The mercury burden of the Czech population: An integrated approach

International Journal of Hygiene and Environmental Health 213 (2010) 243–251

Contents lists available at ScienceDirect

International Journal of Hygiene and Environmental Health journal homepage: www.elsevier.de/ijheh

The mercury burden of the Czech population: An integrated approach a,b a ˇ ˇ ˇ uˇ ˚ rková a , Vladimíra Puklová a,∗ , Andrea Krsková a , Milena Cerná , Mája Cejchanová , Irena Reh a a a a ˚ zena Kubínová , Magdaléna Zimová Jiˇrí Ruprich , Karel Kratzer , Ruˇ a b

National Institute of Public Health, Prague, the Czech Republic Charles University in Prague, Third Faculty of Medicine, the Czech Republic

a r t i c l e

i n f o

Article history: Received 25 May 2009 Received in revised form 1 February 2010 Accepted 7 February 2010 Keywords: Total mercury Monitoring Environmental exposure Dietary intake Biomonitoring

a b s t r a c t In this paper an integrated approach in assessment of the population exposure from various sources of total mercury (THg) oral intake in the Czech Republic is presented. The information on total mercury levels in diet, drinking water, surface urban soil and body fluids and tissues stem from the Czech national Environmental Health Monitoring System (EHMS) operated since 1994. The THg concentration was determined by the special atomic absorption spectrophotometer AMA 254. The data on THg content in food from the sales network were collected in 12 cities. The estimated average dietary intake representing more than 95% of weight of usual diet composition ranged 1–2% of the JECFA/FAO WHO provisional tolerable weekly intake (PTWI) value for total mercury (5 ␮g/kg b.w./week). Data on drinking water quality stem from the nationwide monitoring database. The content of THg in drinking water is generally low; only 0.2% of the Czech population supplied with drinking water from the distribution networks (total of 92% of the population) has a mercury intake from drinking water higher than 1% PTWI and not exceeding 5% PTWI. The estimation of potential mercury intake by unintentional consumption of soil in small children was based on THg content in surface soil of a total of 324 nursery schools in 24 cities and towns. Median value was 0.16 mg/kg. Human biomonitoring was performed in 9 Czech cities. In 2007, the mercury median values in blood of adults (N = 412) were 0.85 and 0.89 ␮g/l in males and in females, respectively; urine median value in adults was 1.10 ␮g/g creatinine. In 2008, the blood median value in children (N = 324) amounted to 0.35 ␮g/l; urine median value is 0.16 ␮g/g creatinine. In children’s hair the median THg value was 0.18 ␮g/g. The correlation between fish consumption and blood THg levels was observed in both adults and children. Also the biomonitoring outputs did not reveal a substantial burden of the population. © 2010 Elsevier GmbH. All rights reserved.

Introduction Population exposure to mercury has been an important issue in view of public health. Mercury is a toxic persistent pollutant; it has been widely spread in the environment. Elemental or inorganic mercury when released into the air or water can be methylated by the microorganisms in the soils or water sediments. Mercury accumulates in animal tissues and increases in concentration through the food chain (UNEP, 2002). Dietary intake is the most important mercury pathway for the general population. A predominant source in diet is fish and seafood products with most of mercury being in the form of highly absorbable methylmercury (ATSDR, 1999). A number of studies were published on dietary exposure to mercury namely from fish consumption (e.g. Mahaffey, 2004;

∗ Corresponding author at: National Institute of Public Health, Department of ˇ Environmental Health, Srobárova 48, Prague 10, 100 42, Czech Republic. Tel.: +420 26708 2602; fax: +420 26708 2924. E-mail address: [email protected] (V. Puklová). 1438-4639/$ – see front matter © 2010 Elsevier GmbH. All rights reserved. doi:10.1016/j.ijheh.2010.02.002

Crépet et al., 2005), but also using Total Diet Studies (Ysart et al., ˜ 2000; Larsen et al., 2002; Munoz et al., 2005). In drinking water almost all mercury is assumed to be in inorganic form of Hg2+ and is absorbed in only small fraction. Risk of intake of organic mercury compounds from drinking water is considered to be improbable (WHO, 2005). A potential source of mercury was previously also thiomersal used as a vaccine preservative. Though the health risk from this exposure was considered low (WHO, 2007), thiomersal was gradually shifted off namely in child vaccines either without substitution or substituted with other preservative medium (NCIRS, 2009). The use of mercury-containing creams and other cosmetic products could be a source of dermal exposure (WHO, 2007), nevertheless, the marketing of mercury-containing cosmetic products has not been permitted within the European Union since 1976. Dental amalgam containing inorganic mercury compounds can be a substantial source of mercury through vapour inhalation and ingestion of small particles. Of the possible sources, fish and seafood containing methylmercury as well as dental amalgam are nowadays thought to be generally the most significant sources of mercury for the non-professionally exposed population,

244

V. Puklová et al. / International Journal of Hygiene and Environmental Health 213 (2010) 243–251

Fig. 1. Map of the sampling sites.

even adverse health effects attributable to amalgam-released mercury is still a matter of exploration (e.g. Zimmer et al., 2002; Mutter et al., 2004; Bellinger et al., 2006). Biomonitoring reflecting all exposure pathways enables to evaluate overall human burden (Becker et al., 2003; Gundacker et al., 2006; Caldwell et al., 2009). This article presents an integrated approach to characterize the population exposure to total mercury (THg) from the monitored exposure pathways which represent food, drinking water, and potentially a surface urban soil. This study continues in description of the overall exposure to environmental pollutants of the Czech population within the Environmental Health Monitoring System (EHMS) and follows-up the previous study on overall cadmium exposure (Puklová et al., 2005). EHMS is coordinated by the National Institute of Public Health (NIPH) in Prague. The system was established in 1991 by governmental resolution and has been run continually since 1994 (Kliment et al., 1997). It involves regular data collection on environmental pollution, population exposure estimates and health risk assessments. Human biomonitoring is an integral part of the EHMS including regular monitoring of selected xenobiotics in human body fluids and tissues. The summary reports have been issued annually and are available at the NIPH web sites www.szu.cz/topics/environmental-health/ environmental-health-monitoring. Methods Sampling The samplings of foodstuffs, urban soil and human body fluids and tissues have systematically been taken following the elaborated standard operational procedures for each medium. This means that the sampling has been carried out in a standard manner each year to ensure data quality and comparability of data series. Sampling of foodstuffs is performed by the regional public health institute’s trained professionals in the sales network (by purchasing the food in shops) in 12 urban areas, i.e. cities and towns of various sizes and characters (Fig. 1). The number of individual monitored food kinds (representing more than 95% of weight of usual diet composition) varied slightly within the monitoring program, ranging from 160 in 1994 to 205 in 2006/2007 per one area

and monitoring period. Samples have been taken in several terms over the year taking into consideration the seasonal character of the sale of certain foods. The food sampling was performed annually till 2003. Since 2004, the sampling and data processing interval has been lengthened from annual to a biennial period. The food samples have been mixed into 4 regional samples (3 urban areas in each region, Fig. 1) and then combined into so-called composite samples. For instance, in 2006/2007 altogether 880 composite samples were prepared out of 3696 individual samples. All samples have been collected and processed in the laboratories of the National Institute of Public Health. The principles of organisation of the monitoring were drawn from recommendations of the WHO (GEMS WHO, 1985). The national food consumption necessary for exposure estimates was determined by several studies: in 1991 (Ruprich et al., 1993) and 1994 (Ruprich et al., 1997) with the aid of a household budget method, and in 2004 by means of an individual food consumption study using a method of twice repeated 24-h recall (Ruprich et al., 2006). Data on drinking water quality have been obtained within the nationwide monitoring of public supply networks since 2004, after a ten-year period of follow-up of the water quality in the selected Czech cities. There are 9.5 mil inhabitants, i.e. 92% of the total population, supplied by controlled drinking water from public water networks. The sampling is carried out from the drinking water pipes intended for human use. A basic unit for assessing the quality of drinking water is the supply zone.1 In 2008, a total of 4020 supply zones were monitored. The majority of samplings were carried out by water network operators, the lesser part by supervising public health authorities. The sampling sites and frequency follow the methodology set in the national Decree No. 252/2004 which is fully harmonized with EU Council Directive 98/83/EC. The sampling

1 The supply zone defined by the national Decree No. 252/2004 is a zone including either several cadastral areas, one cadastral area or its part where a distribution system is located, supplying drinking water that originates from one or more sources and can be considered as approximately the same quality. Water in such a distribution system is supplied by a single water supply system operator or owner for public use.

V. Puklová et al. / International Journal of Hygiene and Environmental Health 213 (2010) 243–251

ˇ is carried out according to the technical norm series CSN EN ISO 5667. Samples have to be representative for drinking water quality throughout the year and for the whole relevant water network. More than 50% of the sampling sites have to be changed each year. The changing sites have been chosen by a random sample method to ensure that no site is excluded from the chance of sampling. The content of mercury in surface soil was determined for the purpose of estimating the potential burden of preschool children caused by unintentional soil consumption. The samples were obtained in playgrounds (except sandpits) in a total of 324 nursery schools in 24 cities and smaller towns differing in size and environmental pollution level during the monitoring period 2002–2006. The cities were chosen so as to cover evenly the country area (Fig. 1). There were monitored all nursery schools in the participated cities and towns. Soil samples were taken from a depth of 0–10 cm at five sampling points in each nursery playground. The sampling was carried out according to the Methods for sampling the soil of free playgrounds, elaborated by NIPH (2000). Chemical analyses were carried out after homogenizing a mixed sample of each playground, so that a total of 324 samples were analysed. In order to assess the level of internal exposure of the general population from the environment, mercury levels were monitored in the whole blood and urine of adults and children, with additional hair analyses in children only. In 1996–2003, monitoring was performed in four urban areas (Fig. 1), selected so as to represent both common provincial towns and industrially polluted cities. Since 2005, another heterogenous set of urban areas has been chosen, including the capital city of Prague, the highly polluted industrial city of Ostrava, medium industrial city of Liberec and two smaller provincial towns Kromˇerˇíˇz and Uherské Hradiˇstˇe (Fig. 1). Adult blood donors aged 18–59 years and school children aged 8–10 years were the monitored population groups sampled in biennial periods; in both groups the number of subjects was about 100 in each urban area per year; in Kromˇerˇíˇz and Uherské Hradiˇstˇe per 50 subjects/year. The protocol of the study was approved by the Ethical Committee of the NIPH. A short questionnaire on medical, life style, occupational and dietary information was filled by blood donors/parents of children. The exclusion criteria were age, gender (to be approx. equal), a professional exposure in adults, resident time <2 years in a city. The sampling in children was performed in healthy subjects during compulsory preventive examinations by a total of 12 paediatricians. Informed consent and parental permission were necessary prior to sampling.

245

Fig. 2. Scheme of the AMA 254 spectrometer. (1) Sampling boat, (2) decomposition furnace, (3) catalytic column, (4) gold amalgamator, (5) releasing furnace, (6) mercury cathode lamp, (7) optical cell system, and (8) detector (Source: Spˇeváˇcková et al., 2004).

for the purposes of further evaluations the value equalling 1/2 of the respective limit of detection was used. The analytical method is described in more detail in Spˇeváˇcková et al. (2004). Analyses of food and human samples were performed in the National Public Health Institute laboratories. Analyses of food were preceded by standardized culinary processing. The procedure was established for each food commodity on the basis of a questionnaire survey of current methods of cooking techniques such as peeling, stewing, boiling and roasting (Ruprich et al., 2000). A correction factor for the culinary processing of food is included in exposure calculations expressing the change in mass by culinary processing (Ruprich et al., 2000). Validation of using the AMA 254 singlepurpose spectrometer for THg determination in hair was provided ˇ by Cejchanová et al. (2008). The laboratories participate in interlaboratory comparison tests on a national and international scale. Proficiency testing by FAPAS (UK) is used for food analysis. The biomonitoring laboratories take part in the annual International Interlaboratory Testing (EQUAS) organized by the German Society for Occupational and Environmental Medicine. Chemical analyses of all media were conducted in accredited, authorized or good laboratory practice certified laboratories. Statistics

Chemical analyses The total mercury concentration in all kinds of samples was determined by atomic absorption spectrophotometry. The special analyzer AMA 254 for mercury determination was used. The advanced mercury analyzer, AMA 254 (Altec Prague, Czech Republic), is a single-purpose atomic absorption spectrometer which permits direct determination of mercury in both liquid and solid materials without preceding mineralization. The sample being introduced into the analyzer is thermally treated. It is initially dried, then combusted in the stream of oxygen. Combustion gases are finally decomposed on the catalytic column at approx. 750 ◦ C. The mercury vapour is trapped quantitatively on the surface of a gold amalgamator. The scheme of AMA 254 is shown in Fig. 2. The light source is a low pressure mercury vapour lamp; detector is a silicon UV photodetector. The working range (with automatic switchover for lower and higher Hg concentrations) is 0.05–40 ng Hg or 40–500 ng Hg. Limits of detection were 0.1 ␮g/kg for food and soil, and 0.2 ␮g/l for drinking water. For blood the detection limit was 0.12 ␮g/l and for urine 0.06 ␮g/l. For hair the detection limit was 0.0012 ␮g/g. If the concentration was below the limit of detection,

Except of dietary exposure the quantitative data processing is based on the calculation of the nonparametric sample characteristics (median, percentile) as most data on environmental pollutant concentrations show a statistical distribution close to the lognormal one. Data from human biomonitoring were processed by the software Statistica 7.1. Kruskal–Wallis test was applied for statistical calculations of gender-related differences and relation of THg blood levels vs. fish consumption. All statistical tests were performed at the significance level of P < 0.05. Results Dietary exposure Total mercury content in most of food commodities was found very low. In the latest monitoring period 2006/2007, in milk and dairy products, butter, vegetable fat, tea, coffee, beer and beverages the mercury content ranged under the limit of detection in the majority of findings (0.1 ␮g/kg). In other monitored food commodities such as flour, bread, pastry, meat, canned meat, poultry,

246

V. Puklová et al. / International Journal of Hygiene and Environmental Health 213 (2010) 243–251

Table 1 Total mercury concentration in the most contaminated food commodities in 20042007. THg concentration (␮g/kg)

Smoked fish Sea fish Marinated fish Freshwater fish Canned fish Delicate fish salads Mushrooms Spice Pork Liver Spinach Rice

N

Min

Average

Max

16 16 16 16 16 8 8 8 16 16 16

32.2 9.4 9.6 7.6 5.8 7.5 0.4 2.6 0.2 0.7 0.2

42.8 39.0 28.2 22.0 19.0 9.5 4.3 3.2 1.3 1.3 0.6

68.2 138.0 39.3 71.4 39.4 11.6 6.5 3.6 5.7 2.2 1.0

N – number of samplings.

salami, potatoes, vegetable, fruits and eggs the total mercury content amounted up to 0.5 ␮g/kg. Quite remarkable concentrations were found in liver, spice, and mushrooms. The most meaningful concentrations were found in the food commodities such as smoked fish, marinated fish, sea fish, fresh water fish, canned fish and delicate (fish) salads. The average THg content of the most contaminated ten food commodities is shown in Table 1. The concentration data obtained in food commodities were converted to exposure data by multiplication of analytical data by the conversion factor for culinary treatment and factor for the amount of consumption. For an average (reference) person (integral lifetime value of 64 kg b.w.) a relatively low dietary exposure to THg from the food basket has been estimated. The single point estimates in particular monitoring years ranged 0.06–0.1 ␮g THg/kg b.w./week. In the latest monitoring period 2006/2007 the THg intake for an average person amounted to about 0.08 ␮g/kg b.w./week. From the particular food commodities the most significant daily exposure dose was found in sea fish followed by delicate (fish) salads, freshwater fish, canned and marinated fish. Relatively remarkable exposure was found from rice and potatoes. The contribution of the most significant food commodities in view of the total mercury dietary exposure dose is shown in Fig. 3. The population diet pattern could change over time in between particular food consumption studies. Therefore the trend in dietary THg exposure doses is expressed with the aid of a model of standardized foodstuff consumption for five population groups: children 4–6 years old; adult males over 18 years of age; adult

females over 18; pregnant and breast-feeding females; and the elderly over 60 years of age (Ruprich et al., 2003). For calculations of exposure doses we implemented the recommended doses of foodstuffs for those specified population groups according to Brázdová et al. (2001) and Ralph (1993). The recommended dose has a standard value for the whole monitoring period, unlike actual food consumption. The results reflect the trend in mercury concentrations in the whole consumer food basket. A gradual increase is apparent (Fig. 4). Exposure from drinking water During the monitoring period 2004–2008, low total mercury levels in drinking water from the Czech public supply networks were detected. Out of a total of 27,512 samples, about 85% findings ranged below the limit of detection 0.2 ␮g/l. A total of 37 isolated values (i.e. 0.13% of all samples) were detected exceeding the limit set by the Czech national drinking water quality standard for total mercury 1 ␮g/l. These findings represented isolated values. No water supply in which the mercury levels exceeded constantly the standard value was identified. As regards the mercury content in drinking water from wells which supply 8% of the Czech population, the available data are from over 2000 public and commercial wells collected by the national drinking water monitoring system. In 2004–2008, out of a total of 5223 samples about 80% of findings were below the limit of detection. The limit value was exceeded in 13 cases, i.e. 0.3% of samples. The maximum value amounted to 3.3 ␮g/l. The range of mercury levels in drinking water from the water supply networks and public wells in the 2004–2008 period are shown in Table 2. The data are presented separately for the larger water supply zones serving a population of more than 5000, and for smaller supply zones. Based on data from the national monitoring we estimated that the mercury intake from drinking water of 99.8% of the supplied population is less than 1% PTWI. Only 0.2% of the population (i.e. about 18,000) supplied by the drinking water public network has a higher mercury intake than 1% PTWI but not higher than 5% of PTWI (upon intake of 2 l piped drinking water daily). Mercury levels in urban surface soil In the monitoring period 2002–2006, the mercury content in surface soil of 324 nursery schools in 24 cities and towns was

Fig. 3. Total mercury mean exposure doses from the most significant food commodities, 2006/2007.

V. Puklová et al. / International Journal of Hygiene and Environmental Health 213 (2010) 243–251

247

Fig. 4. Trend in the model of the total mercury exposure doses, 1994–2006/2007. Table 2 Mercury levels in drinking water from the public supply network in the Czech Republic, 2004–2008. Size of supply zoneYear

Max. value (␮g/l)

AM (␮g/l)

GM (␮g/l)

Under LD

>5000 inh. 2004 2005 2006 2007 2008

2.94 4.79 1.00 2.45 1.00

0.14 0.13 0.13 0.13 0.13

0.10 0.11 0.11 0.11 0.10

1314 1208 1210 1229 1243

<5000 inh. 2004 2005 2006 2007 2008

3.00 3.10 3.19 2.48 3.00

0.15 0.14 0.13 0.14 0.13

0.10 0.11 0.11 0.11 0.10

Public and commercial wells 2004 2005 2006 2007 2008

3.30 2.70 1.00 3.30 2.00

0.15 0.14 0.12 0.12 0.12

0.08 0.10 0.10 0.09 0.09

Exceeded

Total number of samples

Under LD (%)

3 1 0 1 0

1501 1369 1347 1310 1373

88 88 90 94 91

2816 3574 3575 3576 3777

11 5 2 7 7

3583 4317 4257 4068 4387

79 83 84 88 86

515 723 848 1044 1115

3 4 0 4 2

680 904 1073 1267 1299

76 80 79 82 86

Note: LD – limit of detection. AM – arithmetic mean. GM – geometric mean.

monitored. The city median values ranged between 0.07 and 0.62 mg/kg dry matter. The highest median value was found in a city polluted by lead smelting industry. The variability of findings was two orders of magnitude between the particular nursery schools within a city as well as between the cities. The median value of all localities amounted to 0.16 mg/kg with 95th percentile being 0.49 mg/kg. The descriptive statistics of total mercury content in surface soil of nursery school playgrounds is shown in Table 3. Based on the total mercury content in the surface soil of nursery school playgrounds the potential oral exposure of preschool Table 3 The concentration of the total mercury in the surface soil of nursery school playgrounds (mg/kg of dry matter). N = 324 Geom. mean Arithm. mean Standard deviation Median 95th percentile 25th percentile 75th percentile

0.15 0.19 0.17 0.16 0.49 0.09 0.24

children was estimated. A highly conservative residential scenario according to the US EPA (US EPA, 1997) was used. The exposure estimate to mercury through unintentional soil ingestion was obtained by a mathematical model calculating the average daily dose for non-carcinogenic substances (US EPA, 2006). Calculation was based on the median soil mercury value and the probable exposure duration of 210 days/year according to central European conditions. The estimated exposure dose amounted to 1.23E−09 ␮g/kg b.w./day, which is about 0.2% of the JECFA FAO tolerable value for oral intake of the total mercury. The Hazard Index value of 0.002 revealed no significance to health. Mercury levels in human body fluids and tissues Mercury in blood In adults, blood mercury levels (medians) in the period 1996–2003 ranged from 0.53 to 1.31 ␮g/l (in men) and 0.81 to 1.33 ␮g/l (in women), respectively, with no marked trend in time. Significant gender-related differences were observed with higher values in women. Results obtained in 2005 and 2007 in the new set of urban areas appear to have a slightly decreasing time-trend, but longer monitoring period is needed to confirm this. A signifi-

248

V. Puklová et al. / International Journal of Hygiene and Environmental Health 213 (2010) 243–251

Table 4 Blood mercury levels in adults and in children (␮g/l). 1996

1997

1998

1999

2000

2001

2002

2003

2005

Men N Median P 0.95 Range

284 0.79 2.01
291 0.84 3.86
314 0.53 2.22
297 0.78 2.29
300 1.31 3.34 0.17–7 .60

286 0.81 2.84
290 0.80 3.13
290 0.95 2.87
233 0.91 2.66
248 0.85 2.56 0.17–4.29

Women N Median P 0.95 Range

134 0.83 2.04 0.24–4.55

103 0.93 3.35 0.16–6.58

81 0.81 3.50
101 0.94 2.66 0.19–4.47

98 1.33 4.37
114 0.93 3.60 0.26–9.32

107 0.92 4.15
105 0.99 3.51
172 1.16 3.46
163 0.89 2.94
Children N Median P 0.95 Range

380 0.57 1.98
384 0.39 1.25
362 0,38 1.38
354 0.42 1.48
2006

2007

382 0.45 1.39
2008

198 0.35 1.32
N – number of samples. P 0.95 – 95th percentile.
cant increase in blood mercury levels with fish consumption was observed from zero consumption to occasional consumption (less than once per week) and to frequent consumption (one and more portion of fish meal per week, Fig. 5). The similar significant relation was observed also in children from median value of 0.22 ␮g/l by zero consumption to 0.34 ␮g/l by occasional consumption and 0.43 ␮g/l by one and more portion per week. In children, blood mercury levels were about half of adult levels. The gender-related difference in children was not significant. The blood THg levels in adults and children are shown in Table 4. Mercury in urine In adults, median levels in the 1996–2003 period ranged from 0.53 to 0.78 ␮g/g creatinine. The median concentrations obtained in the population monitored since 2005 in the new set of cities had higher values (1.30 and 1.10 ␮g/g creatinine in 2005 and 2007, respectively) than in previous period. Statistically significant gender-related difference was observed with higher values in women. In children, median concentrations in 1996–2003 ranged from 0.25 to 0.43 ␮g/g creatinine. In new urban areas, the median values in children were similar to those in the previous period – 0.26 and 0.16 ␮g/g creatinine in 2006 and 2008, respectively. The urine THg levels in adults and children are shown in Table 5.

Fig. 5. Blood mercury levels in adults in relation to fish consumption, 2007.

Mercury in children’s hair The total hair mercury levels in Czech children are relatively low. In the years 1996–2008 the median values ranged 0.13–0.26 ␮g/g. In all monitoring years, the 95th percentile values were below the

Table 5 Urine mercury levels in adults and children (␮g/g creatinine). 1996 Men N Median P 0.95 Range

247 0.61 2.79
Women N Median P 0.95 Range

114 1.29 4.66
Children N Median P 0.95 Range

435 0.25 2.54
1997

397 0.38 2.56
N – number of samples. P 0.95 – 95th percentile.
1998

2000

2002

2003

2005

294 0.51 2.70
275 0.63 5.23
251 0.44 5.39
246 0.63 4.93
165 0.84 5.13
170 0.90 4.72
73 0.99 13.27
84 0.90 7.07
84 1.05 11.81
76 1.09 10.52
113 2.18 10.37
109 1.57 8.55
384 0.35 3.15
349 0.43 3.94
270 0.28 4.46
399 0.27 4.22
1999

393 0.28 2.40
2006

364 0.26 2.19
2007

2008

312 0.16 1.01
V. Puklová et al. / International Journal of Hygiene and Environmental Health 213 (2010) 243–251

249

Table 6 Hair mercury levels in children (␮g/g).

Children N Median P 0.95 Range

1996

1997

1998

1999

2000

2001

2002

2003

2006

2008

412 0.23 0.54 0.05–1.36

372 0.20 0.54 0.05–0.92

359 0.16 0.30 0.05–0.53

360 0.16 0.37 0.05–0.60

343 0.26 0.84 0.05–10.60

325 0.20 0.72 0.06–1.31

319 0.20 0.50 0.06–0.93

292 0.14 0.50 0.02–1.98

372 0.13 0.28 0.04–0.74

316 0.18 0.61 0.05–7.49

N – number of samples. P 0.95 – 95th percentile.
value 1 ␮g/g considered as safe limit (US EPA, 1997). The hair THg levels in children are shown in Table 6. No gender-related differences were found. The positive correlation between the groups of children never eating fish (0.15 ␮g/g) and frequently eating fish (one portion and more per week, 0.20 ␮g/g) was observed. Discussion Besides a certain number of drinking water samplings, no systematic data are available on possible exposure to mercury in the rural environment or in small municipalities with all its specific traits, e.g. partial home production of foodstuffs or individual sources of drinking water. There are about 750,000 private wells in the Czech Republic; however, they are not an object of a routine control. The estimate of mercury burden does not implicate the rural population. The mercury dietary intake found for the average Czech person was lower than the intake of an average adult (70 kg) presented in the European expert’s assessment (SCOOP, 2004) for the 13 European Union which states 5.53 ␮g/day (0.553 ␮g THg/kg b.w./week). JECFA/FAO WHO established the provisional tolerable weekly intake for the total mercury as 5 ␮g/kg b.w./week (JECFA, 1978). No more than 3.3 ␮g/kg b.w./week should be present as methylmercury (expressed as mercury). The estimates for the Czech average person thus represent 1.1–1.9% of the WHO PTWI value; it amounted to 1.7% PTWI in the latest monitoring period 2006/2007. Although the Czech Republic is a landlocked country with only freshwater fish production and all seafood is imported, fish and fish products are a major source of mercury exposure. Nevertheless, fish consumption has been traditionally low in the Czech population; in the last decade it has been slowly increasing (5.9 kg/person in 2008). The majority of consumed fish and fish products comprise sea fish, with fresh water fish representing a smaller proportion (Ruprich et al., 2004). A higher dietary intake of mercury is naturally expected in children due to relatively greater food consumption per unit of body mass. Moreover, the latest individual food consumption study (Ruprich et al., 2006) revealed the highest frequency of fish and fish products consumption among children out of all population groups. In that survey 18% and 14% of children aged 4–6 and 7–10 years, respectively, were fish consumers. In adult women and men it was 12% and 11%, respectively (Ruprich et al., 2006). The most vulnerable group represents pregnant women due to transplacental transport of methylmercury and possibility of fetal impairment resulting in neuropsychic disorders in child. The Czech Scientific Food Committee released a recommendation for public in 2004 which advises especially women in fertile age, pregnant and breastfeeding women and small children to consume kinds of fish with low mercury content. It is generally assumed that in fish and fish products about 90% of mercury is in a form of methylmercury (SCOOP, 2004). In 2004 the Joint FAO/WHO Expert Committee on Food Additives established a provisional weekly tolerable intake (PTWI) of 1.6 ␮g/kg b.w./week

for methylmercury in order to protect the developing fetus from neurotoxic effects (JECFA, 2004). Later, JECFA clarified that lifestages other than the embryo and fetus may be less sensitive to the adverse effects of methylmercury (JECFA, 2006). Though it is known that adults are more tolerant to neurotoxic effects, this conclusion cannot be drawn for children and adolescents. Therefore, the tolerable intake established in 2004 has been applied to a general population (WHO, 2007). This PTWI was used also as Reference Dose by EFSA (2004). The US EPA set a more strict Reference Dose in IRIS (last revised in 2001) for chronic oral exposure to methylmercury of 0.1 ␮g/kg b.w./day for developmental neuropsychological impairment. Since the inorganic mercury in food is considerably less toxic as methylmercury, the worst scenario is that total mercury intake equals the methylmercury intake. Under this scenario the estimated dietary exposure dose for the average Czech person amounted to about 5.4% PTWI or 12.3% RfD in 2006/2007. Of course, the situation can be different in cases of greater fish consumption. In the Czech Republic the number of recreational fishermen has always been very high, at over 100,000 persons. For these persons (and their families) the additional fish meat consumption should be considered – about 10 g/person/day (Ruprich et al., 2004), with resulting higher mercury intake. According to the estimate of methylmercury usual intake distribution based on the concentration and on fish consumption data from the current individual food consumption study the proportion of the Czech population which could be highly exposed to methylmercury doses exceeding the tolerable intake PTWI is less than 0.01% with a probability of 97.5% ˇ uˇ ˚ rková, 2006). (Ruprich and Reh WHO recommends a guideline value of 6 ␮g/l for inorganic mercury in drinking water (WHO, 2006). Within the monitoring period, the highest value amounting to 4.8 ␮g/l was detected in the water main supplying 11,500 inhabitants. The control sampling which followed within one month did not confirm exceeding of the standard in that supply network. There was no temporal exemption granted by the public health protection authority due to increased mercury content in drinking water. In Europe, little evidence is available about exposure to mercury in soils (WHO, 2007). A recent study on soil mercury content in six European cities revealed high variability of surface soil mercury content, both within each locality and between cities, the median values ranging between 0.06 to 1.2 mg/kg dry matter (Rodrigues et al., 2006). In Polish city gardens from a typical urban environment with only diffuse pollution sources and traffic influence the mercury surface soil content ranged 0.052–0.29 mg/kg (Dabkowska ˇ and Roˇzanski, 2007). The values found in the nursery playgrounds in Czech urban environment (median 0.16 mg/kg, 95th percentile 0.49 mg/kg) are in agreement with those findings. Biologically significant limit values (HBM values) for blood and urine mercury levels were established by the German Commission on Human Biological Monitoring (Ewers et al., 1999). In fact, two levels of limits (HBM I and HBM II) are used for health-related population exposure. HBM values for mercury are the same for both sexes of adults and children. In blood, the current HBM I value (no

250

V. Puklová et al. / International Journal of Hygiene and Environmental Health 213 (2010) 243–251

Table 7 The mercury levels in blood and urine in the biomonitoring surveys. Population group

N

Matrix

Type of value

Value

Source

Adults 18–65 years Women 16–49 years Children 1–5 years Children 1–5 years Children 6–11 years Women 20–29 years Adults 18–69 years Adults 18–69 years Women 16–49 years Adults 18+ years Adults 18–65 years

383 1709 705 1879 1790 563 4645 4730 1240 2342 152

Urine Blood Blood Blood Blood Blood Blood Urine Blood Blood Blood

Median GM GM Median Median Median Median Median Median Median GM

0.78 ␮g/g cr. 1.02 ␮g/l 0.34 ␮g/l 0.3 ␮g/l 0.4 ␮g/l 0.74 ␮g/l 0.6 ␮g/l 0.3 ␮g/g cr. 0.9 ␮g/l 0.38 ␮g/l 2.38 ␮g/l

Apostoli et al. (2002) Schober et al. (2003) Schober et al. (2003) Caldwell et al. (2009) Caldwell et al. (2009) Caldwell et al. (2009) Becker et al. (2002) Becker et al. (2003) Vupputuri et al. (2005) Son et al. (2009) Gundacker et al. (2006)

Note: N – number of samples. GM – geometric mean.

adverse population health risk below this value can be expected) is 5 ␮g/l and the HBM II (adverse population health risk above this value can be expected) is 15 ␮g/l. In 2007, only two adult persons (0.7%) exceeded the blood mercury threshold value 5 ␮g/l. However, National Research Council in 2000 recommended a lower limit value (3.4 ␮g/l) for women in reproductive age because of an increased risk of neurotoxic effects on the fetus in case of potential pregnancy. In the period 1996–2003, a total of 52 women had blood mercury level in excess of 3.4 ␮g/l; however, not all of them were in reproductive age. The reference value in blood (the higher bound of CI 95th percentiles) established for the period 2001–2003 was 3.45 ␮g/l for the total adults (Batáriová et al., 2006). This reference value was higher than that of 2.0 ␮g/l proposed for the German adults (Wilhelm et al., 2004). Also a higher value was found in 2007 in Czech adults who never eat fish (0.5 ␮g/l) in comparison to that obtained by GerES III (never fish consumption 0.3 ␮g/l, Becker et al., 2002). The comparison of the mercury levels found in the Czech Republic with selected biomonitoring outputs from other countries enables Table 7. The Czech blood levels are somewhat higher in comparison to German study results, similar to those obtained in Italy and in the US population. The range of values found for Austrian population (0.34–9.97 ␮g/l blood, Gundacker et al., 2006) is similar to the range observed in Czech population (see Table 4), the mean value was nevertheless higher in Austrian adults (2.38 ␮g/l in 2002/2004) than in Czech men and in women (0.85 and 0.95 ␮g/l in 2007). The HBM values for mercury in urine are 5 ␮g/g (HBM I) and 20 ␮g/g (HBM II) creatinine for both adults and children. In 2005 and 2007, only 7% of persons exceeded the HBM I value of 5 ␮g/g creatinine. The median value (95th percentile) in the 1994–2001 period was 0.19 ␮g/g (0.68 ␮g/g) (Beneˇs et al., 2003). The reference value (the higher bound of CI 95th percentiles) established for the 2001–2003 period was 6.8 ␮g/g creatinine for total adults, with significantly higher reference value in women (12 ␮g/g creatinine) than in men (5.4 ␮g/g creatinine) (Batáriová et al., 2006). The Czech reference values were higher than those obtained for German population (Wilhelm et al., 2004). Hair mercury concentration is the preferred biomarker for evaluating mercury exposure for extended periods of time (Bencko, 1995). In hair of children, positive correlation was found between hair mercury levels and fish consumption never versus one portion and more per week. Similarly, positive correlation in Czech conditions was found also by Kruzikova et al. (2008). In the Czech Republic the state of dentition is mainly examined by the dental health surveys, i.e. dental caries prevalence, the number of teeth that are filled due to caries for particular population age groups, but regardless of the filling quality. The Czech Dental Chamber recommends, in agreement with the Council of European Dentists resolution from 2007, the use of amalgam (pre-dosed, with a stable constitution, in contrast to hand-made amalgam mixes) as

a suitable dental filling material. There can be noted that no representative statistics is available but the use of amalgams is probably still largely spread in CR. Conclusions Estimates of mercury intake from food as the major exposure source are relatively favourable for the Czech Republic in comparison with intake values in other European countries due to still lower fish consumption in the Czech population. In view of the recognized benefits from fish consumption it is actually desirable to increase fish consumption in the Czech population. The public health authorities make efforts to educate the public, namely women of fertile age, about the importance of eating fish together with informed choice of fish types in terms of mercury content. A higher dietary mercury intake is expected in children due to relatively greater food consumption per unit of body mass. Population exposure from drinking water in public supply networks is not expected to pose adverse health effects. Potential oral exposure of preschool children from unintentional ingestion of soil on children’s playgrounds in the urban environment was found to be insignificant. Estimated exposure to mercury in the Czech population at present does not signalize any increased risk for public health. Mercury levels in body fluids and hair do not represent any substantial burden, and have no marked trend in time. However, particular population groups under risk (e.g. pregnant women, fishermen, above average fish consumers) and residents in spot-contaminated localities should be focused on. Acknowledgments The authors of the paper would like to express their appreciation and thanks to the cooperating public health regional centres for their efforts in performing the monitoring activities and the Ministry of Health for a kind financial support. References Agency for Toxic Substances and Disease Registry (ATSDR), 1999. Toxicological Profile for Mercury. Department of Health and Human Services, Public Health Service, Atlanta, GA, USA. Apostoli, P., Cortesi, I., Mangili, A., Elia, G., Drago, I., Gagliardi, T., Soleo, L., Valente, T., Sciarra, G.F., Aprea, C., Ronchi, A., Minia, C., 2002. Assessment of reference values for mercury in urine: the results of an Italian polycentric study. Sci. Total Environ. 1–3, 13–24. ˇ ˇ ˇ Batáriová, A., Spˇeváˇcková, V., Beneˇs, B., Cejchanová, M., Smíd, J., Cerná, 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. J. Hyg. Environ. Health 4, 297–308.

V. Puklová et al. / International Journal of Hygiene and Environmental Health 213 (2010) 243–251 Becker, K., Schulz, Ch., Kaus, S., Seiwert, M., Seifert, B., 2003. German Environmental Survey 1998 (GerES III): environmental pollutants in the urine of the German population. Int. J. Hyg. Environ. Health 1, 15–24. Bellinger, D.C., Trachtenberg, F., Barregard, L., Tavares, M., Cernichiari, E., Daniel, D., McKinlay, S., 2006. Neuropsychological and renal effects of dental amalgam in children. JAMA 295, 1775–1783. Bencko, V., 1995. Use of human hair as a biomarker in the assessment of exposure to pollutants in occupational and environmental settings. Toxicology 101, 29–39. ˇ Beneˇs, B., Sladká, J., Spˇeváˇcková, V., Smíd, J., 2003. Determination of normal concentration levels of Cd, Cr, Cu, Hg, Pb, Se and Zn in hair of the child population in the Czech Republic. Cent. Eur. J. Publ. Health 11, 184–186. Brázdová, Z., Ruprich, J., Hrubá, D., Petráková, A., 2001. Dietary guidelines in the Czech Republic III: challege for the 3rd millenium. Cent. Eur. J. Publ. Health 9 (1), 30–34. Caldwell, K.L., Mortensen, M.E., Jones, R.L., Caudill, S.P., Osterloh, J.D., 2009. Total blood mercury concentrations in the U.S. population: 1999–2006. Int. J. Hyg. Environ. Health 6, 588–598. ˇ Cejchanová, M., Spˇeváˇcková, V., Kratzer, K., Wranová, K., Spˇeváˇcek, V., Beneˇs, B., 2008. Determination of mercury and methylmercury in hair of the Czech children’s population. Biol. Trace. Elem. Res. 2, 97–105. Crépet, A., Tressou, J., Verger, P., Leblanc, J.Ch., 2005. Management options to reduce exposure to methyl mercury through the consumption of fish and fishery products by the French population. Regul. Toxicol. Pharmacol. 2, 179–189. ˇ Dabkowska, H., Roˇzanski, S., 2007. Mercury content in garden soils of urban agglomeration. Gnest. Int. J. 3, 237–241. EFSA, 2004. Opinion of the scientific panel on contaminants in the food chain on a request from the commission related to mercury and methylmercury in food. EFSA J. 1, 1–14. Ewers, U., Krause, C., Schulz, C., Wilhelm, M., 1999. Reference values and human biological monitoring values for environmental toxins. Int. Arch. Occup. Environ. Health 72, 255–260. GEMS/Food WHO, 1985. Guidelines for the Study of Dietary Intake of Chemical Contaminants. WHO Offset Publication No. 87. Gundacker, C., Komarnicki, G., Zödl, B., Foerster, C.H., Schuster, E., Wittmann, K., 2006. Whole blood mercury and selenium concentrations in a selected Austrian population: does gender matter? Sci. Total Environ. 1, 76–86. Joint FAO/WHO Expert Committee on Food Additives JECFA, 1978. Evaluation of Certain Food Additives and Contaminants: Twenty-second Report of the Joint FAO/WHO Expert Committee on Food Additives. WHO Technical Report Series No. 631, Geneva, WHO. Joint FAO/WHO Expert Committee on Food Additives JECFA, 2004. Methylmercury. Safety Evaluation of Certain Food Additives and Contaminants. Sixty-first Report of the Joint FAO/WHO Expert Committee on Food Additives. Technical Report Series 922, Geneva, WHO, pp. 132–139. Joint FAO/WHO Expert Committee on Food Additives JECFA, 2006. Methylmercury. Summary and Conclusions of the 67. Joint FAO/WHO Expert Committee on Food Additives. WHO Technical Report Series 940, Geneva, WHO. ˇ sma, P., Ruprich, J., Cerná, ˇ Kliment, V., Kubínová, R., Kazmarová, H., Havlík, B., Siˇ M., Kodl, M., 1997. System of monitoring the environmental impact on population health of the Czech Republic. Cent. Eur. J. Publ. Health 5, 107–116. Kruzikova, K., Modra, H., Kensova, R., Skocovska, B., Wlasow, T., Svoboda, T., Svobodova, Z., 2008. Mercury in human hair as an indicator of the fish consumption. Neuroendocrinol. Lett. 5, 675–679. Larsen, E.H., Andersen, N.L., Moller, A., Petersen, A., Mortensen, G.K., Petersen, J., 2002. Monitoring the content and intake of trace elements from food in Denmark. Food Addit. Contam. 1, 33–46. Mahaffey, K.R., 2004. Fish and shellfish as dietary sources of methylmercury and the ω-3 fatty acids, eicosahexaenoic acid and doxosahexaenoic acid: risks and benefits. Environmen. Res. 95, 414–428. ˜ Munoz, O., Bastias, J.M., Araya, M., Morales, A., Orellana, C., Rebolledo, R., Velez, D., 2005. Estimation of the dietary intake of cadmium, lead, “mercury, and arsenic by the population of Santiago (Chile) using a total diet study. Food Chem. Toxicol. 11, 1647–1655. Mutter, J., Neumann, J., Sadaghiani, C., Walach, H., Drasch, G., 2004. Amalgam studies: disregarding basic principles of mercury toxicity. Int. J. Hyg. Environ. Health 4, 391–397. NIPH, 2000. Methodology of sampling of soil from free playgrounds in purpose of studies of microbiological processes and for physical- and chemical analyses. NIPH Prague (In Czech) Available at: http://www.szu.cz/uploads/ documents/chzp/puda/legislativa venkovni hraci plochy/Odber vzorku pudy VHP.pdf. National Centre for Immunisation Research and Surveillance of Vaccine Preventable Diseases (NCIRS), 2009. Thiomersal Fact Sheet. NCIRS, Last Updated 2009. Available at: http://www.ncirs.usyd.edu.au/facts/thiomersal%20Aug%2009.pdf. ˇ Puklová, V., Batáriová, A., Cerná, M., Kotlík, B., Kratzer, K., Melicherˇcík, J., Ruprich, J., ˇ uˇ ˚ rková, I., Spˇeváˇcková, V., 2005. Cadmium exposure pathways in the Czech Reh urban population. Cent. Eur. J. Publ. Health 1, 11–19. Ralph, J., 1993. Dietary reference values. In: Garrow, J.S., James, W.P.T. (Eds.), Human Nutrition and Dietetics, 9th ed. Churchil Livingstone, Edinburgh, pp. 792–793.

251

Rodrigues, S., Pereira, M.E., Duarte, A.C., Ajmone-Marsan, F., Davidson, C.M., Grˇcman, H., Hossack, I., Hursthouse, A.S., Ljung, K., Martini, C., Otabbong, E., Reinoso, R., Ruiz-Cortés, E., Urquhart, G.J., Vrˇscˇ aj, B., 2006. Mercury in urban soils: a comparison of local spatial variability in six European cities. Sci. Total Environ. 2–3, 926–936. ˇ ˇ uˇ ˚ rková, I., 1993. Food Basket Ruprich, J., Cernoeviˇ cová, M., Kopˇriva, V., Resová, D., Reh for the Czech Republic, Exposure Factors – CZ 1991. National Institute of Public Health, Prague, ISBN 80-90034-0-0. (in Czech). ˇ uˇ ˚ rková, I., Ruprich, J., Dofková, M., Kleinwachterová, H., Kopˇriva, V., Resová, D., Reh 1997. Food Basket for the Czech Republic, Exposure Factors – CZ 1994. National Institute of Public Health, Prague, ISBN 80-7071-058-6. (in Czech). ˇ uˇ ˚ rková, I., 2000. Food Basket Ruprich, J., Dofková, M., Kopˇriva, V., Resová, D., Reh for the Czech Republic, Exposure Factors – CZ 1997. National Institute of Public Health, Prague, ISBN 80-7071-166-3. (in Czech). Ruprich, J., et al., 2003. Health Effects from Burden by Chemicals in Food Chains. Technical Report from Environmental Health Monitoring System. National Institute of Public Health, Prague (in Czech). http://www.chpr.szu.cz/ monitor/tds03c/tds03c.htm. ˇ uˇ ˚ rková, I., Drápal, J., Kozáková, M., 2004. Attitude of the Scientific Ruprich, J., Reh Committee for Food, Methylmercury in Fish and Fish Products. National Institute of Public Health, Prague (in Czech). (http://www.chpr.szu.cz/vedvybor/ dokumenty/stanoviska/stan 2004 5 deklas Hg rev1.pdf). ˇ uˇ ˚ rková, I., Slamˇeníková, E., Resová, D., 2006. IndividRuprich, J., Dofková, M., Reh ual Food Consumption – National Study SISP04. Centre for the Hygiene of the Food Chains, National Institute of Public Health, Prague (in Czech). (http://www. chpr.szu.cz/spotrebapotravin.htm). ˇ uˇ ˚ rková, I., 2006. The Probabilistic Model of the Total Mercury Intake Ruprich, J., Reh in the Czech population, As Methylmercury Intake Estimate. National Institute of Public Health, Prague (in Czech). http://www.chpr.szu.cz/vedvybor/ dokumenty/stanoviska/stan 2006 5 deklas Hg rev1 add1.pdf. Schober, S.E., Sinks, T.H., Jones, R.L., Boler, P.M., McDowell, M., Osterloh, J., Garrett, E.S., Canady, R.A., Diplom, Ch.F., Sun, Y., Joseph, C.B., Mahaffey, K.R., 2003. Blood mercury levels in US children and women of childbearing age, 1999–2000. JAMA 289, 1667–1674. Scientific cooperation SCOOP, 2004. Assessment of the Dietary Exposure to Arsenic, Cadmium, Lead and Mercury of the Population of the EU Member States. Reports on Tasks for Scientific Cooperation. EC Directorate-General Health and Consumer Protection, Brussels. Son, J., Lee, J., Paek, D., Lee, J., 2009. Blood levels of lead, cadmium, and mercury in the Korean population: results from the second Korean national human exposure and bio-monitoring examination. Environmental Research 6, 738–744. ˇ ´ M., 2004. SamSpˇeváˇcková, V., Korunová, V., Cejchanová, M., Vobecky, pling procedure and a radio-indicator study of mercury determination in whole blood by using an AMA 254 atomic absorption spectrometer. Anal. Bioanal. Chem. 380, 346–350, Available at: http://www.springerlink.com/ content/xpyhaejy65wr7tv5/. United Nations Environment Programme (UNEP), 2002. Global Mercury Assessment. UNEP Chemicals Mercury Programme. UNEP, Available at: www.chem.unep.ch/mercury/Report/Final%20Assessment%20report.htm. US EPA, 1997. Exposure Factors Handbook. Environmental Protection Agency, Office of Research and Development, National Center for Environmental Assessment, Washington Office, Washington, DC, EPA/600/P-95/002Fa-c. US EPA, 2006. Child-Specific Exposure Factors Handbook (External Review Draft) 2006. U.S. Environmental Protection Agency, Washington, DC, EPA/600/R06/096A. Vupputuri, S., Longnecker, M.P., Daniels, J.L., Guo, X., Sander, D.P., 2005. Blood mercury level and blood pressure among US women: results from the National Health and Nutrition Examination Survey 1999–2000. Environmental Research 2, 195–200. WHO, 2005. Mercury in Drinking-water. Background Document for Development of WHO Guidelines for Drinking-water Quality. WHO, WHO/SDE/WSH/05.08/10. WHO, 2006. Guidelines for Drinking-water Quality, First addendum to third edition. WHO, Geneva. WHO, 2007. Joint WHO/Convention Task Force on the Health Aspects of Air Pollution. Health Risks of Heavy Metals from Long Range Transboundary Pollution. WHO Regional Office for Europe, Copenhagen, Denmark, http://www.euro.who.int/document/E91044.pdf. 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. Ysart, G., Miller, P., Croasdale, M., Crews, H., Robb, P., Baxter, M., de L’Argy, C., Harrison, N., 2000. 1997 UK total diet study – dietary exposures to aluminium, arsenic, cadmium, chromium, copper, lead, mercury, nickel, selenium, tin and zinc. Food Addit. Contam. 9, 775–786. Zimmer, H., Ludwig, H., Bader, M., Bailer, J., Eickholz, P., Staehle, H.J., Triebig, G., 2002. Determination of mercury in blood, urine and saliva for the biological monitoring of an exposure from amalgam fillings in a group with self-reported adverse health effects. Int. J. Hyg. Environ. Health 3, 205–211.