Environmental Research Section A 84, 100}107 (2000)
doi:10.1006/enrs.2000.4088, available online at http://www.idealibrary.com on
Mercury Exposures in Riverside Amazon Communities in ParaH , Brazil1
Elisabeth C. de Oliveira Santos,* Iracina Maura de Jesus,* Edilson da Silva Brabo,*
Edvaldo C. Brito Loureiro,* Artur F. da Silva Mascarenhas,- Judith Weirich,*
Volney de M. Ca9 mara,? and David ClearyA
*MS/FUNASA/Instituto Evandro Chagas-IEC, Servimo de Ecologia Humana e Meio Ambiente, 66090 BeleH m, ParaH , Brazil;
-Secretaria Executiva de InduH stria, ComeH rcio e Mineram a o do ParaH -SEICOM. ?Universidade Federal do Rio de Janeiro-UFRJ,
NuH cleo de Estudos de SauH de Coletiva, and AEuropean Union
Received April 13, 1999
problems are emerging. In addition, this process has
degraded the regional environment and increased
exposures to toxic pollutants. One of the most signi7cant of these pollutants is mercury because of its use
in artisanal or small-scale gold mining processes in
Amazo9 nia (Ca9 mara and Corey, 1992).
Mercury pollution from gold mining is an old concern. As early as the 18th century, gold was known
to exist in the TapajoH s River, in what is now the state
of ParaH . In 1901, the Bulletin of the ‘‘Museu Paraense de HistoH ria Natural e Ethnogra7a’’ published
a report of a geological exploration in the Monte
Alegre region and the Tropas River, a tributary of
the Tapajos River. In 1958, mining activity in that
area was well underway. Mining activity accelerated
in 1962 when the 7rst airstrip was built, providing
easy transport of more sophisticated equipment and
more rapid inmigration of miners (Rodrigues et al.,
1994). In the 1970’s, with the inception of the
Plan of National Integration and the construction
of the Transamazo9 nica and CuiabaH }SantareH m
highways, mining activities proliferated to the point
of arousing the attention of federal and state
authorities.
According to estimates by FeH lix (1987) and the
‘‘Secretaria Estadual de InduH stria, ComeH rcio e Mineraia o do ParaH (SEICOM),’’ the production of gold in
Brazil was 9.6 tons in 1972 and increased steadily to
111 tons per year by the late 1980’s (SEICOM, 1995;
DNPM/SEICOM, 1998). Since then, mining activity
in the TapajoH s River has decreased signi7cantly,
largely due to the depletion of readily extractable
alluvial gold deposits and decreases in the price of
gold on the international market.
The quantity of mercury used in mining has been
and can be inferred from the amount of gold extracted from this region since it is estimated that approximately 1.3 to 2.0 kg of mercury is used for the
Cross-sectional studies were conducted in three
riverside communities in the state of ParaH in the
Amazon Region of Brazil. Two of these communities, Bras1H lia Legal and Sa o Lu1H s do TapajoH s, are
located in a watershed exposed to mercury pollution from artisanal gold mining, and the third, Santana do Ituqui, is outside this area. The studies were
based upon a community census and included all
persons volunteering for participation in each community. All participants were surveyed by questionnaire for health history, occupation, residence, and
diet. Clinical evaluations were also done on each
subject, along with collection of blood, urine, feces,
and hair. Fish samples were collected to represent
the most frequently consumed species. Mercury was
measured in hair and Ash. Although no signs or
symptoms of overt mercury intoxication were observed, persons in Bras1H lia Legal and Sa o Lu1H s do
TapajoH s had higher levels of mercury in hair than
residents of Santana de Ituqui, located out of the
risk area. Levels of mercury in Ash were below Brazilian health guidance limits, but the high rates of Ash
consumption among these populations raise concerns for the possible effects of chronic exposure,
especially among young children and women of
childbearing age. 2000 Academic Press
Key Words: mercury; epidemiology; gold mining;
Ash; Amazon.
INTRODUCTION
Unregulated human settlement and development
in the Amazon region of Brazil over the past 30 years
have destroyed large areas of the forest and changed
the natural state of the environment. As a result,
endemic diseases are expanding and new health
1
This paper was presented at Mercury as a Global Pollutant: 5th
International Conference, Rio de Janeiro, Brazil, May 23}28, 1999.
100
0013-9351/00 $35.00
Copyright 2000 by Academic Press
All rights of reproduction in any form reserved.
HUMAN HEALTH IN AN AREA OF MERCURY POLLUTION
extraction of 1 kg of gold (Rodrigues et al., 1994).
Mercury from mining adds signi7cantly to the natural amounts present in tropical forest soils, which
can be released by deforestation. The naturally occurring levels and distribution of mercury in the
Amazon region is not known since no studies were
conducted in the TapajoH s region before gold mining
began (Veiga, 1994). The impact of mercury on
human health depends on its form (organic vs inorganic), mode of exposure, and extent of biomagni7cation through the food chain (Brabo, 1992; Akagi
et al., 1995a; Malm et al., 1995; Kehrig et al., 1998;
Boischio et al., 1998).
Investigating the health impacts of mercury exposure in the Amazon requires a clear understanding
of general conditions in the area, including knowledge of endemic diseases, since there can be clinical
overlaps between the symptoms of mercury exposure and certain infectious diseases (Santos, 1993;
Santos et al., 1995). This study is part of an ongoing
evaluation of public health in different regions of the
Amazon to understand patterns of exposure and to
determine risks of intoxication by mercury, including when necessary, differentiating these effects
from other health conditions that may be present in
the exposed populations of the region.
MATERIALS AND METHODS
Area and Population
Between 1994 and 1997 the Instituto Evandro
Chagas (IEC) developed, in the State of ParaH , a program of research in several riverside communities,
including BrasmH lia Legal, Sa o LumH s do TapajoH s, and
Santana de Ituqui. BrasmH lia Legal and Sa o LumH s do
TapajoH s are located in the basin of the TapajoH s River,
a major tributary of the Amazon and a watershed
known to be at risk of contamination by mercury
originating from gold mining activity in the upper
region of the watershed. They belong to the municipal districts of Aveiro and Itaituba, respectively.
Santana do Ituqui, in contrast, is located on the
Amazonas River in the municipal district of SantareH m outside of this risk area (Fig. 1).
All three communities share the same tropical
climate and depend economically on subsistence agriculture and 7shing enterprises. The populations
are of mixed ethnicity, comprising caboclos (descended from European and indigenous groups). Fish and
cassava 8our constitute the primary food source in
all three communities. Drinking water is taken from
wells and the river without treatment; sanitary
wastes are discharged into the water or buried in
cesspools near habitated areas.
101
BrasmH lia Legal is located in the southwest region of
the State of ParaH on the western bank of the TapajoH s
River, 825 km away from BeleH m, the capital of the
State of ParaH . The only means of transportation to
and from this community is by water. Sa o LumH s do
TapajoH s is located in southwestern ParaH on the eastern bank of the TapajoH s River, 912 km away from
BeleH m. It can be reached by road (public buses) or
waterway. However, frequent 8ooding makes
ground transportation less reliable. Most of the
population consumes untreated water taken from
6 to 10 m below ground. Santana de Ituqui, in the
lower Amazon, is located in the municipal district of
SantareH m in the west of ParaH . It is on the southern
bank of the Amazonas River below its con8uence
with the TapajoH s River, 625 km from the city of
BeleH m. It can be reached by either the SantareH m}
Curuauna highway, via a local highway (in poor
repair), or by water. Drinking water comes from four
wells with depths varying from 6 to 27 m. The river
is used for bathing and domestic purposes. Human
waste is deposited in dug cesspools and garbage is
burned or dumped.
Research Design
The study utilized clinical, epidemiological, and
laboratory methods to evaluate exposures and
health status among populations at risk for mercury
exposure, a toxic element with cumulative toxicity.
Our goal was to determine levels of mercury in the
population, to investigate 7sh as a source of exposure, and to understand the nature of exposure to
individuals based on dietary habits.
All age groups were studied with special emphasis
on children under 14 years and women of childbearing age, which together constituted approximately
50% of the total population in each community.
These two groups were oversampled because of the
hazards of mercury exposure for fetal development.
The selection of methods was based upon knowledge
of the conditions in which these communities live. In
addition to the research, free medical care was provided to those identi7ed during the examinations.
Study Procedures
The community-based studies began with a community census, conducted by visit to each site 30
days before the full study. In every community, the
study began with an invitation to all residents to
participate followed by registration and assurance of
informed consent from all volunteering subjects.
A questionnaire was utilized to collect information
102
SANTOS ET AL.
FIG. 1. Map of study area.
HUMAN HEALTH IN AN AREA OF MERCURY POLLUTION
on life conditions, residential history, occupation,
dietary habits, and health history. Assessment of
individual health status was done by examination,
including a physical exam to assess speech and
sight, static and dynamic equilibrium, motor coordination, tactile sensitivity, muscle tone, and posture.
Biological specimens were collected from participants, including urine (upon clinical indication),
hair (from all for measurement of mercury), blood
(from all subjects for studies of malaria prevalence
by thick-7lm diagnosis, hematocrit, selected blood
chemistries, and mercury levels), and feces (from all
for direct assessment of intestinal parasites). Fish
samples were collected from the same sites where
local 7shermen in each community harvested 7sh for
consumption.
Laboratory Procedures
Hair samples were collected from the same area of
the scalp in each person. Hair locks of at least 100
hairs were selected and cut 1 cm from the scalp, with
scissors made of nonoxidizable steel, and stored in
white paper envelopes at room temperature until
analysis (Campos and Piveta, 1993). At the laboratory, hair samples were washed 7rst with neutral
detergent and water and then with acetone. Samples
were dried in a desiccator under vacuum, transferred into a glass 8ask, and minced. Samples of 10 to
20 mg were utilized for analysis according to the
procedures of Akagi et al. (1995b, 1996). For the
analysis of mercury in 7sh samples, portions of edible tissue were removed from each specimen in the
7eld and frozen individually in plastic bags for later
analysis. At the laboratory, 0.5 g of tissue from each
sample was analyzed for total mercury levels by the
methods of Akagi et al. (1995), after acid digestion
using a mixture of nitric, perchloric, and sulfuric
acids. Mercury determinations in hair and 7sh were
made by 8ameless atomic absorption spectrophotometry using a mercury analyzer HG-3500.
RESULTS
Descriptive epidemiological and clinical information and the results of mercury analyses in the three
communities are presented here. The mercury levels
in hair were evaluated by age groups for each of the
populations. An analysis was also done in a smaller
group of individuals in the same populations for
whom residence time in each community was in the
same range of years as age. Overall, no signs or
symptoms indicative of mercury intoxication (dysfunctions of speech or vision, static and dynamic
103
equilibrium, motor coordination, tactile sensitivity,
muscle tone, or segmentary posture were observed).
No prevalent cases of malaria were seen.
BrasnH lia Legal
In BrasmH lia Legal, there were 541 inhabitants
living in 107 houses, according to our preliminary
census. From this population, 246 individuals were
interviewed, representing 46.0% of the local population. The composition of the sample contained 94
(38.2%) males and 152 (61.8%) females. Their ages
varied from 1 to 74 years old; we oversampled persons under 14 years old (59.0%) according to our
study interests in pediatric exposures. Women between 15 and 49 years old (59) were assumed to be
fertile. This constituted another priority of the research and so we oversampled this group as well
to include 59% of the women in this age group,
according to the community census. The men in the
community were very much involved in 7shing,
small-scale agriculture, and the collection of wood,
spending a great deal of time away from the settlement. The daily diet is composed of 8our, rice, and
7sh. Ox meat or game is consumed at the most
weekly. Seasonal fruits are frequently eaten, as are
vegetables (mainly tomatoes and onions), milk, and
eggs. Breast feeding is frequent.
The average mercury level in carnivorous 7sh consumed in BrasmH lia Legal was 0.529 lg/g (0.029 to
1.460 lg/g) and that for noncarnivorous 7sh was
0.106 lg/g (range 0.011 to 0.349 lg/g).
At the time of examination, 20 people (9.1%) had
diarrhea and 69 (31.1%) had multiple parasites in
feces. Of the 78 women older than 14 years old, 30
(38.4%) reported one or more spontaneous abortions
during their fertile life.
Sa o LunH s do TapajoH s
The preliminary census in Sa o LumH s do TapajoH s
recorded 509 inhabitants living in 97 houses. Three
hundred and thirty-three individuals (65.0%) participated in the study. The population sample consisted of 139 males (41.7%) and 194 females (58.3%).
Their ages ranged from 6 months to 98 years old,
with 73.3% younger than 24 years old. Eighty-three
women were between 15 and 49 years old and they
represented 85.6% of the women in that age group,
according to the community’s preliminary census.
The daily diet included 8our, rice, and 7sh. Ox meat
or game was eaten at most three times a month.
Seasonal fruits were eaten frequently, most vegetables rarely, tomato and onion a little more fre-
104
SANTOS ET AL.
quently, and milk and eggs less often. Breast feeding
constitutes a frequent habit of infant nutrition.
The average mercury level in carnivorous 7sh consumed by this population was 0.487 lg/g (0.094 to
1.647 lg/g) and that for noncarnivorous 7sh was
0.068 lg/g (range 0.007 to 0.190 lg/g).
At the time of examination, 13 people (4.0%) had
diarrhea and 280 (85.0%) had polyparasitism in
feces. Of the 95 women older than 14 years old, 32
(33.68%) reported one or more spontaneous abortions during their fertile life.
Santana de Ituqui
The preliminary census recorded 425 inhabitants
in Santana de Ituqui living in 64 houses. Three
hundred and twenty-six individuals (77.0%) participated. The population sample contained 153
males (46.9%) and 173 females (53.1%). The ages
ranged from 1 to 79 years old with 68.7% of the
population less than 24 years old. Seventy-two
women were between 15 and 49 years old, and they
represented 92.3% of all the women in that age
group. The daily diet consisted of 8our, rice,
and 7sh. Ox meat or chicken was eaten once a week,
game rarely, fruits of the season frequently, greens
and vegetables very rarely, and milk and eggs less
often.
The average mercury level in carnivorous 7sh consumed in this place was 0.124 lg/g (0.032 to 0.367
lg/g) and that for noncarnivorous 7sh was 0.052 lg/g
(range 0.010 to 0.186 lg/g). Breast feeding constitutes a frequent habit.
At the time of examination, 8 people (2.5%) had
diarrhea and 110 (34.3%) had polyparasitism in
feces. Of the 86 women over 14 years of age, 30
(34.88%) reported one or more spontaneous abortions during their fertile life.
Mercury Levels in Hair
Table 1 presents data on the levels of Hg in the
hair samples by age group and weekly consumption
of 7sh meals. In BrasmH lia Legal, hair samples taken
from 220 individuals had an average mercury of
11.75 lg/g, a standard deviation of 7.95, and a range
of 0.53 to 49.99 lg/g. The median values tended to
increase over age until 40 years old, after which they
tended to decrease. In this population, 124 individuals (56.4%) had mercury levels below 11.75 lg/g
whereas 96 individuals (43.6%) had levels above
11.75 lg/g. The frequency of 7sh consumption was
relatively low in this village, ranging from 8 to 14
times, with an average of 10 times per week.
TABLE 1
Bras1H lia LegalEMean Values of Mercury in Hair,
According to Age Group and Weekly Consumption of
Fish, Amazon, State of ParaH , Brazil, 1994
Age
group
n
Mean Hg
lg/g
SD
Range
lg/g
Fish
consumption
0}5
6}10
11}15
16}20
21}25
26}30
31}35
36}40
41}45
46}50
51}55
56}60
61}65
'65
Total
30
68
33
12
10
9
16
12
1
7
8
5
6
3
220
5.84
13.06
14.20
13.39
15.25
11.06
12.57
14.21
11.70
7.06
11.53
4.93
11.33
7.45
11.75
4.91
6.98
8.91
4.97
9.99
5.49
9.01
7.27
0
4.37
1.71
2.58
3.96
7.79
7.95
1.09}20.46
0.70}35.80
1.22}47.00
5.56}19.90
1.40}29.50
3.70}21.40
2.84}37.20
5.0}33.0
11.70
1.02}14.24
3.57}49.99
0.53}7.07
5.01}15.94
2.78}16.46
0.53}49.99
10
10
10
12
12
10
10
9
10
13
10
8
14
10
10
In Sa o LumH s do TapajoH s (Table 2) the average mercury level in hair samples of 327 individuals was
19.91 lg/g, the standard deviation was 11.96, and
the range was 0.10 to 94.50 lg/g. The median values
increased by age up to 15 years old. In older persons
the mercury levels tended to be lower. In this population, 183 (56.5%) had mercury values at or below
the average, whereas 43.5% (144) had levels of
TABLE 2
Sa o Luiz do TapajoH sEMean Values of Mercury in Hair,
According to Age Group and Weekly Consumption of Fish,
Amazon, State of ParaH , Brazil, 1995
Age
group
n
Mean Hg
lg/g
SD
Range
lg/g
Fish
consumption
0}5
6}10
11}15
16}20
21}25
26}30
31}35
36}40
41}45
46}50
51}55
56}60
61}65
'65
Total
75
74
51
21
21
15
15
15
14
8
4
6
3
5
327
21.06
22.10
23.24
19.11
15.68
15.34
18.98
14.31
15.13
21.71
15.40
17.13
12.60
12.60
19.91
14.38
12.38
12.38
9.14
7.42
9.35
9.84
7.75
8.90
13.27
6.55
6.92
13.21
7.77
11.96
0.10}94.50
2.40}52.50
3.90}61.80
2.10}33.60
1.73}32.0
3.90}34.90
5.10}38.0
2.60}27.80
3.20}33.60
4.0}47.10
5.90}20.60
7.90}27.60
3.8}27.80
3.20}20.80
0.10}94.50
13
14
14
14
14
14
14
12
14
14
14
14
13
12
13
105
HUMAN HEALTH IN AN AREA OF MERCURY POLLUTION
TABLE 3
Santana de ItuquiEMean Values of Mercury in Hair,
According to Age Group and Weekly Consumption of Fish,
Amazon, State of ParaH , Brazil, 1996
Age
group
n
Mean Hg
lg/g
SD
Range
lg/g
Fish
consumption
0}5
6}10
11}15
16}20
21}25
26}30
31}35
36}40
41}45
46}50
51}55
56}60
61}65
'65
Total
37
81
62
25
17
16
19
18
10
10
9
6
4
7
321
3.67
4.44
4.47
5.0
3.34
4.69
5.36
3.44
4.18
4.02
5.39
4.37
4.15
3.61
4.33
1.84
1.95
1.93
2.02
1.35
1.88
2.41
1.35
1.38
1.35
2.40
1.38
3.27
1.82
1.94
0.50}8.50
0.40}10.90
2.0}11.60
2.5}9.60
1.30}7.10
1.70}9.20
1.90}9.60
1.20}6.0
2.70}6.80
1.90}6.40
2.30}9.0
3.10}6.90
1.90}9.0
0.70}5.70
0.40}11.60
11
13
12
14
14
14
12
12
12
12
14
14
14
13
13
19.91 lg/g. In this population, the number of weekly
7sh meals varied from 12 to 14 times, with an average of 13 times.
In the hair samples from 321 individuals of Santana de Ituqui (Table 3) the average mercury level
was 4.33 lg/g, the standard deviation was 1.94, and
the range was 0.40 to 11.60 lg/g. The median values
did not vary by age (Fig. 2). In this population, 188
(60.0%) had mercury values at or below the average,
FIG. 3. Distribution of mercury in hair samples, when age
is equal to residence time, in three riverside communities of
Amazon, State of ParaH , Brazil, 1994}1996.
whereas 133 individuals (40.0%) had levels above
4.33 lg/g. The number of meals in which 7sh was
consumed varied from 11 to 14 times, with an average of 13 times.
An additional analysis was done with 152 individuals from BrasmH lia Legal, 256 from Sa o LumH s do
TapajoH s, and 167 from Santana de Ituqui to examine
the effect of residence time, taken as an indicator of
length of exposure (Fig. 3). The results of this analysis are by age group, since age and length of residence were within the same range for all persons in
these subsets (Tables 4}6).
In BrasmH lia Legal the average level of Hg increased
to 14.22 lg/g in the oldest age/longest resident group
and then decreased to 9.82 lg/g in the lowest
age/residence group. In this population 81 (53.3%)
TABLE 4
Bras1H lia LegalEMean Levels of Mercury in Hair, by Age
and Length of Residence, 1994
FIG. 2. Distribution of mercury in hair samples, according to
age group, in three riverside communities of Amazon, State of
ParaH , Brazil, 1994}1996.
Age
group
n
Mean Hg
lg/g
SD
Range
lg/g
Fish
consumption
0}5
6}10
11}15
16}25
26}35
36}55
Total
30
56
24
13
14
15
152
5.84
13.63
12.41
14.22
11.67
9.82
11.39
4.91
6.98
6.10
8.40
8.43
7.44
7.35
1.09}0.46
0.70}35.80
1.22}22.70
1.40}29.50
2.84}37.20
1.02}33.0
0.70}37.20
10
10
9
12
10
10
9.89
106
SANTOS ET AL.
TABLE 5
DISCUSSION AND CONCLUSIONS
Sa o Luiz do TapajoH sEMean Levels of Mercury in Hair,
By Age and Length of Residence, 1995
Age
group
n
Mean Hg
lg/g
SD
Range
lg/g
Fish
consumption
0}5
6}10
11}15
16}25
26}35
36}55
Total
75
67
47
28
21
18
256
21.06
22.36
23.90
18.51
18.29
14.69
20.96
14.38
12.43
12.21
7.84
9.41
8.39
12.29
0.10}94.50
2.40}52.50
3.90}61.80
4.40}33.40
5.10}34.90
2.60}27.80
0.10}94.50
13
14
14
14
14
13
13.35
had mercury values at or below the average, whereas 71 (46.7%) had levels above 11.39 lg/g. The number of weekly 7sh meals was the lowest among the
three populations, between 9 and 12 times, with an
average of 9.89 times.
In Sa o LumH s do TapajoH s, the mean concentration of
Hg, 20.96 lg/g, was slightly higher than the average
found for the total sample. In persons residing between zero and 5 years (similar age range), the average hair level was 21.06 lg/g, rising to 23.90 lg/g in
the age/residence group 11 to 15 years old, and decreasing thereafter. In the total population, 148
(57.88%) had hair mercury levels at or below the
average, whereas 108 (42.2%) had levels above
20.96 lg/g. The number of weekly 7sh meals was 13
to 14, with an average of 13.35.
In Santana de Ituqui, the average hair Hg level
was 4.28 lg/g, with little variation by age or time of
residence. In the total population, 95 individuals
(56.9%) had mercury values at or below average,
whereas 72 individuals (43.1%) had levels above
4.28 lg/g. The number of weekly 7sh meals
varied from 11 to 14 times, with an average of 12.5
times.
TABLE 6
Santana de ItuquiEMean Levels of Mercury in Hair,
by Age and Length of Residence, 1996
Age
group
n
Mean Hg
lg/g
SD
Range
lg/g
Fish
consumption
0}5
6}10
11}15
16}25
26}35
36}55
Total
37
55
26
25
12
12
167
3.67
4.62
4.20
4.30
5.12
3.95
4.28
1.84
1.87
1.53
1.94
2.42
1.13
1.85
0.50}8.50
1.60}10.90
2.20}8.80
1.30}8.00
1.70}9.20
2.30}6.30
0.50}10.90
11
13
12
14
14
12
12.53
Over the past 35 years the basic social and
economic infrastructure of the Amazon region has
remained underdeveloped. Due to accelerated population growth in this region, the problems of healthcare delivery have been aggravated (Santos et al.,
1992). The risks to health from mercury exposure
are not monitored by the health services, since
health professionals are not trained to distinguish
mercury toxicity in the context of the endemic diseases of the region (Santos et al., 1995).
In Brazil, it was not foreseen that artisanal mining activity, a relatively simple type of mineral extraction initiated almost 40 years ago that provides
employment for populations lacking many economic
opportunities, could become a signi7cant source of
pollution. The intensive and extensive use of mercury in the extraction of gold has impaired the quality of rivers and streams, polluting the 7sh, and in
this way threatening the health of resident populations. Increased concentrations of mercury can now
be measured in the aquatic biota of the TapajoH s
River and in many persons living in riverside communities in this ecosystem. Since 7sh are the main
source of protein for riverside communities, they are
a potential vehicle for exposure to mercury and especially methylmercury.
The diet of the communities in this study was very
limited and poor in vegetables. The weekly consumption of 7sh, measured by the number of reported 7sh
meals, is high. Although no obvious signs of protein
or caloric de7ciency were observed in these populations, possible micronutrient or vitamin de7ciencies
were not studied. Intake of vitamin A and selenium
may be particularly important to assess, since de7ciencies of vitamin A can cause vision problems that
may be confused with mercury intoxication, and
selenium is known to protect against mercury absorption and toxicity. Although no severe health
problems and no prevalent cases of infectious disease were found, the frequent diarrheal disease and
polyparasitism found in these populations could
interfere with the absorption or retention of methylmercury from the diet.
The levels of mercury in hair of the populations in
the communities known to be at risk of mercury
contamination from gold mining (BrasmH lia Legal and
Sa o LumH s do TapajoH s) are elevated and appear to
increase with age and length of residence in these
communities up until 25 years. A hypothesis to explain these results could be that mercury levels in
individuals younger than 25 years are higher than
in older persons because of differences in in utero
HUMAN HEALTH IN AN AREA OF MERCURY POLLUTION
exposure. People older than 30 years were born to
mothers not yet exposed to mercury (since gold mining began about 30 years ago), whereas, persons
born within the last 25 years were born with elevated mercury levels, as a consequence of the in utero
transfer of mercury from their mothers.
The pattern of mercury exposures in the population of Santana de Itaqui, located outside of the risk
area, is very different. Overall levels of mercury are
low, similar to those reported in unexposed persons
in North America, and there is little change with age
or length of residence.
During our clinical investigation no acute signs or
symptoms of mercury-related intoxication were
found in the examined individuals. Many persons in
this survey had hair mercury levels above guidelines
proposed to protect against developmental neurotoxicity (WHO, 1990). We recommended further
studies using tests capable of detecting signs of subclinical intoxication, as well as detailed investigation
of dietary patterns. Such studies have suggested an
association between mercury exposures and adverse
neurodevelopmental outcomes in the same region
(Grandjean et al., 1999). Further research is also
appropriate because of the high rates of 7sh consumption among these populations to obtain information on patterns of accumulation and elimination of
mercury and methylmercury in these at-risk populations.
ACKNOWLEDGMENTS
Financial support for this study was provided by MS/FNS/Instituto Evandro Chagas and European Union. This study was
conducted in accordance with national (National Council of
Health) and institutional (Scienti7c Ethics Council) guidelines for
the protection of human subjects.
REFERENCES
Akagi, H., Malm, O., Kinjo, Y., Harada, H., Branches, F. J. P.,
Pffeifer, W. C., and Kato, H. (1995a). Methylmercury pollution
in the Amazon. Sci. Total Environ. 5, 85}95.
Akagi, H., Malm, O., Branches, F. J. P., Kinjo, Y., Kashima, Y.,
Guimara es, J. R. D., Oliveira, R. B., Haraguchi, K., Pfeiffer,
W. C., Takizawa, Y., and Kato, H. (1995b). Human exposure to
mercury due to goldmine in the TapajoH s River Basin, Amazon,
Brazil: Speciation of mercury in human hair, blood, and urine.
Water Air Soil Pollut. 80, 85}94.
Akagi, H., Malm, O., and Branches, F. J. P. (1996). Human
exposure to mercury due to mining in the Amazon, Brazil
;a review. Environ. Sci. 3, 199}211.
Boischio, A. A., Henschel, D., and Barbosa, A. C. (1995). Mercury
exposure through 7sh consumption by the upper Madeira River
population Brazil;1991. Ecosyst. Health 1, 177}192.
107
Brabo, E. S. (1992). ‘‘Contaminaia o por MercuH rio dos Rios Crepori
e MarupaH ;Bacia do TapajoH s.’’ Dissertaia o de Mestrado,
BeleH m, Universidade Federal do ParaH .
Ca9 mara, V. M., and Corey, G. (1992). ‘‘Epidemiologia e Meio
Ambiente.’’ O Caso dos Garimpeiros de Ouro no Brasil,
Metepec, MeH xico, Centro Panamericano de Ecologia Humana
e SauH de/ECO/OPS.
Campos, R. C., and Pivetta, F. (1993). MeH todos de Coleta e AnaH lise
de Amostras de Sangue, Urina e Cabelo para Dosagem de
MercuH rio. In ‘‘MercuH rio em AD reas de Garimpos de Ouro, SeH rie
Vigila9 ncia 12’’ (Volney de Magalha es Ca9 mara, Eds.), pp. 65}80.
Metepec, MeH xico. Ed. ECO/OPS.
DNPM, Departamento Nacional da Produia o Mineral; SEICOM,
Secretaria de Estado de InduH stria, ComeH rcio e Mineraia o.
(1998). ‘‘SumaH rio mineral.’’ BeleH m, ParaH .
FeH lix, J. T. (1987). AnaH lise da Produia o de Ouro no Brasil. Brasil
Mineral 48, 90}96.
Grandjean, P., White, R. F., Nielsen, A., Cleary, D., and Santos,
E. O. (1999). Methylmercury neurotoxicity in Amazonian children downstream from gold mining. Environ. Health Perspect.
107, 587}592.
Kehrig, H. A., Malm, O., Akagi, A., Guimara es, J. R. D., and
Torres, J. P. M. (1998). Methylmercury in 7sh and hair samples
form the Balbina Reservoir, Brazilian Amazon. Environ. Res.
77, 84}90.
Malm, O., Branches, F. J. P., Akagi, H., Castro, M. B., Pfeiffer,
W. C., Harada, M., Bastos, W. R., and Kato, H. (1995).
Mercury and methylmercury in 7sh and human hair from
the TapajoH s River Basin, Brazil. Sci. Total Environ. 175,
141}145.
Rodrigues, R. M., Mascarenhas, A. F. S., Ichiara, A. H., Souza,
T. M. C., Bidone, E. D., Bellia, V., Hacon, S., Silva, A. R. B.,
Braga, J. B., and Filho, B. S. (1994). ‘‘Estudo dos Impactos
Ambientais Decorrentes do Extrativismo Mineral e Poluiia o
Mercurial no TapajoH s-PreH -DiagnoH stico.’’ Rio de Janeiro, SeH rie
Tecnologia Ambiental, Ed, CETEM/CNPq.
Santos, E. O. (1993). Garimpagem e SauH de. In ‘‘Conseque9 ncias da
Garimpagem no A: mbito Social e Ambiental da Amazo9 nia’’
(A. Mathis and R. Rehaag, Eds.), pp. 36}37. Ed. Cejup,
BeleH m.
Santos, E. O., Rosa, J. F. T., Jesus, I. M., and Loureiro,
E. C. B. (1992). A SauH de das Populaio es da Amazo9 nia
Brasileira. In ‘‘Enfoque Integral de la Salud Humana en la
Amazonia.’’ (L. Yarzabal, C. Espinal, and L. E. Aragon, (Eds.),
pp. 95}156. Associaia o de Universidades Amazo9 nicas/Universidad Central de Venezuela, Imprenta UCV, Caracas.
Santos, E. O., Loureiro, E. C. B., Jesus, I. M., Brabo, E. S., Silva,
R. S. U., Soares, M. C. P., Ca9 mara, V. M., Souza, M. R., and
Branches, F. (1995). DiagnoH stico das Condiio es de SauH de de
uma Comunidade Garimpeira na Regia o do TapajoH s, Itaituaba,
ParaH -Brasil (1992). Cadern. Sau& de Pu& blica 11, 212}225.
SEICOM, Secretaria de Estado de InduH stria, ComeH rcio e Mineraia o. (1995). SumaH rio mineral. BeleH m, ParaH .
Veiga, M. M., Meech, J. A., and Onate, N. (1994). Deforestation:
A major source of mercury pollution in the Amazon. Nature 368,
816}817.
World Health Organization (WHO) (1990). Methylmercury. IPCS
Environmental Health Criteria Document 101, Geneva,
WHO.