International Journal of Hygiene and Environmental Health 216 (2013) 243–249

Contents lists available at SciVerse ScienceDirect

International Journal of Hygiene and

Environmental Health
journal homepage: www.elsevier.com/locate/ijheh

Reference values for lead, cadmium and mercury in the blood of adults from the
metropolitan area of Sao Paulo, Brazil
Rúbia Kuno a,∗ , Maria Helena Roquetti a , Kerstin Becker b , Margarete Seiwert b , Nelson Gouveia c
a
Companhia Ambiental do Estado de São Paulo, CETESB, Av. Prof. Frederico Hermann Jr., 345, Sao Paulo, SP 05459-900, Brazil
b
German Federal Environment Agency (UBA), Dessau-Roßlau/Berlin, Germany
c
Faculdade de Medicina da Universidade de São Paulo, Departamento de Medicina Preventiva, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: Human biomonitoring is an important tool for the evaluation of environmental exposure to contami-
Received 8 December 2011 nants. The data that are obtained from these studies might be compared to appropriate reference values
Received in revised form 20 May 2012 (RVs) in a specific population. The RVs were derived from the rounded values of the upper limit of the
Accepted 24 May 2012
95th confidence interval of the 95th percentile for lead (Pb), cadmium (Cd) and mercury (Hg) in blood
from adults in the metropolitan area of Sao Paulo (MASP), Brazil to investigate the association between
Keywords:
blood metals and sociodemographic and lifestyle factors. Blood samples from 653 nonsmoking blood
Reference values
donors without occupational exposure to the studied metals were collected in 2006. Our evaluations
Lead/blood
Cadmium/blood
distinguished a younger group (18–39 years) and an older group (40–65 years). RVs in the younger group
Mercury/blood were 60 ␮g Pb/L and 4 ␮g Hg/L for men and 47 ␮g Pb/L and 4 ␮g Hg/L for women. RVs in the older group
Environmental monitoring were 80 ␮g Pb/L and 5 ␮g Hg/L for men and 63 ␮g Pb/L and 6 ␮g Hg/L for women. The RV for Cd was
Environmental quality 0.6 ␮g/L for adults aged 18–65 years. Pb and Cd levels demonstrated a significant association with sex
Environmental health and age. Male blood contained 50% more Pb, and the older group exhibited 23% more Pb. Fish consump-
tion and amalgam fillings were primarily related to Hg levels. RVs for lead were similar to the Czech
Republic and Germany but higher than the US population. The RV for Cd in Brazil was well below the
RVs of these countries. The RVs for Hg in Brazil were similar to the US but higher than Germany and the
Czech Republic.
© 2012 Elsevier GmbH. All rights reserved.

Introduction Alternatively, tolerable limits are established based on health

Human biomonitoring (HBM) is an Environmental Health tool
to support consistent decisions. HBM results should be compared
to appropriate references values (RVs) or tolerable limits. RVs are
established statistically and represent the upper limit of the back-
ground exposure of the general population. Our study calculated
RVs as the rounded upper value of the 95% confidence interval of
the 95th percentile. These values describe the body’s chemical load
statistically, they do not distinguish between hazardous and non-
hazardous levels (Schulz et al., 2007a). RVs are used to discover
and compare populations with high exposures to contaminants in
environmental and health risk assessments.
Some countries derive RV for the general population using data
from surveys of large numbers of individuals, such as the German
Environmental Survey (GerES) (Becker et al., 2002; UBA, 2010) and
the National Health and Nutrition Examination Survey (NHANES,
CDC, 2010).
∗ Corresponding author. Tel.: +55 11 3133 4068; fax: +55 11 3133 3982.
E-mail addresses: rkuno@terra.com.br, rubiak@cetesbnet.sp.gov.br (R. Kuno).
effects. The Human Biomonitoring Commission of the German Fed-
eral Environment Agency (Umweltbundesamt – UBA) has defined
health-related HBM values. An HBM I value is the threshold below
which adverse health effects are not expected in the general pop-
ulation. Concentrations above the HBM II value indicate a risk
of health effects, and exposure sources should be identified and
reduced. Concentrations between HBM I and HBM II require a
re-examination of the analytical results and an investigation of
possible contamination sources (Schulz et al., 2007b).
RVs from other countries, primarily the US and Germany, are
used to compare data from exposure assessment studies in indi-
viduals in contaminated sites due to the absence of data on the
background levels of contaminants in the Brazilian population.
The Environmental Agency of Sao Paulo State (CETESB) has
identified a number of contaminated industrial waste sites in the
metropolitan area of Sao Paulo (MASP). A survey in 2009 revealed
that approximately 50% of the 2904 contaminated sites in Sao Paulo
State are located in the MASP. Metals are among the main contam-
inants in these contaminated sites, which also include aromatic
solvents, liquid fuels, polycyclic aromatic hydrocarbons (PAHs) and
halogenated solvents (CETESB, 2009).

1438-4639/$ – see front matter © 2012 Elsevier GmbH. All rights reserved.
http://dx.doi.org/10.1016/j.ijheh.2012.05.010
244 R. Kuno et al. / International Journal of Hygiene and Environmental Health 216 (2013) 243–249
Lead (Pb), cadmium (Cd) and mercury (Hg) are naturally present
in the environment, but the most important human exposures
result from human activity (McKelvey et al., 2007). These met-
als are important environmental contaminants due to their severe
effects on human health. The present study proposed reference
values for Pb, Cd and Hg in the blood of the adult population in
MASP and investigated the association between blood metals and
sociodemographic and lifestyle factors in this population.
Materials and methods
Study design and period
We performed a cross-sectional study to determine the concen-
trations of lead, cadmium and mercury in the blood from a sample
of adults who lived in the MASP (blood donors). Blood samples were
collected from November 15 to December 16, 2006.
Study group
A sample size of 533 individuals was calculated (precision of
5%) based on the standard deviation of the blood lead levels that
were obtained in a pilot study (Kuno et al., 2007). In order to allow
for possible losses, during field work we sought to recruit more
blood donors to ensure the calculated sample size was reached. The
following inclusion criteria were applied: blood donors who lived
in one of the municipalities within the MASP for at least a month
and who were nonsmokers without occupational exposure to lead,
cadmium or mercury. In 2006, the MASP had almost 19 million of
inhabitants (SEADE, 2006).
Sample collection and metals analysis
Blood was collected using venous phlebotomy and stored in
sodium heparin anticoagulant vacutainers that were prescreened
for background lead, cadmium and mercury contamination. Two
vacutainer tubes were collected with 4–5 mL of whole blood each
one in addition to the 400 mL collected for donation. One tube
was used for Pb and Cd determinations and the other was sent to
UBA for Hg determination. Immediately after collection, samples
were kept and transported refrigerated to the Cetesb’s laboratory,
where they were kept frozen until the analysis. The samples sent
to Germany were shipped in containers suitable for this type of
material with dry ice. Duplicate lead, cadmium and mercury anal-
yses were performed on each sample, and the arithmetic mean of
the duplicates was recorded for each participant. Blood lead and
cadmium were assayed using graphite-furnace atomic absorption
spectrometry in a Perkin Elmer SIMA 6000 with AS 72 according to
Kummrow et al. (2008) at CETESB’s laboratory. The limits of quan-
tification (LOQ) for lead and cadmium were 2.5 ␮g/L and 0.10 ␮g/L
respectively. The limits of detection (LOD) were 0.65 ␮g/L for lead
and 0.026 ␮g/L for cadmium. Mercury determination was assayed
using cold vapor atomic spectrometry (Schaller, 1988) in the Ger-
man Federal Environment Agency in Berlin as part of a cooperation
agreement between the UBA and CETESB. The LOQ for mercury was
0.2 ␮g/L and the LOD was 0.3 ␮g/L.
Lyphocheck® Whole Blood Metals Control Level 1 (Bio-Rad,
Irvine, USA) and SeronormTM Trace Elements in Whole Blood Level
1 (Sero AS, Billingstad, Norway) reference materials were used for
lead and cadmium analytical quality assurance. The standard mate-
rial for mercury was provided by IPASUM (Institute and Out-Patient
Clinic for Occupational, Social and Environmental Medicine of the
University Erlangen-Nuremberg) at a concentration of 1.08 and
2.88 ␮g/L.
Questionnaire and informed consent
Basic demographic data and information on lifestyle, including
educational level, leisure activities, income, fish and alcohol con-
sumption, source of drinking water and possible environmental
exposures, were collected by personal interviews for the estima-
tion of population exposure to contaminants. The Ethics Committee
of the School of Medicine of the University of Sao Paulo and The
National Commission of Ethics in Research (CONEP) of the National
Health Council, an agency of the Brazilian Ministry of Health,
approved the research protocol. Participation in the study was vol-
untary, and all subjects signed a Free Informed Consent Term.
Statistical analysis
A descriptive analysis was performed for the sociodemographic
characteristics of the study group and the blood metal concen-
trations. Concentrations below the LOQ were set to 1/2 LOQ for
computational purposes. The statistical parameters must be inter-
preted with caution when a large proportion of blood metal
concentrations were below the LOQ.
Bivariate regression analyses were performed between blood
lead and mercury concentrations (PbB and HgB, dependent vari-
ables) and each of the following determinants (independent
variables): sex, age group, education level, fish consumption, alco-
hol consumption, source of drinking water, medication intake,
chewing gum use, amalgam fillings, recent house refurbishment,
garden at home, industrial plant near home, income per capita, civil
status, last blood donation, total of donations and blood pressure.
The most important covariables (p ≤ 0.2) in bivariate regression
were included in the multivariate models. The multiple regression
procedure identified the significant covariables using a stepwise
approach. The final models retained variables that were predictive
of blood metal levels at a significance level of ˛ = 0.05.
For bivariate and multivariate regression analyses, the blood
metal concentrations were log-transformed (natural logarithms)
because this transformation produced approximately normally dis-
tributed values. The regression coefficients (b) were transformed
back (exp(b)) to facilitate the interpretation. A back-transformed
coefficient expresses the relative increase in the exposed group’s
blood metal concentration compared to the unexposed (or refer-
ence) group.
Only a descriptive analysis for the total study group was con-
ducted for Cd in blood (CdB). Bivariate or multiple regression
analyses were not performed for Cd because most concentrations
were below the LOQ.
An RV was computed as the rounded upper value of the 95%
confidence interval of the 95th percentile (P95) for the reference
group. The Statistical Package for the Social Sciences (SPSS) version
13.0 was used for all statistical analyses.
Results
Study participants included 653 blood donors from nine blood
centers in the MASP. We collected 653 blood samples, but 114 sam-
ples were not analyzed for Pb and Cd, and 60 samples were not
analyzed for Hg. Therefore, 539 samples were analyzed for Pb and
Cd, and 593 were analyzed for Hg. We obtained results for all three
metals for 488 samples. Losses were due to technical problems that
precluded analysis.
Donors were of both sexes, aged between 18 and 65 years and
residents in the MASP. The general characteristics of the study
population are shown in Table 1. Blood donors resided in 18 munic-
ipalities of the MASP, the majority of subjects resided in Sao Paulo
city. The mean income of the study population was R$ 530,00
 R. Kuno et al. / International Journal of Hygiene and Environmental Health 216 (2013) 243–249 245

Table 1 exhibited 80%, 60% and 45% more Hg, respectively, compared to
Characteristics of the population.
donors who never ate fish.
N (%) People with amalgam fillings exhibited 24% higher HgB lev-
Sex els compared to donors who had no fillings. The 40–65-year-old
Men 364 (55.7) donors exhibited approximately 19% more Hg compared to the
Women 289 (44.3) 18–30-year-old donors.
Total 653 (100) Education level was negatively associated with HgB. Individu-
Age
als with a lower education level had lower HgB levels compared
18–39 466 (71.4)
40–65 187 (28.6) to the highest education group. People with basic (0–4 years)
Total 653 (100) and middle (5–11 years) education levels demonstrated approx-
Education level imately 30% less HgB than people in the highest education level
0–4 years 131(20.2)
(≥12 years).
5–11 years 379 (58.5)
≥12 years 138 (21.3) Table 6 shows the proposed RVs for blood metal concentrations
Total 648 (100) that were derived from the 95% confidence intervals of the 95th
Income ($ Reaisa ) percentiles.
Up to 350b 259 (42.2)
351–1225 325 (52.9)
1226–2800 28 (4.6)
2801–3501 2 (0.3) Discussion
Total 614 (100)
a
$ Reais = Brazilian currency (1 USD = 2.15 Reais). Blood donors are an attractive alternative for the examination
b
R$ 350 refers to the minimum wage at that time. of blood contaminants within a population. Several environmental
contaminant exposure studies of the general population have used
Table 2 blood donors as a sample group. Studies conducted in Barcelona
Pb, Cd and Hg concentrations (␮g/L) in the blood of the study group. (Torra et al., 1997), Brescia/Italy (Braga, 1992), Orava and Priev-
idza/Slovak Republic (Buchancová et al., 1994), Bari/Italy (L’Abbate
Lead Cadmium Mercury
et al., 1991) and Belgium (Ducoffre et al., 1990) have used blood
N 539 539 593 donors to assess metal concentrations in the blood of the gen-
AM 27.1 0.10a 1.40
eral population. In Belgium, blood donors were chosen as a sample
STDDEV 15.5 0.15 1.40
GM group from 1978 to 1985. A pilot survey conducted in 1978 demon-
Males 28.0 0.08a 0.95 strated that no significant differences were observed between the
Females 18.6 0.10a 1.07 PbB levels of a population group selected at random and a group of
Total 23.7 0.08a 0.98
blood donors (Ducoffre et al., 1990). Other countries also use blood
95%CI 22.4 – 24.7 0.08a –0.09a 0.90–1.05
Minimum 1.3 <0.10 0.10
donors to establish reference values. The Czech Republic has rou-
Maximum 131.0 1.70 12.40 tinely conducted biomonitoring to assess population exposure to
N < LOQ 0 348 9 environmental contaminants using blood donors aged 18–58 years,
LOQ 2.5 0.10 0.2 children aged 5–6 and 8–10 years, nursing mothers and autopsy
Notes: AM, arithmetic mean; STDDEV, standard deviation; GM, geometric mean; material since 1994 (Černá et al., 2007).
95%CI, CI P 95 = 95% confidence interval of the total geometric mean; LOQ, limit of The RVs and geometric means of Pb, Cd and Hg in the present
quantification; N < LOQ, number of values below LOQ.
a
study demonstrated that the population was not exposed to very
Statistical parameter below LOQ. Values below LOQ were set to LOQ/2.
high environmental levels of these metals. Sex and age were
strongly associated with the concentrations of these metals in
whereas the mean income at MASP was R$ 824,00 in 2006 (Dantas, blood, especially for lead. Therefore, we chose to stratify the RVs
2007). The income was expressed in Reais, the Brazilian currency according to these variables for lead and mercury.
(1 USD = 2.15 Reais). Table 7 shows the RVs for some population groups worldwide.
Table 2 illustrates the distributions of PbB, CdB and HgB concen- The RVs for lead in the present study for the oldest group, 80 ␮g/L
trations (␮g/L) in the entire study group. for men and 63 ␮g/L for women were similar to the Czech Republic
Table 3 shows the relationship between potential determinants and very close to the German population aged 18–69 years. The RV
of PbB and HgB. Sex, age, education level, the consumption of for adults in Germany was derived using data from a survey that
alcoholic beverages, use of chewing gum, garden at home and was conducted between 1997 and 1999, and this data has not been
blood pressure were significantly associated with lead in the blood updated. The RVs that were derived for the MASP are smaller than
(p ≤ 0.05). Education level, fish consumption, the ingestion of med- the proposed RVs in seven areas of Italy in 2000 (Apostoli et al.,
ications and amalgam fillings were significantly associated with 2002). However, the RVs are higher than the 2003–2004 NHANES
HgB. (CDC, 2009) values, primarily the RVs in the older group. The differ-
Tables 4 and 5illustrate the results of multiple linear regres- ences between studies may be due to the gap between the historical
sion analyses for lead and mercury, respectively. The final model periods in which restrictions on the use of leaded gasoline were
(Table 4) revealed the two variables that significantly contributed adopted. Brazil suspended the use of lead in gasoline in the 1980s,
to lead in the blood were sex (men exhibited approximately but the gasoline that is used in small aircraft still contains this metal
50% more lead than women) and age (older group, 40–65 years as an additive (Paoliello and De Capitani, 2007; Lemos et al., 2008).
old, contained approximately 23% more lead in their blood than Additionally, the differences may be due to the inclusion of smokers
blood donors between 18 and 39 years). No other potential in the studies cited above.
explanatory variables remained significant after controlling for sex Low levels of CdB were observed in our study group, and 65%
and age. were below the LOQ. The exclusion of nonsmokers is not suffi-
Fish consumption was the primary contributor to HgB levels cient to explain the smaller values compared to studies worldwide.
(Table 5). Donors who ate fish daily or more than once a week This difference suggests lower environmental exposure to this
showed 107% more HgB than donors who never ate fish. People metal. The main sources of Cd in nonsmokers are food and envi-
who ate fish once a week, 2–3 times a month and once a month ronmental factors. Data from environmental monitoring programs
246 R. Kuno et al. / International Journal of Hygiene and Environmental Health 216 (2013) 243–249

Table 3
Relationship between variables and blood lead and mercury concentrations.

Variables Lead Mercury

N Exp (ˇ) P N Exp (ˇ) P

Sex <0.001 0.874

Male 316 1.51 328 0.99
Female 223 1.00 265 1.00
Age <0.001 0.078
18–39 392 1.00 418 1.00
40–65 147 1.21 175 1.13
Education level 0.015 <0.001
0–4 years 112 1.28 114 0.60
5–11 years 313 1.11 341 0.67
≥12 years 109 1.00 133 1.00
Fish consumption 0.977 <0.001
Daily or more than once a week 24 1.14 30 2.26
Once a week 89 1.05 103 1.88
2–3 times per month 68 1.08 74 1.73
Once a month 231 1.03 258 1.50
None 109 1.00 115 1.00
Alcoholic consumption 0.048 0.177
Daily or more than once a week 25 1.51 24 1.36
Weekend 218 1.15 234 1.09
None 281 1.00 323 1.00
Drinking water 0.773 0.061
Mineral 169 1.04 186 0.98
Well water 21 1.40 24 1.15
Water supply network 333 1.00 371 1.00
Use of medication 0.282 0.043
Yes 123 0.94 138 1.17
No 404 1.00 448 1.00
Chewing gum 0.002 0.731
Yes 259 0.87 286 1.02
No 268 1.00 299 1.00
Amalgam fillings 0.345 <0.001
Yes 362 0.95 400 1.33
No 150 1.00 163 1.00
Recent house refurbishment 0.691 0.794
Yes 205 1.02 230 1.02
No 322 1.00 356 1.00
Garden at home 0.030 0.386
Yes 23 1.28 25 0.87
No 504 1.00 560 1.00
Industrial plant near home 0.332 0.401
Yes 162 0.95 166 1.06
No 358 1.00 409 1.00
Income (Reais) 0.128 0.180
Up to 350 220 1.17 235 0.65
351–1225 262 1.10 299 0.76
>1226 24 1.00 28 1.00
Marital status 0.359 0.291
Married 287 1.00 313 1.00
Single, widowed, divorced 251 0.96 279 0.93
Last blood donation 0.075 0.077
1st time 254 0.92 276 0.89
>3 months 277 1.00 306 1.00
Number of donations 0.676 0.676
1st time 242 0.92 258 0.97
1–5 230 1.01 255 1.12
>5 46 1.00 51 1.00
Blood pressure <0.001 0.770
High 106 1.22 125 0.98
Normal 429 1.00 464 1.00

Note: Bivariate analysis with log-transformed blood metal concentrations.

demonstrate that the MASP has low levels of Cd (surface water,
groundwater and soil) (CETESB, 2004, 2006, 2007a,b; Lemos et al.,
2008). The LOQ of the analytical methods that are used in differ-
ent studies might partially explain the variability of the results.
Batáriová et al. (2006) considered the influence of LOQ on the higher
mean of Cd in the blood of Czech nonsmokers compared to Ger-
man nonsmokers. The LOQ of Cd determination in the German
survey was 0.12 ␮g/L, but the LOQ of Cd in the Czech Republic was
0.20 ␮g/L. The LOQ in our study was 0.10 ␮g/L which is closer to the
German LOQ.
The RV for CdB in other studies can only be compared to our RV
if the study population was restricted to nonsmokers. The data of
German nonsmokers from 1997 to 1999 yielded an RV of 1 ␮g/L.
Czech nonsmokers exhibited an RV of 1.1 ␮g/L that was derived
from data that was collected during 2001–2005, an RV of 0.5 from
data in 2005, and an RV of 0.3 ␮g/L from data in 2007. Our RV of
0.6 ␮g/L falls within this range.
Hg means were markedly below the limits based on health
effects in our study. Our RVs for HgB were similar to the val-
ues in NHANES (USA), and above to the German (2 ␮g/L) and the
 R. Kuno et al. / International Journal of Hygiene and Environmental Health 216 (2013) 243–249 247

Table 4 The RV for Hg in the subgroup with fish consumption up to 3

Results of the multiple linear regression of lead concentrations in blood.
Model Exp (ˇ)
Sex: male (reference: female) 1.52
Age group: 40–65 years (reference: 18–39 years) 1.23
Note: Stepwise regression analysis with log-transformed PbB.
Table 5
Results of the multiple linear regression of mercury concentrations in blood.
Model Exp (ˇ)
Eat fish daily or more than once a week (reference: no fish) 2.07
Eat fish once a week (reference: no fish) 1.81
Eat fish 2–3 times per month (reference: no fish) 1.61
Eat fish once a month (reference: no fish) 1.45
0–4 years of schooling (reference: ≥12 years) 0.67
5–11 years of schooling (reference: ≥12 years) 0.71
Amalgam fillings (reference: no amalgam filling) 1.24
Age group: 40–65 years (reference: 18–39 years) 1.19
Note: Stepwise regression analysis with log-transformed HgB.
Czech Republic. Recent data indicate that the medians of blood
mercury in Czech adults in recent years (2005 and 2007) are sim-
ilar to the medians of the donors in the MASP. However, the P95
of the Czech population in 2005 did not exceed 4 ␮g/L, and this
value was below 3 ␮g/L in 2007 (Puklová, 2008). These values are
clearly below the health-based German HBM I value. These data
are important because the derived RV (upper limit of 95% of P95)
for mercury in our study for both men and women aged 40–65
(5.1 ␮g/L and 6.1 ␮g/L, respectively) were above the threshold for
adults in Germany (HBM I, 5 ␮g/L). Of all of the donors, 1.3% (8
cases) exceeded the HBM I. However, the latest survey in the Czech
Republic revealed that only 0.7% (2 cases) exceeded HBM I (Puklová,
2008). This limit was based on data for women of reproductive age
but is also recommended for other groups.
times per month, 4 ␮g/L, in our study was twice the RV in Germany.
P Most of the blood donors had low fish consumption (n = 447). Nev-
<0.001 ertheless, a strong positive gradient of fish consumption frequency
<0.001 and HgB levels was observed (Table 5). Fish consumption was most
strongly associated with increasing levels of HgB, which is consis-
tent with previous studies (McKelvey et al., 2007; Gundacker et al.,
2006; Oskarsson et al., 1996). Donors who ate fish daily or more
than once a week exhibited levels over 100% higher than donors
who ate no fish. Education level, amalgam fillings and age were
P also strongly related to Hg blood concentrations (Table 5).
The influence of education may be related to the habit of eating
<0.001
<0.001
fish because people with higher education levels had higher con-
<0.001 centrations of HgB. New Yorkers (McKelvey et al., 2007) with higher
<0.001 incomes have higher levels of mercury. These authors assumed that
<0.001 wealthier individuals consumed more expensive fish, which is at
<0.001
the top of food chain, with higher levels of Hg. Income was not a
0.002
0.013 significant explanatory variable for HgB in our study, which may be
due to the small number of donors with higher income. Gundacker
et al. (2006) observed that Austrian blood donors exhibited a pos-
itive association between education and Hg in men. These authors
suggested that individuals with higher education consumed more
fish because of the positive association between seafood consump-
tion and education level.
Amalgam fillings were also associated with HgB, which is con-
sistent with previous studies (Oskarsson et al., 1996; Gundacker
et al., 2006). Gundacker et al. (2006) observed that age was posi-
tively correlated to HgB. This finding was confirmed in our study;
the older group exhibited approximately 20% higher Hg levels than
the younger group. Conversely, older Austrian men, 51 and 65 years
old, displayed lower levels of Hg, but this effect was not observed
in women. These authors suggested a different pattern of fish con-
sumption at different ages.
Fish consumption and dental amalgam were the main routes of
mercury exposure in the present study. The use of dental amal-
gam in Brazil has decreased in recent years, and the content of

Table 6
Lead, cadmium and mercury in the blood of the study group and proposed reference values (␮g/L).

Metal/groups N P10 P50 GM P90 P95 95%CI (P95) RV

Lead
Men
18–39a 234 15.05 26.46 26.46 46.63 54.76 50.20–59.73 60
40–65b 82 18.56 32.94 32.94 58.47 68.81 59.30–79.84 80
Total 316 16.37 27.55 28.01 48.85 65.93 56.10–75.70 76e
Women
18–39 158 9.02 17.61 17.61 34.38 41.56 36.68–47.09 47
40–65 65 10.65 21.26 21.26 42.43 51.61 42.21–63.10 63
Total 223 10.10 18.00 18.60 35.73 44.20 36.90–48.90 49e
Total lead 539 12.60 23.60 23.65 44.00 56.20 48.90–66.10 66e
Cadmium
Nonsmoking adults
18–65 539 0.05d 0.05d 0.082d 0.30 0.50 0.40–0.60 0.6
Mercury
Men
18–39 234 0.33 0.94 0.94 2.61 3.57 2.96–4.30 4
40–65 94 0.43 1.14 1.14 3.27 4.05 3.21–5.10 5
Total 328 0.40 0.91 0.986 2.71 3.955 3.20–4.30 4e
Women
18–39 185 0.37 0.94 0.94 2.42 3.16 2.68–3.71 4
40–65 80 0.36 1.10 1.10 3.33 4.56 3.41–6.10 6
Total 265 0.40 0.97 1.002 2.50 4.20 3.10–5.00 5e
c
Rare fish consumption 447 0.40 0.86 0.916 2.40 3.60 2.70–4.20 4
Total Hg 593 0.40 0.94 0.993 2.66 4.10 3.50–4.40 4e

Notes: GM, geometric mean; 95%CI (P95), 95% confidence interval of the 95th percentile; RV, reference value.
a
Individuals 18–39 years of age.
b
Individuals 40–65 years of age.
c
Fish consumption ≤3 times per month.
d
Statistical parameter below limit of quantification.
e
This RV should not be used if the sex and age of the person are known.
248 R. Kuno et al. / International Journal of Hygiene and Environmental Health 216 (2013) 243–249

Table 7
Reference values proposed for lead, cadmium and mercury in the blood (␮g/L) of several adult populations.

Local and year Population group Lead Cadmium Mercury Reference

Several countries Do not eat fish – – 2 ± 1.8 a

Nordberg et al. (1992)
After 1976 Low fish consumption 5 ± 3.8a
Medium fish consumption 7.5 ± 4.8a
High fish consumption 42.7 ± 26.7a
No information about fish 5 ± 2.8a
consumption
Czech Republic Male 95b – – Kliment (2000)
1996–1999 Female 80b – –
Total – 1.2b 2.8b
2001–2003 18–59 years old Batáriová et al. (2006)
Male 80b 3.5b 3.1b Lustigová and Puklová (2006)
Female 65b 3b 4b
Total 75b 3b 3.5b Puklová (2008)
Smoker 80b 4.5b 3.6b
Nonsmoker 75b 1.1b 3.4b

2005 Male 35c – 0.91c

Female 27c – 1.16c
Total – – –
Smoker – 1.3c –
Nonsmoker – 0.5c –

2007 Male 33c – 0.8c

Female 24c – 0.9c
Total – – –
Smoker – 1c –
Nonsmoker – 0.3c –
Germany 18–69 years old Schulz et al. (2007a, 2011) and
Wilhelm et al. (2004)
d
1997/1999 Male 90 – –
Female 70d – –
Total – – –
Nonsmoker – 1d –
Eat fish ≤3× month – – 2d
Brazil – Londrina Male 12–140.4e – – Paoliello et al. (2001)
1994–1996 56f
Female 12–134.9e – –
54f
Total 12–137.2e – –
55f
USA – NHANES Male 60 (54–65)g 1.3 (1.2–1.5)g – CDC (2005)
1999–2000 Female 40 (37–42)g 1.3 (1.1–1.4)g –
Total 52 (48–55)g 1.5 (1.4–1.6)g –
2001–2002 Female/16–49 years old – – 7.1 (5.3–11.3)g
Male 53 (50–55)g 1.4 (1.2–1.8)g –
Female 36 (30–38)g 1.4 (1.2–1.6)g –
Total 46 (42–49)g 1.6 (1.3–1.8)g –
Female/16–49 years old – – 4.6 (3.7–5.9)g
Italy – Brescia Male – 0.64 ± 0.5a Apostoli and Alessio (1991)
Female – 0.43 ± 0.36a
Italy – seven areas Male 45.1 ± 27.4a – – Apostoli et al. (2002)
2000 100.5h
Female 30.6 ± 16.7a – –
61h
Belgium – Flandres 50–65 years old 39.6 (38.4–40.9)i 0.42 (0.4–0.44)i – Schroijen et al. (2008)
2002–2006
a
Mean ± SD.
b
95%CI of P95.
c
Median.
d
Upper limit of 95%CI of P95.
e
Mean ± 2SD.
f
Geometric mean.
g
P95 (95%CI).
h
P95.
i
Geometric mean (95%CI).

mercury in fish should be prioritized because some individuals had
concentrations above the limit value (HBM I) of 5 ␮g/L.
Conclusion
This study derived RVs for PbB, CdB and HgB in Brazil. The
background levels of exposure to these metals were assessed in
a population in one of the most industrialized regions in the
country. Our background levels were compared to other popula-
tions worldwide.
Despite the low consumption of fish among this popula-
tion, it is possible that fish in the market at MASP are more
contaminated with Hg. This metal is a very persistent and
biocumulative contaminant, even released in the past to envi-
ronment it can remain in aquatic systems and enter into food
chain.
 R. Kuno et al. / International Journal of Hygiene and Environmental Health 216 (2013) 243–249
The low levels of lead should be result of the environmental
control actions implemented in the last decades to avoid the release
of lead in the State of Sao Paulo.
Biomonitoring studies should be performed continuously to
enable an assessment of time trends and update RVs to estimate
the exposure. Further studies in Brazil should be conducted to pro-
pose RVs for children and women of childbearing age because these
population subgroups are highly susceptible to the effects of metals.
Acknowledgements
We thank the staff of Colsan – Associação Beneficente de Coleta
de Sangue and Setor de Análises Toxicológicas – Cetesb.
References
Apostoli, P., Alessio, L., 1991. Identification of reference values for metals in the
general population. Rev. Bras. Toxicol. 4, 7–10.
Apostoli, P., Baj, A., Bavazzano, P., Ganzi, A., Neri, G., Ronchi, A., Soleo, L., Di, L.L.,
Spinelli, P., Valente, T., Minoia, C., 2002. Blood lead reference values: the results
of an Italian polycentric study. Sci. Total Environ. 287, 1–11.
Batáriová, A., Spĕváčková, V., Beneš, B., Čejchanová, M., Šmíd, J., Černá, M., 2006.
Blood and urine levels of Pb Cd, and Hg in the general population of the
Czech Republic and proposed reference values. Int. J. Hyg. Environ. Health 209,
359–366.
Becker, K., Kaus, S., Krause, Ch., Lepom, P., Schulz, Ch., Seiwert, M., Seifert, B., 2002.
German Environmental Survey 1998 (GerES III): environmental pollutants in
blood of the German population. Int. J. Hyg. Environ. Health 205, 297–308.
Braga, M., 1992. Non-parametric age related centile curves for blood lead levels in
blood donors from Northern Italy. Sci. Total Environ. 120, 173–184.
Buchancová, J., Knizková, M., Hýllová, D., Vrlík, M., Mesko, D., Klimentová, G.,
Gáliková, E., 1994. Content of the selected trace elements (Al, As, Cd, Cu, Fe,
Hg, Mn, Ni, Pb Zn) in blood, urine and hair of blood donors without occupational
exposure to these metals. Cent. Eur. J. Public Health 2, 82–87.
Centers for Disease Control and Prevention (CDC), 2005. Third National Report on
Human Exposure to Environmental Chemicals. Centers for Disease and Preven-
tion, Atlanta, GA, Available at: http://www.cdc.gov/exposurereport.
Centers for Disease Control and Prevention (CDC), 2009. Fourth National Report on
Human Exposure to Environmental Chemicals. Centers for Disease and Preven-
tion, Atlanta, GA, Available at: http://www.cdc.gov/exposurereport/.
Centers for Disease Control and Prevention (CDC), 2010. National Health and Nutri-
tion Examination Survey. Centers for Disease and Prevention, Atlanta, GA,
Available at: http://www.cdc.gov/nchs/nhanes.htm.
Černá, M., Spĕvácková, V., Batáriová, A., Šmíd, J., Čejchanová, M., Ocadlíková, D.,
Bavorová, H., Beneš, B., Kubínová, R., 2007. Human biomonitoring system in the
Czech Republic. Int. J. Hyg. Environ. Health 210, 495–499.
Companhia Ambiental do Estado de São Paulo (CETESB), 2004. Relatório de qual-
idade de aguas subterrâneas do estado de São Paulo 2001–2003. CETESB,
Sao Paulo, Brazil, http://www.cetesb.sp.gov.br/agua/aguas-subterraneas/111-
publicacoes-e-relatorios (in portuguese).
Companhia Ambiental do Estado de São Paulo (CETESB), 2006. Relatório de qual-
idade de águas interiores do estado de São Paulo 2005. CETESB, Sao Paulo,
Brazil, http://www.cetesb.sp.gov.br/agua/aguas-superficiais/35-publicacoes-/-
relatorios (in portuguese).
Companhia Ambiental do Estado de São Paulo (CETESB), 2007. Relatório
de qualidade de águas interiores do Estado de São Paulo 2006. Com-
panhia Ambiental do Estado de São Paulo, São Paulo, Available at:
http://www.cetesb.sp.gov.br/agua/aguas-superficiais/35-publicacoes-/-
relatorios (in portuguese).
Companhia Ambiental do Estado de São Paulo (CETESB), 2007b. Relatório de qual-
idade de águas subterrâneas no estado de São Paulo 2004–2006. CETESB,
Sao Paulo, Brazil, http://www.cetesb.sp.gov.br/agua/aguas-subterraneas/111-
publicacoes-e-relatorios (in portuguese).
Companhia Ambiental do Estado de São Paulo (CETESB), 2009. Relação de
áreas contaminadas – Novembro de 2009. CETESB, Sao Paulo, Brazil,
http://www.cetesb.sp.gov.br/areas-contaminadas/rela%E7%F5es-de-
%E1%A1reas-contaminadas/4- (in portuguese).
Dantas, F., 2007. Desempenho de SP desde 1995 é muito pior que
o do país. O Estado de S. Paulo, 25 set. 2007, Sao Paulo, Brazil,
http://www.estadao.com.br/noticias/impresso,desempenho-de-sp-desde-
1995-e-muito-pior-que-o-do-pais,56522,0.htm (in portuguese).
Ducoffre, G., Claeys, F., Bruaux, P., 1990. Lowering time trend of blood lead levels in
Belgium since 1978. Environ. Res. 51, 25–34.
249
Gundacker, C., Komarnicki, G., Zödl, B., Forster, C., Schuster, E., Wittmann, K., 2006.
Whole blood mercury and selenium concentrations in a selected Austrian pop-
ulation: does gender matter? Sci. Total Environ. 372, 76–86.
Kliment, V., 2000. System of Monitoring the Environmental Impact on Pop-
ulation Health of the Czech Republic: Summary Report 1999. National
Institute of Public Health, Prague, http://www.szu.cz/topics/environmental-
health/environmental-health-monitoring.
Kummrow, F., Silva, F.F., Kuno, R., Souza, A.L., Oliveira, P.V., 2008. Biomonitoring
method for the simultaneous determination of cadmium and lead in whole blood
by electrothermal atomic absorption spectrometry for assessment of environ-
mental exposure. Talanta 75, 246–252.
Kuno, R., Roquetti-Humaytá, M.H., Kummrow, F., Kitano, G.M., Silva, F.F., Yanagi,
Y., 2007. Factors related to blood lead levels in a group of individuals without
occupational exposure. In: XIth International Congress of Toxicology (ICT XI),
Montreal, Canada, Abstracts. . . 1 CD ROM.
L’Abbate, N., Cassano, F., Rana, F., Gagliardi, T., Giacomantonio, A., 1991. Blood lead
and erythrocyte protoporphyrin levels in the general population of an area in
Southern Italy. Med. Lav. 82, 336–340.
Lemos, M.M.G., Modesto, R.P., Ruby, E.C., Toffoli, F.F., Souza, J.B., Barbour,
E.D.A., Freitas, E.A., Silva, C.M., Oliveira, D.E.C., Casarini, D.C.P., 2008.
Valores da condição de qualidade dos solos da Bacia Hidrográfica do
Alto Tietê (UGRHI 6), Região Metropolitana de São Paulo. ABAS, Brazil,
http://ojs.c3sl.ufpr.br/ojs2/index.php/asubterraneas/article/view/23659 (in
portuguese).
Lustigová, M., Puklová, V. (Eds.), 2006. Environmental Health Monitoring System in
the Czech Republic. Summary Report – 2005. National Institute of Public Health,
Prague http://www.szu.cz/topics/environmental-health/environmental-
health-monitoring.
McKelvey, W., Gwynn, R.C., Jeffery, N., Kass, D., Thorpe, L.E., Garg, R.K., Palmer, C.D.,
Parsons, P.J., 2007. A biomonitoring study of lead, cadmium, and mercury in the
blood of New York City adults. Environ. Health Perspect. 115, 1435–1441.
Nordberg, G., Brune, D., Gerhardsson, L., Grandjean, P., Vesterberg, O., Wester, P.O.,
1992. The ICOH and IUPAC international programme for establishing reference
values of metals. Sci. Total Environ. 120, 17–21.
Oskarsson, A., Schültz, A., Skerfving, S., Hallén, I.P., Ohlin, B., Lagerkvist, B.J., 1996.
Total and inorganic mercury in breast milk in relation to fish consumption and
amalgam in lactating women. Arch. Environ. Health 51, 234–241.
Paoliello, M.M.B., Gutierrez, P.R., Turini, C.A., Matsuo, T., Mezzaroba, L., Barbosa,
D.S., Carvalho, S.R.Q., Alvarenga, A.L.P., Rezende, M.I., Figueiroa, G.A., Leite,
V.G.M., Gutierrez, A.C., Lobo, C.R., Cascales, R.A., 2001. Valores de referência
para plumbemia em uma população urbana do Sul do Brasil. Rev. Panam. Salud
Publica 9, 315–319 (in portuguese).
Paoliello, M.M.B., De Capitani, E.M., 2007. Occupational and environmental human
lead exposure in Brazil. Environ. Res. 103, 288–297.
Puklová, V. (Ed.), 2008. Environmental Health Monitoring System in the Czech
Republic. Summary Report – 2007. National Institute of Public Health, Prague
http://www.szu.cz/topics/environmental-health/environmental-health-
monitoring.
Sistema Estadual de Análise de Dados (SEADE), 2006. Projeção da população resi-
dente em 1◦ de julho: Região Metropolitana de São Paulo – 2006. SEADE, São
Paulo, Brazil, http://www.seade.gov.br/produtos/projpop/index.php (in por-
tuguese).
Schaller, K.H., 1988. Mercury. In: Angerer, J., Schaller, K.H. (Eds.), Analyses of Haz-
ardous Substances in Biological Materials: Methods for Biological Monitoring,
2. VCH-Verlagsgesellschaft, Weinheim, pp. 195–211.
Schroijen, C., Baeyens, W., Schoeters, G., Den Hond, E., Koppen, G., Bruckers, L.,
Nelen, V., Van De Mieroop, E., Bilau, M., Covaci, A., Keune, H., Loots, I., Kleinjans,
J., Dhooge, W., Van Larebeke, N., 2008. Internal exposure to pollutants mea-
sured in blood and urine of Flemish adolescents in function of area of residence.
Chemosphere 71, 1317–1325.
Schulz, C., Angerer, J., Ewers, U., Kolossa-Gehring, M., 2007a. The German human
biomonitoring commission. Int. J. Hyg. Environ. Health 210, 373–382.
Schulz, C., Conrad, A., Becker, K., Kolossa-Gehring, M., Seiwert, M., Seifert, B., 2007b.
Twenty years of the German Environmental Survey (GerES): human biomon-
itoring – temporal and spatial (West Germany/East Germany) differences in
population exposure. Int. J. Environ. Health 210, 271–297.
Schulz, C., Wilhelm, M., Heudorf, U., Kolossa-Gehring, M., 2011. Update of the
reference and HBM values derived by the German Human Biomonitoring Com-
mission. Int. J. Hyg. Environ. Health 215, 26–35.
Torra, M., Rodamilans, M., Montero, F., Farré, C., Corbella, J., 1997. Estudio de la
exposición al plomo en la población de Barcelona: evolución cronológica entre
1984 y 1995. Med. Clin. (Barcelona) 108, 601–603.
Umweltbundesamt (UBA), 2010. Health and Environmental Hygiene:
German Environmental Survey (GerES). Umweltbundesamt, Berlin,
http://www.uba.de/gesundheit-e/survey/index.htm.
Wilhelm, M., Ewers, U., Schulz, Ch., 2004. Revised and new reference values for some
trace elements in blood and urine for human biomonitoring in environmental
medicine. Int. J. Hyg. Environ. Health 207, 69–73.