Document Outline

). Lead exposure can afflict both children and

adults, but the greatest concern is for children, who will
experience symptoms at a significantly lower blood lead
levels (BLLs) than the adults. The most commonly used index
of lead exposure is the measurement of lead in blood. In 1991,
the US Centers for Disease Control and Prevention (CDC)
established acceptable BLLs of 100

μg/L for children (CDC

1991

). Recent epidemiology studies have found that BLLs

below 100

μg/L, a previously thought safe level, can still

result in significant cognitive impairment in children, and
there may be no threshold for adverse effects (Canfield et al.

2003

; Lanphear et al.

1998

Over the past several decades, childhood lead exposure has

attracted great attention in the USA, and childhood lead ex-
posure prevention efforts are sometimes called a victory in
light of the dramatic reductions in BLLs in children. The
major cause of the observed decline in BLLs is most likely
the removal of 99.8% of lead from gasoline the removal of

Responsible editor: Philippe Garrigues

J. Chen
Global Application Laboratory, PerkinElmer Analytical Sciences,
Shelton, CT 06484, USA

J. Chen

J. Xu

X. Liu

Shenzhen Center for Chronic Disease Control,
Shenzhen 518020, China

Y. Tong (

College of Physics and Technology, Shenzhen University,
Shenzhen 518060, China
e-mail: yongpeng@szu.edu.cn

Y. Li

M. Tan

Y. Li

Shanghai Institute of Applied Physics,
Chinese Academy of Sciences,
Shanghai 201800, China

Environ Sci Pollut Res
DOI 10.1007/s11356-012-0837-9

lead from soldered food cans (CDC

1991

; Pirkle et al.

1994

)

and community awareness programs. At the same time, how-
ever, this problem has not been a subject of concern in most
developing countries including China. BLLs in China are
generally higher than those in other countries due to the heavy
lead pollution (Shen et al.

1999

; Wang and Zhang

Numerous cases of lead poisoning have been documented
across the country, and research studies indicate that vehicle
exhausts, industrial emissions, and coal combustion are the
three major lead pollution sources in Chinese cities (Cheng
and Hu

). Recently, evidence reported that lead in child-

ren

’s blood is mainly caused by coal-fired ash in Shanghai,

China, by tracing lead isotopic composition (Liang et al.

The environmental Pb, especially Pb from gasoline deposited
during earlier times in large industrial cities, may also threaten
children

’s health. The evaluation of the risk of environmental

Pb is very important nowadays in China.

Lead has four naturally occurring stable isotopes

(

204

Pb,

206

Pb,

207

Pb, and

208

Pb). Only

204

Pb is non-

radiogenic; the other three are the end-products of uranium
(

238

235

U) and thorium (

232

Th) radioactive decay. The isoto-

pic characteristics of lead are dependent upon the original
composition and age of the ore bodies (Dickin

1995

). Because

the isotopic composition of lead is not affected to a measurable
extent by physical and/or chemical fractionation processes,
lead isotopic composition provide a powerful tool for charac-
terizing sources and pathways of lead exposure (Chen et al.

; Gulson et al.

; Zheng et al.

2004

To determine recent children

’s blood lead levels and iden-

tify sources of lead exposure, a blood lead screening survey
program with stratified cluster sampling method for children
residing in the Pearl River Delta (PRD) region was conducted
from Dec 2007 to Jan 2008. The level of lead and the stable
lead isotope ratios in venous blood were measured by using
inductively coupled plasma mass spectroscopy (ICP-MS).
Moreover, various urban environmental samples relating to
lead exposure sources, including residential dusts, vehicle
exhaust particles, and water, etc, were also evaluated to iden-
tify possible pathways of lead in blood.

2 Materials and methods

2.1 Selection of subjects

The PRD region is situated in the southern part of Guangdong
Province and is one of the most economically developed areas
in South China. Two megacities, Guangzhou and Shenzhen,
are located over a territory about 150 km in length and 100 km
in width. Figure

shows the geographic map of PRD region.

As its core metropolitan cities, they demonstrate minor differ-
ences in terms of economic structure, growth dynamics, de-
gree of internationalization, and transport infrastructure to

some extent. Focused on these important subpopulations, a
blood lead screening survey program was conducted from
Dec 2007 to Jan 2008. In this program, we provided multi-
stage method under the condition of stratified clustered sam-
pling, and the sampling frame was organized by the type of
land use. As a practical matter, area of residence was catego-
rized as urban, residential, and rural sites for the purposes of
this study, and then distinct subgroups from large population
were randomly selected out of the two cities. The detailed
information of children was examined to ascertain the back-
ground characteristics which were given in Table

A total of 761 children, 368 residing in Guangzhou and

393 residing in Shenzhen, all aged between 2 and 12 years

Fig. 1 The geographic map of PRD with the location of Guangzhou and
Shenzhen cities

Table 1 Background information for the 761 participants entered this
survey program

Characteristics

Guangzhou (n0368)

Shenzhen (n0393)

Age in years (range)

6.87±2.40 (2

–12)

6.98±2.33 (2

–12)

Gender

Male

188 (51)

204 (52)

Female

180 (49)

189 (48)

Area of residence

Urban

212 (58)

218 (55)

Residential

118 (32)

139 (35)

Rural

38 (10)

36 (9)

Values represent mean±SD or number of subjects (percent)

Age, gender, and area of residence were not significantly different between
Guangzhou and Shenzhen as evaluated by Student

’s t test

Environ Sci Pollut Res

with a median age of around 7 years, were considered for
this study. All of the subjects in the two groups, defined by
different sampling city which involved both 24 kindergarten/
elementary schools, respectively, were almost similar in age,
sex, areas of residence, traffic density, life style, and socioeco-
nomic status. Subjects did not report any accidental or occupa-
tional exposure to lead with no symptoms suggesting lead
toxicity, and therefore, the main source of the detected lead in
blood is expected to be exposures through environmental me-
dia, such as air, water, food, and dusts.

The protocol for this survey was approved by the Institu-

tional Ethnic Board of China National Center for Disease
Control and Prevention. The survey program was explained
to parents or guardians of the children, and their written
consent together with a filled predesigned questionnaire were
obtained before sampling. Participants living in the current
communities less than 1 year were excluded from this pro-
gram. After arm-washing and cleaning with alcohol swabs,
approximately 3.0 mL venipuncture blood samples were
obtained from each child and collected into lead-free BD
Vacutainer tubes containing sodium heparin as an anticoagu-
lant by trained nurses (Becton Dickinson, Franklin Lakes, NJ,
USA). The blood specimens were directly transported under
ice-cold conditions to the lab for analysis.

With the aim mainly to identify sources of lead exposure in

terms of normal children in this program, typical environmen-
tal samples, such as water, vehicle exhaust particles, and
residential dusts were also evaluated. Among them, 12 water
samples, 66 residential dusts samples from Shenzhen, and 59
dusts samples from Guanzhou were collected locally at the
selected sampling sites, respectively. A total of 15 brands of
paints samples were obtained from local organizations of En-
vironment Protection Authority.

2.2 Laboratory analyses

Blood specimens were diluted 20-fold with a dilute solution
containing 0.1% (v/v) ultrapure nitric acid (Tamapure-AA 10,
TAMA Chemicals, Japan) and 0.1% Triton X-100 in de-
ionized water (Milli-Q A10, Milipore, Billerica, MA, USA).
Indium and bismuth at 10 ng m/L concentration were added
online by a T-type canal as mixed internal standard solution.
An ELAN DRC II (PerkinElmer-SCIEX, Concord, Ontario,
Canada) ICP-MS with dynamic reaction cell (DRC

™) was

used for the measurements under standard mode. The samples
were introduced by a peristaltic pump and a baffled quartz
cyclonic spray chamber was used along with a concentric
nebulizer of 0.4 mL/min uptake rate because of its excellent
sensitivity and reproducibility. Standard addition calibration
provided in the ELAN software was used for calibration with
concentrations ranging from 0 to 50

μg/L. The field reagent

blanks normally showed less than 0.5

μg/L lead in final

solution, which assured negligible contamination problems.

The accuracy of the developed method was evaluated by ana-
lyzing Seronorm

™ Trace Elements Whole Blood L-1, L-2, and

L-3 reference materials (SERO AS, Billingstad, Norway). The
measured results (28.8, 396, and 497

μg/L, respectively)

showed excellent agreement with the certified values (27.6,
393, and 503

μg/L, respectively) in all lead levels.

For lead isotope ratio measurement by ICP-MS, nitric acid

extraction of blood specimens was employed instead of direct
dilution in order to minimize the matrix influence. Briefly,
1 mL venous blood was placed into a 15-mL pre-cleaned
polypropylene centrifuge tubes, to which 9 mL of 5%
HNO3 (v/v) was added and loosely screwed with cap. The
tubes were heated in a water bath maintained at 90 C for
30 min and were centrifuged after that. The supernatant sol-
utions were directly decanted into autosampler vials or were
further diluted with de-ionized water to obtain a lead concen-
tration below the limit of 25

μg/L. SRM 981 (Common Lead

Isotopic Standard, National Institute of Standards and Tech-
nology, Gaithersburg, MD) was used to determine the mass
bias correction factors of isotope ratios. For routine analyses
of lead isotope ratios following the procedure given in the
previous work (Chen et al.

), without any chemical lead

purification of the resulting solutions, the typical uncer-
tainty level was about 0.07% for

207

Pb/

206

Pb, 0.09% for

208

Pb/

206

Pb, and 0.3% for

206

Pb/

204

Pb, respectively.

More detailed descriptions about the digestion program for

the environmental samples such as water, paints, vehicle
exhaust particles, and residential dusts can also be found
elsewhere in the same previous study (Chen et al.

2.3 Statistical analyses

BLLs were log transformed to improve normality. Statistical
analyses were performed using SPSS version 13.0 software
(SPSS Inc., Chicago, IL, USA). On the basis of BLLs, sub-
jects were checked by factors focus on age, gender, and lead
isotope ratios.

3 Results

3.1 BLLs in children

There were 761 children at the age of 2

–12 years from

Guangzhou and Shenzhen enrolled in the study. Table

shows

the geometric mean (GM) of BLLs distributed among age and
the BLLs in this work was expressed in micrograms per liter.
As corresponded to the community intervention level used in
the USA, Table

also provided the prevalence of elevated

BLLs which was calculated from the percentage of the num-
ber of children with higher BLLs than 100

μg/L. Overall in

this period, the geometric mean was 57.05

μg/L (95% CI,

55.55~58.58). The 64.5% of the children had higher BLLs

Environ Sci Pollut Res

than 50

μg/L, and 9.6% of the children had BLL higher than

100

μg/L.

Among the gender groups of total 761 children, 392 par-

ticipants were boys and 369 were girls. The geometric means
were 58.37

μg/L (95% CI, 55.53~62.65) and 56.14 μg/L

(95% CI, 53.61~58.70) respectively, which demonstrated
significant difference (P<0.05). Among the community
groups examined, 368 participants were from Guangzhou
and 393 were from Shenzhen. The geometric means were
found to be 58.71

μg/L (95% CI, 56.75~60.74) and

55.32

μg/L (95% CI, 53.09~57.64), respectively, which also

demonstrated significant difference (P<0.001).

3.2 Characteristics of the blood lead isotope ratios

This blood lead screening survey program was designed to
select participants who had been living in current communities
for more than 1 year. We differentiated them into two sub-
groups by which city they resided. We reported the blood lead
isotope ratios in children corresponded to different BLLs
which were divided by criteria of per 10

μg/L BLLs upward

shift. Figure

shows the comparison of lead isotope ratios in

blood collected in Guangzhou and Shenzhen, respectively.

From a general overall view, the blood lead isotope

ratios of the children in both the cities were tightly
clustered when the BLLs were lower than100

μg/L. By

using BLLs at the concentration of 100

μg/L as a thresh-

old, the mean average of blood lead isotope ratios in
Guangzhou were calculated as ranging between 0.859±0.002
to 0.861±0.004 for

207

Pb/

206

Pb and 2.108±0.004 to 2.110±

0.003 for

208

Pb/

206

Pb. In Shenzhen, it ranged between 0.858±

0.002 to 0.860±0.003 for

207

Pb/

206

Pb and 2.107±0.003 to

2.108±0.004 for

208

Pb/

206

Pb, respectively. It is worth noting

that lead isotope ratios of

207

Pb/

206

Pb and

208

Pb/

206

Pb in both

two cities had a similarly weak trend of going upward while

Table 2 The BLLs and the
prevalence of elevated
BLLs for children at the
age of 2

–12 years collected

in Guangzhou and Shenzhen
after 10 years of phasing
out of leaded gasoline

Age

No.

BLLs (GM)

95% Confidence interval

Percentage BLLs

≥100 μg/L

54.22

44.07~66.71

11.8

57.83

52.59~63.59

7.9

57.42

51.00~64.64

13.2

55.27

51.51~59.30

8.2

61.16

56.24~66.50

12.4

130

59.04

55.40~62.92

8.5

124

56.38

52.66~60.37

9.7

55.51

51.91~59.37

7.7

57.03

52.61~61.83

10.7

11~

51.61

46.33~57.50

8.5

Guangzhou

368

58.71

56.75~60.74

10.3

Male

188

59.09

56.44~61.86

13.8

Female

180

58.31

55.43~61.33

6.7

Shenzhen

393

55.32

53.09~57.64

8.9

Male

204

58.37

54.93~62.02

9.3

Female

189

52.30

49.55~55.21

8.5

Total

761

57.05

55.55~58.58

9.6

Fig. 2 A comparison of lead isotope ratios in blood for children at the
age of 2

–12 years collected in Guangzhou and Shenzhen after 10 years

of phasing out of leaded gasoline

Environ Sci Pollut Res

the children

’s BLLs increased. Isotope ratios of

207

Pb/

206

and

208

Pb/

206

Pb in both two cities were showed significant

different in 60

–70 μg/L group (P<0.05).

For the children with BLLs higher than 100

μg/L, our

findings clearly indicated that the isotopic compositions in
blood are quite different with obvious higher values of
0.863±0.003 and 0.862±0.003 for

207

Pb/

206

Pb in Guangzhou

and Shenzhen, respectively. For

208

Pb/

206

Pb, the values were

2.113±0.004 and 2.111±0.004, respectively. Since the geo-
metric mean value for children

’s BLLs in Guangzhou

(58.71

μg/L) were significantly higher than that of in

Shenzhen and (55.32

μg/L), the higher value of isotopic

composition in blood for both

207

Pb/

206

Pb and

208

Pb/

206

in Guangzhou correlated with higher children

’s BLLs coin-

cided with the weak trend of lead isotope ratios moving up
with children

’s BLLs. Furthermore, this fact combined with

clustered isotopic composition in children

’s blood means that

it can be applied to study the different contributions from
target media of lead exposure.

3.3 Isotopic composition in possible sources

The most important source of lead exposure for children
nowadays is from a wide range of environmental factors,
and Table

showed the lead concentrations and isotopic

composition in possible sources, including residential dusts,
vehicle exhaust particles, and water, etc.

4 Discussions

As hazards of lead exposure were recognized, the removal
of lead from gasoline was properly viewed as public health
triumph today and in the past decades BLLs have fallen

significantly in a number of countries (Pirkle et al.

1994

;

Shen et al.

1999

; Dixon et al.

2009

). It had been reported

that the mean BLLs of Chinese children was 93

μg/L (37–

254

μg/L), and 33.8% (9.6–80.5%) of the subjects had

BLLs higher than 100

μg/L with sampling time from 1994

to March 2004. Nine of the 27 provinces or cities in China
had reported average BLLs

≥100 μg/L. In boys, measured

average BLL was 96

μg/L, significantly higher than that

measured for girls

’ (89 μg/L, P<0.001) (Wang and Zhang

). When compared with literature data, our results

suggested that both the average BLLs (57

μg/L) and prev-

alence of elevated BLLs (9.6%) continuously decreased
remarkably. Such a decline of the order of about 35% 3 years
later in BLLs can be attributed to a number of factors.

As we know, the Chinese government started the phase-

out of leaded gasoline in 1991 nationally, and the sales and
uses of leaded gasoline were totally banned in Guangzhou
since October 1997 and in Shenzhen since January 1998.
Besides, several measures were taken to reduce the air
pollutant emissions. The Environmental Protection Plan in
the PRD region was developed in 2004. These programs
were effective in reducing some air pollutant emissions,
particularly primary particles from industrial sources in the
PRD region (Chan and Yao

2008

). All these policy changes

helped to reduce the environmental lead pollution and con-
tributed mainly to the significant decreasing trend in BLLs
in children. Meanwhile, great efforts were made on the food
industry in China to decrease food lead concentration with
strict examination and approval of food and packaged goods
containing lead, which also might have helped to reduce
children's exposure to lead. In addition, nutritional factors
are thought to play an important role in lead accumulation
and poisoning (Tahiri et al.

2000

). It is believed that child-

ren's diet and hygiene habits such as not washing their hands

Table 3 Pb concentrations and isotopic compositions of the urban environment

Sample

No.

Pb (mg/kg)

207

Pb/

206

208

Pb/

206

Mean

Range

Mean

Range

Mean

Range

Residential dust (Shenzhen)

181±117

–452

0.862±0.002

0.859

–0.866

2.116±0.005

2.109

–2.123

Vehicle exhaust past

11,913±2,899

8,076

–19,466

0.904±0.006

0.896

–0.911

2.194±0.009

2.184

–2.204

Residential dust (Guangzhou)

303±340

–1,026

0.866±0.005

0.859

–0.875

2.114±0.008

2.106

–2.132

Paint

6,185±15,879

–61,478

0.859±0.016

0.820

–0.877

2.113±0.020

2.057

–2.141

Water

0.34±0.18

0.13

–0.73

0.838±0.005

0.830

–0.845

2.093±0.012

2.069

–2.107

Regional background

0.837±0.010

0.819

–0.845

2.084±0.009

2.071

–2.096

Aerosol

2,000

–26,500

0.860±0.004

0.855

–0.866

2.119±0.007

2.110

–2.128

Eolian dust

0.856±0.004

0.852

–0.862

2.105±0.005

2.098

–2.112

Fankou lead ore

0.855±0.006

0.851

–0.859

2.113±0.012

2.105

–2.122

Chen et al. (

)

The data are both for Guanzhou and Shenzhen in micrograms per kilogram

Zhu et al. (

2001

)

Environ Sci Pollut Res

carefully before meals, hand-to-mouth behaviors, frequently
eating preserved eggs or canned food, and so on, are important
contributors to children's body lead burden in China (Shen et al.

1999

). Due to the enhanced knowledge of preventing lead

exposure, combined with increased dairy milk intake and pub-
lic health education, the intestinal absorption of lead in children
were reduced effectively.

However, we should pay attention to the fact that the

BLLs for children in the age group of 12

–60 months in the

USA between 1999 and 2004 was 20

μg/L. Eight percent of

the children (2,155) were

≥50 μg/L; 1.71% were ≥100 μg/L

(Dixon et al.

2009

). Our results strongly indicated that

children

’s BLLs together with prevalence of elevated BLLs

in PRD region of China are much higher than those of their
counterparts in developed countries at the similar sampling
time, which suggested that children's lead exposure preven-
tion and controlling would still be a long-term mission in
China.

Based on the consideration that the lead isotope ratios are

significantly more sensitive tracers than elemental concentra-
tions, the isotopic composition of lead was used to identify the
sources of exposure and gain insight into the prevention
management in this study. The lead isotope ratios in children

’s

blood samples corresponded to different BLLs in comparison
to those of lead-exposure-related samples plotted in Fig.

. In

order to observe the contribution from the decay of Th, taking
the Th-rich environment of China into account,

208

Pb/

206

and

207

Pb/

206

Pb ratios were used as Y and X axes, respectively.

The lead growth curve in Fig.

was taken from literature

(Cumming and Richards

1975

). The data regarding the past

vehicle exhaust was taken from a previous study, which was
done in 1995 when alkyl lead additives as antiknock com-
pounds were in use (Chen et al.

As can be seen in Fig.

, lead isotopic composition

distinctively fell into three different plots. The natural re-
gional background lead isotopic composition in PRD is

characterized by low values for both

207

Pb/

206

Pb (0.819

–

0.845) and

208

Pb/

206

Pb (2.071

–2.096) ratios. This was

closely related to the signature of drinking water collected
in PRD with the values of

207

Pb/

206

Pb ranges from 0.830 to

0.845 and

208

Pb/

206

Pb ranges from 2.069 to 2.107. Thus, it

is clearly understood that the drinking water showed the
characteristics inherited from natural regional background.

On the contrary, the isotopic composition of past vehicle

exhaust went in the opposite direction with extremely high
values of

207

Pb/

206

Pb (0.896

–0.911) and

208

Pb/

206

(2.184

–2.204). This could be due to the fact that the Pb in

alkyl lead additives as antiknock compounds used in China
exclusively originated from the Australian geologically old
mines of Broken Hill (Wang et al.

2002

). It had also been

reported that lead contamination in the environment of the
neighboring PRD region could be linked to the input of a
high

207

Pb/

206

Pb anthropogenic source, such as Australian

ore, in the leaded gasoline (Duzgoren-Aydin et al.

2004

Those vehicle exhaust past, also presented in Table

, were

considerable produced from leaded gasoline feedstocks, and
its isotopic signatures which located quite away from residen-
tial dust and children

’s blood, indicated it is no longer the main

source of lead exposure.

Based on the data of Table

, the average

207

Pb/

206

value of the Shenzhen residential dust is 0.862 and that of
Guangzhou is 0.866. Actually, residential dust, aerosol, and
vehicle exhaust now samples are similar Pb sources as no Pb
in gasoline now. From our previous result, it is known that
those Pb-containing dusts ( residential dust, aerosol, and
vehicle exhaust now) are mainly contributed coal-fired ash
(Liang et al.

) and flying soil dust which polluted by

deposited gasoline Pb. The average

207

Pb/

206

Pb value of

coal-fired ash is 0.859 (Liang et al.

), and the average

207

Pb/

206

Pb value of gasoline Pb should be 0.904 as we

determined before (Chen et al.

). So it can be simply

calculated that there is about 6.7% past used gasoline Pb
embedded in Shenzhen residential dust and about 15.6% in
Guangzhou dust, respectively. Considering the sea as a sink
of environmental gasoline Pb which had been accumulated
for more than 20 years before 1998, the environmental
gasoline Pb of Shenzhen (0 km faraway from the sea) spreads
more quickly than that of Guangzhou (150 km faraway from
the sea).

The isotopic content of lead introduced into a child body

remains constant over time and changes only if another source
of lead with differing lead isotopic content is introduced
(Gulson et al.