. Nat Clin Pract Cardiovasc Med 2009;6:3644.
Dong GH, Qian Z, Xaverius PK, et al. Association between long term air pollution and
increased blood pressure and hypertension in China. Hypertension 2013;61:57884.
Laden F, Schwartz J, Speizer FE, Dockery DW. Reduction in ne particulate air pollution and mortality. Am J Respir Crit Care Med 2006;173:66772.
Chen X, Wang Y. Tracking of blood pressure from childhood to adulthood: a systematic review and meta-regression analysis. Circulation 2008;117:317180.
Brunekreef B, Holgate S. Air pollution and health. Lancet 2002;360:123342.