Evaluation of the use of human hair for biomonitoring the deficiency of essential and exposure to toxic elements

SC IE N CE OF T H E TOT AL E N V I RO N ME N T 4 0 5 ( 2 00 8 ) 3 7 0–3 76

a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m

w w w. e l s e v i e r. c o m / l o c a t e / s c i t o t e n v

Evaluation of the use of human hair for biomonitoring the deficiency of essential and exposure to toxic elements Jairo L. Rodrigues, Bruno L. Batista, Juliana A. Nunes, Carlos J.S. Passos, Fernando Barbosa Jr. ⁎ Laboratório de Toxicologia e Essencialidade de Metais, Depto. de Análises Clínicas, Toxicológicas e Bromatológicas, Faculdade de Ciências Farmacêuticas de Ribeirão Preto-USP, Avenida do Café s/n, Monte Alegre, 14040-903, Ribeirão Preto-SP, Brazil

AR TIC LE I N FO

ABS TR ACT

Article history:

Monitoring the nutritional status of essential elements and assessing exposure of

Received 27 March 2008

individuals to toxic elements is of great importance for human health. Thus, the

Received in revised form 4 June 2008

appropriate selection and measurement of biomarkers of internal dose is of critical

Accepted 6 June 2008

importance. Due to their many advantages, hair samples have been widely used to assess

Available online 15 July 2008

human exposure to different contaminants. However, the validity of this biomarker in evaluating the level of trace elements in the human body is debatable. In the present study,

Keywords:

we evaluated the relationship between levels of trace elements in hair and whole blood or

Hair

plasma in a Brazilian population. Hair, blood and plasma were collected from 280 adult

ICP-MS

volunteers for metal determination. An ICP-MS was used for sample analysis. Manganese,

Trace metals

copper, lead and strontium levels in blood varied from 5.1 to 14.7, from 494.8 to 2383.8, from

Biomarkers

5.9 to 330.1 and from 11.6 to 87.3 μg/L, respectively. Corresponding levels in hair varied from

Nutrition

0.05 to 6.71, from 0.02 to 37.59, from 0.02 to 30.63 and from 0.9 to 12.6 μg/g. Trace element

Blood

levels in plasma varied from 0.07 to 8.62, from 118.2 to 1577.7 and from 2.31 to 34.2 μg/L for

Exposure

Mn, Cu and Sr, respectively. There was a weak correlation (r = 0.22, p < 0.001) between lead levels in hair and blood. Moreover, copper and strontium levels in blood correlate with those levels in plasma (r = 0.64 , p < 0.001 for Cu) and (r = 0.22, p < 0.05 for Sr). However, for Cu, Mn and Sr there was no correlation between levels in hair and blood. Our findings suggest that while the idea of measuring trace elements in hair is attractive, hair is not an appropriate biomarker for evaluating Cu, Mn and Sr deficiency or Pb exposure. © 2008 Elsevier B.V. All rights reserved.

1.

Introduction

Dietary habits and environmental conditions may partly affect trace element levels in both tissues and biological fluids, and consequently their participation in numerous biochemical mechanisms (Dona et al., 2006). Essential elements are those that are required by an organism to maintain its normal physiological function. Without the essential elements, the organism cannot complete its normal life cycle or achieve normal healthy growth; many such elements are key components of metalloenzymes or are involved in crucial biological

functions, such as oxygen transport, free radical scavenging, or hormonal activity (Parsons and Barbosa, 2007). On the other hand, many nonessential elements are so ubiquitous in the environment that they are easily detected in human body tissues and fluids. Some are relatively benign, but others, such as Pb, Cd, Hg and As, are quite toxic even at concentrations considered trace (Barbosa et al., 2006a,b; Parsons and Barbosa, 2007). Monitoring the nutritional status of essential elements and assessing exposure of individuals to toxic elements are of critical importance in human health. Today, the assessment of human exposure to background levels of trace elements in

⁎ Corresponding author. Tel.: +55 16 36024701. E-mail address: [email protected] (F. Barbosa). 0048-9697/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.scitotenv.2008.06.002

S CIE N CE OF T H E TOT AL E N V I RO N ME N T 4 0 5 ( 2 00 8 ) 3 7 0–3 76

the environment through measurement of those chemicals or their metabolites in human specimens is termed biomonitoring (Angerer et al., 2007; Parsons and Barbosa, 2007). Most clinical methods used to diagnose trace element deficiencies or to assess environmental or occupational exposure to toxic elements rely on the analysis of blood, serum/ plasma, and/or urine specimens. However, the choice of the appropriate specimen depends on several factors, such as toxicokinetics (time of appearance and residence time of the biological parameter), the convenience or invasiveness of the specimen collection procedure, and the potential for specimen contamination. Thus, the appropriate selection and measurement of biomarkers is of critical importance for health care management purposes, public health decision making, and primary prevention activities. Several alternative, i.e., non-traditional, specimen matrices including saliva, hair, and nails (Wilhelm et al., 1994; Nowak and Chmielnicka, 2000; Wilhelm et al., 2002; Pereira et al., 2004; Barbosa et al., 2006a,b; Slotnick and Nriagu, 2006) that permit non-invasive collection procedures have been explored, Hair is a biological specimen that is easily and noninvasively collected, inexpensive, and easily stored and transported to the laboratory for analysis. These attributes make hair an attractive biomonitoring substrate, at least superficially (Barbosa et al., 2005). These advantages have led to the widespread use of trace element analysis of hair samples to assess wildlife and human exposure to different contaminants present in the environment (Schuhmacher et al., 1991; Wilhelm et al., 1994; Schuhmacher et al., 1996; Sen and Chaudhuri, 1996) or at the workplace (Ashraf et al., 1994). However, hair analysis is subject to certain limitations, such as the occurrence of exogenous contamination. This contributes to a differential increase in the total contents of different contaminants (Bencze, 1990; Miekeley et al., 1998; ATSDR, 2001; Frisch and Schwartz, 2002). The main sources of exogenous contaminants are deposits of sebum, sweat, polluted air residues or residues of cosmetic or pharmaceutical products. Some other constraints on the use of hair analysis have also been pointed out (Bozsai, 1992; ATSDR, 2001; Seidel et al., 2001; Harkins and Susten, 2003) These constraints include the lack of scientific knowledge about the kinetics of trace element incorporation in hair and the insufficiency of epidemiological data to support predictions concerning the health effects, of a specific concentration of each element in hair. Moreover, given the growing use of hair analyses in health studies, an assessment of the biomarker validation criteria, which include the correlation of the levels found in this specimen with those found in blood or plasma, is called for. The aim of this paper was to evaluate the use of hair as a biomarker of Sr, Zn and Cu deficiency and/or Pb exposure. Thus, the relationship between the level of these elements in hair with their levels in whole blood or plasma was obtained in an adult Brazilian population.

2.

Materials and methods

2.1.

Population

We studied 280 healthy adults (47% women and 53% men) between 18 and 60 years of age from 3 different Brazilian states

371

(São Paulo, Minas Gerais and Pará). Ethical approval was obtained from the Ethics Committee of the University of São Paulo at Ribeirão Preto (Brazil).

2.2.

Sample collection

2.2.1.

Blood and hair collection

A trained Brazilian nurse collected a 4-mL blood sample from each participant. Blood samples were collected in trace-metalfree evacuated tubes (BD Vacutainer®) containing heparin as an anticoagulant. Two mL of blood was then pippeted into an eppendorf tube (2 mL volume) previously cleaned in a 100 clean room and immediately frozen at −20 degrees Celsius before analysis. For plasma separation, 2 mL of blood samples were centrifuged (1000 ×g for 6 min). The plasma fraction was then pipetted into an eppendorf tube (2 mL volume) previously cleaned in a 100 clean room and was immediately frozen at −20 °C before analysis. On the same day as blood collection, hair samples were taken from the occipital area of the head, close to the scalp. The lock of hair was stapled at the base and stored in labeled Ziploc bags. Hair samples were cut into 1 cm lengths and washed before analysis. From each 1-cm hair sample collected, 20 mg was weighed for trace element determination.

2.2.2.

Hair washing

Hair was washed according to the method proposed by Ohmori (1984), with acetone, water and acetone. After washing, samples were dried in a class — 100 laminar flow hood before analysis.

2.3.

Sample analysis

2.3.1.

Reagents

All reagents used were of analytical-reagent grade except HNO3, which was previously purified in a quartz sub-boiling still (Kürner) before use. A clean laboratory and laminar-flow hood capable of producing class 100 were used for preparing solutions. High purity de-ionized water (resistivity 18.2 MΩ cm) obtained from a Milli-Q water purification system (Millipore, Bedford, MA, USA) was used throughout. All solutions were stored in high-density polyethylene bottles. Plastic bottles and glassware were cleaned by soaking in 10% (v/v) HNO3 for 24 h, rinsed five times with Milli-Q water and dried in a class 100 laminar flow hood before use. All operations were performed on a clean bench.

2.3.2.

Instrumentation

All measurements were made with an ICP-MS (Elan DRC II PerkinElmer, Norwalk, CT) with high-purity argon (99.999%, White Martins, Brazil). A Meinhard concentric nebulizer (Spectron/Glass Expansion, Ventura, CA, USA) connected to a cyclonic spray chamber was used. A radiofrequency (rf) with 1100 watts of power was selected in pulse mode with autolens one. Sample data were acquired by using 20 sweeps/reading, 1 reading/replicate and a dwell time of 50 ms. Argon nebulizer gas flow rate was optimized daily from 0.5 to 0.9 L min- 1. Data were acquired in counts per second (cps). The following isotopes were selected: 63Cu, 55Mn, 208Pb, 88Sr.

372

SC IE N CE OF T H E TOT AL E N V I RO N ME N T 4 0 5 ( 2 00 8 ) 3 7 0–3 76

Table 1 – Analytical performance for the determination of trace elements in reference hair, blood and plasma samples (EQAS) from the Institut National de Sante' Publique du Quebec (INSP) Analyte

Cu Pb Mn

ICP 03H09 hair (n = 5)

ICP 03B06 blood (n = 5)

EQAS 06S06 serum (n = 5)

Found value (μg/g)

Target value (μg/g)

Found value (μg/L)

Target value (μg/L)

Found value (μg/L)

Target value (μg/L)

79.4 ± 0.8 5.1 ± 0.3 5.7 ± 0.2

84.0 ± 4.4 4.60 ± 0.69 6.40 ± 0.58

2385 ± 23 391 ± 3 9.7 ± 0.2

2350 ± 114 396 ± 26 10.4 ± 1.2

1942 ± 18 7.9 ± 0.1 4.3 ± 0.3

1890 ± 111 8.20 ± 0.83 3.97 ± 1.05

Multielement stock solutions containing 1000 mg L− 1 of each element were obtained from Perkin-Elmer (PerkinElmer, Norwalk, CT). Rhodium was used as the internal standard at the concentration of 10 μg/L Rh. The internal standard was diluted from a 1000 mg/L stock standard solution (PerkinElmer, Norwalk, CT, USA).

Triton X-100. Rhodium was added as internal standard to get a 10 μg L− 1 final concentration. After that, samples were directly analyzed by ICP-MS. The method's detection limit was 0.05, 0.06 and 0.01 μg/L for Cu, Sr and Mn, respectively. The detection limits were obtained as 3 SD/slope of the 20 consecutive measurements of the matrix-matching blanks.

2.3.3. Determination of trace elements in hair, whole blood and plasma

2.3.4.

Trace element levels in hair were determined by Inductively Coupled Plasma Mass Spectrometry according to the method proposed by Rodrigues et al. (2008). Briefly, hair samples (20 mg), were accurately weighed into (15 mL) conical tubes. Then, 1 ml of 25% m/v tetramethylammonium hydroxide (TMAH) solution was added to the samples and incubated at room temperature overnight. Following this, the volume made up to 10 mL with a solution containing 1% v/v HNO3. Rhodium was added as an internal standard to get a final concentration of 10 μg/L. After that, samples were directly analyzed by ICP-MS. The method's detection limits were 0.8, 3.5, 2.5, and 1.0 ng/g for Pb, Cu, Sr, and Mn, respectively. Trace metal levels in whole blood were determined by Inductively Coupled Plasma-Mass Spectrometry (ICP-MS) according to the method proposed by Palmer et al. (2006). In brief, blood samples (200 μL), were pipetted into (15 mL) conical tubes. Then, the volume made up to 10 mL with a solution containing 0.5% v/v HNO3 + 0.005% v/v Triton X-100. Rhodium was added as internal standard to get a 10 μg L− 1 final concentration. After that, samples were directly analyzed by ICP-MS. The method's detection limits were 0.02, 0.09, 0.007, and 0.01 μg/L for Pb, Cu, Sr and Mn, respectively. Trace element levels in plasma were determined by Inductively Coupled Plasma-Mass Spectrometry (ICP-MS). For this, we used the method proposed by Goullé et al. (2005) with little modification. Briefly, plasma samples (500 μL), were pipetted into (15 mL) conical tubes. Then, the volume made up to 10 mL with a solution containing 0.5% v/v HNO3 + 0.005% v/v

Quality control of the results

Quality control (QC) of trace metal determination in clinical specimens was ensured by analyzing Reference Materials provided by the Institut National de Santé Publique du Québec, Canada (INSP-external quality assessment scheme) (EQAS) for trace elements in blood, plasma and hair. Reference samples were analyzed before and after ten ordinary samples. Obtained values together with reference values are shown in Table 1. Reference values for strontium were not provided by the INSP.

2.3.5.

Statistical analysis

Since the data are non-parametric, Spearman's correlation was used to evaluate the correlation between trace element levels in all biomarkers. All statistical analyses were performed using SPSS (SPSS, Inc., Chicago, IL).The statistical significance was set at p < 0.05.

3.

Results

Table 2 summarizes the mean and the range of trace element concentrations found in the different biomarkers evaluated in the population of the present study. Manganese levels in hair (Mn-H), blood (Mn-B) and plasma (Mn-P) of the studied population varied from 0.05 to 6.71 μg/g, from 5.1 to 14.7 μg/L and from 0.07 to 8.62 μg/L, respectively. Manganese mean levels were 1.3 μg/g, 8.9 μg/L, and 2.1 μg/L in hair, blood and plasma, respectively. These levels are in agreement with previous publications dealing with other

Table 2 – Concentrations of the trace elements present in the population of this study Parameters Hair Mn Blood Mn Plasma Mn Hair Cu Plasma Cu Blood Cu Hair Pb Blood Pb Hair Sr Blood Sr Plasma (μg g− 1) (μg L− 1) (μg L− 1) (μg g− 1) (μg L− 1) (μg L− 1) (μg g− 1) (μg L− 1) (μg g− 1) (μg L− 1) Sr (μg L− 1) Mean Standard Deviation Median Range

1.3 2.4

8.9 4.1

2.1 1.3

5.7 4.0

922.3 168.1

1184.5 287.9

2.5 3.7

115.2 68.7

1.6 1.5

31.0 11.5

15.4 4.2

0.7 0.05–6.7

7.5 5.1–14.7

2.0 0.07–8.6

5.9 0.02–37.6

908.7 118–1578

1149.1 495–2384

1.5 0.02–31

93.8 6–330

1.3 1–12

28.4 12–87.3

14.3 2–34

S CIE N CE OF T H E TOT AL E N V I RO N ME N T 4 0 5 ( 2 00 8 ) 3 7 0–3 76

373

Fig. 3 – The relationship between strontium levels in blood and hair (r = 0.06; p = 0.92) (n = 280).

Fig. 1 – The relationship between biomarkers of copper: (a)-copper in hair and in blood (r = 0.05; p = 0.43); (b)-copper in blood and in plasma (r = 0.69; p < 0.001) (n = 280). populations (Pereira et al., 2004; Goullé et al., 2005). There were no statistical correlations between Mn levels in hair and blood (r = 0.100, p = 0.08), Mn levels in blood and plasma (r = 0.122, p = 0.06) or between Mn levels in plasma and hair (r = 0.07, p = 0.21). The levels of copper in hair (Cu-H), blood (Cu-B) and plasma (Cu-P) varied from 0.02 to 37.6 μg/g, from 494.8 to 2383.8 μg/L and from 118.2 to 1577.7 μg/L, respectively. Copper mean levels were 5.7 μg/g, 1184.5 μg/L, and 922.3 μg/L for hair, blood and plasma, respectively. Again these levels are in the same range found in previous studies with other populations (Goullé et al., 2005; Cerna et al., 2007). The relationship between biomarkers for copper is shown in Fig. 1. For this element, a significant

Fig. 2 – The relationship between lead levels in blood and in hair (r = 0.22; p < 0.001) (n = 280).

statistical correlation was found only between levels in blood with plasma levels (r = 0.64, p < 0.001), Fig. 1b. Levels of lead found in hair (Pb-H) and blood (Pb-B) varied from 0.1 to 30.6 μg/g and from 5.9 to 330.1 μg/L, respectively. There was a weak significant statistical correlation between these biomarkers (r = 0.22, p < 0.001) as can be seen in Fig. 2. Mean Pb-B and Pb-H levels were 115.2 μg/L and 2.5 μg/g, respectively. For strontium, the levels found in hair (Sr-H), blood (Sr-B) and plasma (Sr-P) varied from 0.9 to 12.6 μg/g, from 11.6 to 87.3 μg/L and from 2.3 to 34.2 μg/L, respectively. A weak statistical correlation was found between Sr-B levels and the corresponding levels in plasma (r = 0.22, p < 0.05). On the other hand, there was no correlation between Sr-B levels and Sr-H levels (r = 0.06, p = 0.92) as shown in Fig. 3, or between Sr-P and Sr-H levels (r = 0.02, p = 0.54) Few differences in the coefficients of correlation were observed for any elements when data were analyzed after separation by sex.

4.

Discussion

To our knowledge, this study is the first describing the relationship between essential and toxic trace element levels in hair and those levels in blood and plasma in a wide adult population composed of men and women. The trace element levels found in hair, blood and plasma collected from our population presented high variation and were in the same range as the levels found in other studies with different populations (Bárány et al., 2002; Bryan et al., 2007; D'Haese et al., 1997; Goullé et al., 2005; Pereira et al., 2004; Mortada et al., 2002; Cerna et al., 2007; Sen et al., 1996; Caroli et al., 1994). Blood or plasma/serum are the most common specimens used as biomarkers of internal dose to diagnose deficiency or exposure to toxic elements. For instance, serum/plasma copper concentration is a reliable biomarker of copper deficiency, falling to very low concentrations in copperdeficient individuals (Kumar et al., 2007). In our study, copper levels in blood varied from 494.8 to 2383.8 μg/L and in plasma from 118.2 to 1577.7 μg/L. There was a very good correlation between the two markers. These findings suggest that Cu–B could reflect the same copper status as Cu–P. The same wide

374

SC IE N CE OF T H E TOT AL E N V I RO N ME N T 4 0 5 ( 2 00 8 ) 3 7 0–3 76

variation was observed for copper levels in hair (0.02 to 37.6 μg/g). However, there was no statistical correlation between Cu–H levels with those levels in Cu–P or Cu–B. This suggests that copper in hair may not reflect dietary intake of this essential element and is also probably not sensitive to marginal copper status. Serum or plasma manganese concentrations appear to be somewhat sensitive to large variations in manganese intake, but longer studies are needed to evaluate the usefulness of serum manganese concentrations as a biomarker of manganese status. An advantage of whole blood manganese concentration over plasma or serum manganese as a biomarker of deficiency is that slight hemolysis of samples can markedly increase plasma or serum manganese concentrations. However, according to the literature, whole blood manganese seems to be extremely variable. Even so, the most common method used to estimate changes in manganese metabolism and status is to measure its concentration in whole blood or serum/plasma. In the present study, a wide variation in the concentrations of Mn in blood was observed (5.1–14.7 μg/L) which is in line with reference values published for other populations (Caroli et al., 1994). We also observed the same wide variation for Mn in plasma (0.07–8.62 μg/L) and in hair (0.05–6.71 μg/g). However, a correlation between Mn–B and Mn–P levels was not observed. Furthermore, Mn–H levels did not correlate with those levels in blood or plasma. Over the last few decades, whole blood has been the primary biological fluid used for assessment of Pb exposure, both for screening and diagnosis and for long-term biomonitoring. Although Pb–B measurements reflect recent exposure, they may also represent past exposures as a result of Pb mobilization from bone back into blood (Barbosa et al., 2005). In those subjects without excessive exposure to Pb, 45–75% of the Pb in blood may have come from bone (Barbosa et al., 2005). In exposed children, however, it has been reported that 90% or more of Pb in the blood may be contributed by Pb in bone (Barbosa et al., 2005). Plasma/serum Pb levels in nonexposed and exposed individuals reported in older publications range widely, from 0.02 to 14.5 μg/L (Versieck and Cornelis, 1988). This is probably due to inappropriate collection methods, analytical instrumentation, and methods for Pb determination. The development and use of more sensitive analytical instrumentation, especially inductively coupled plasma mass spectrometry (ICP-MS), has resulted in determinations of Pb in plasma and serum specimens with much lower detection limits and with better accuracy. More recent data, also based on ICP-MS methods, have shown Pb–P levels < 1.0 μg/L in nonexposed individuals (Caroli et al., 1994; Barbosa et al., 2006a,b). However, the use of advanced analytical techniques is not the only essential requirement for ensuring accurate and reliable Pb–P data. Contamination of the specimen may occur in the preanalytical phase, namely, during collection, manipulation, or storage. Since the sampling conditions in the present study did not allow for accurate measurement of lead levels in plasma, Pb–P levels were not reported. However, previous studies published by our group demonstrated that the relationship between Pb–B and Pb–P is curvilinear. The two quantities are related by the line y = 0.0006x1492 (y = Pb–P, and x = Pb–B) (Barbosa et al., 2006a).

In the present study, there was a weak correlation between Pb–B levels and Pb levels in hair. Our findings are in line with those reported in two recent studies: one by Stupar et al. (2007) of young Slovenian males and another by Wilhelm et al. (2002) of children from Germany. The weak correlation between the two markers may be attributed to the large variation in Pb intake and the different kinetics of Pb appearance in hair and blood. However, it is evident from the foregoing studies that the correlation between Pb–H and Pb–B varies tremendously (r = 0.03–0.76). It appears that the level of environmental or occupational lead pollution considerably affects the strength of this correlation. Niculescu et al. (1983) and Clayton and Wooler (1983) found the strongest correlations for heavily exposed lead-battery workers, while the correlations were much lower in the control population. A similar exponential accumulation of Pb in hair with simultaneous linear increase of Pb–B was reported by both teams of researchers. An interesting observation was reported in a study with children living in a highly polluted town in Russia (Esteban et al., 1999). In that study, the mean content of Pb-B and Pb-H was 98 μg/L and 7.2 μg/g, respectively. The whole population's correlation between log Pb-H and log Pb-B was statistically significant (r = 0.45, p < 0.05),.If highly exposed children were excluded, however, the significance of the correlation was lost (r = 0.07). Thus, it seems that hair may be a better biomarker in cases of high exposure. Hair structure contains a high amount of sulfur because the amino acid cysteine is a key component of the keratin proteins in hair fiber. The sulfur in cysteine molecules in adjacent keratin proteins links together in disulfide chemical bonds. Some heavy metals, such as mercury and lead, have a high affinity to sulfur. This could in part explain the weak correlation found between Pb–B and Pb–H levels in our study and the strong correlation observed in other studies between Hg in blood and Hg in hair (Berglund et al., 2005). At the average growth rate of roughly 1 cm per month, consecutive 1-cm segments of hair recapitulate average monthly blood levels. There is an approximately 20-day lag between the concentration of trace elements in the first centimeter next to the scalp and the corresponding average monthly blood level (Clarkson and Magos, 2006). Thus, fluctuation in the dietary intake of essential elements or the exposure to toxic elements over time could explain part of the lack of correlation between hair and blood levels, since blood and hair were collected from each volunteer on the same day. Moreover, the composition of trace elements in food is highly variable. Most of the volunteers of the present study have a typical diet without variations over time, however. On the other hand, factors such as age, interactions between elements and genetics may to a greater or lesser extent modify the metabolism of the trace element and its mobilization from the blood to the hair compartment (Chojnacka et al., 2006; Khalique et al., 2005; Paschal et al., 1989). In other studies, strong correlations were observed between Hg levels in hair and in blood (Berglund et al., 2005). Moreover, this biomarker is currently used to evaluate Hg exposure (Li et al., 2008; Guentzel et al., 2007). However, on the basis of our results, hair is not an appropriate biomarker for other elements such as those evaluated in the present study (Pb, Sr, Mn and Cu).

S CIE N CE OF T H E TOT AL E N V I RO N ME N T 4 0 5 ( 2 00 8 ) 3 7 0–3 76

In summary, while the idea of measuring trace elements in hair is attractive, our findings suggest that the use of this biomarker is only appropriate for some specific elements.

Acknowledgements The authors are grateful to Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) for financial support and fellowships.

REFERENCES Angerer J, Eweres U, Wihelm M. Human biomonitoring: state of art. Int J Hyg Environ Health 2007;210(3-4):201–28. Ashraf W, Jaffar M, Mohammad D. Trace metal contamination study on scalp hair of occupationally exposed workers. Bull Environ Contam Toxicol 1994;53:516–23. ATSDR – Agency for Toxic Substance and Disease Registry. Hair Analysis Panel Discussion: Exploring the State of the Science. June 12-13. Summary Report. Agency for Toxic Substances and Disease Registry, Division of Health Education and Promotion. Atlanta, Georgia, 2001. Available at: http://www.atsdr.cdc.gov/ HAC/hair_analysis, accessed on march 03rd, 2008. Bencze K. What contributions can be made to biological monitoring of hair analysis? Part I. Fresenius J Anal Chem 1990;337:867–76. Bárány E, Bergdahl IA, Eric L, Lundh T, Samuelson G, Schutz A, et al. Trace element levels in whole blood and serum from Swedish adolescents. Sci Total Environ 2002;286:129–41. Barbosa Jr F, Tanus-Santos JE, Geriach RF, Parsons PJ. A critical review of biomarkers used for monitoring human exposure to lead: advantages, limitations, and future needs. Environ Health Perspect 2005;113:1669–74. Barbosa Jr F, Ramires I, Rodrigues MHC, Saint' Pierre TD, Curtius AJ, Buzalaf MR, et al. Contrasting effects of age on the plasma/ whole blood lead ratio in men and women with a history of lead exposure. Environ Res 2006a;102:90–5. Barbosa Jr F, Rodrigues MHC, Buzalaf MR, Krug FJ, Gerlach RF, Tanus-Santos JE. Evaluation of the use of salivary lead levels as a surrogate of blood lead or plasma lead levels in lead exposed subjects. Arch Toxicol 2006b;80:633–7. Berglund M, Lind B, Bjornberg KA, Palm B, Einarsson O, Vahter M. Inter-individual variations of human mercury exposure biomarkers: a cross-sectional assessment. Environ Health 2005;4:1–11. Bozsai G. Quality control and assurance in hair analysis. Michochem J 1992;46:159–66. Bryan CE, Cristopher SJ, Balmer BC, Wells RS. Establishing baseline levels of trace elements in blood and skin of bottlenose dolphins in Sarasota Bay, Florida: implications for non-invasive monitoring. Sci Total Environ 2007;388:325–42. Caroli S, Alimonti A, Coni E, Petrucci F, Senofonte O, Violante N. The assessment of reference values for elements in human biological tissues and fluids: a systematic review. Crit Rev Anal Chem 1994;24:363–98. Cerna M, Spevackova V, Batariova A, Smid J, Cejchanova M, Ocadlikova D, et al. Human biomonitoring system in the Czech Republic. Int J Hyg Environ Health 2007;210:495–9. Chojnacka K, Gorecka H, Gorecki. The effect of age, sex, smoking habit and hair color on the composition of hair. Environ Toxicol Pharmacol 2006;22:52–7. Clarkson TW, Magos L. The toxicology of mercury and its chemical compounds. Crit Rev Toxicol 2006;36:609–62.

375

Clayton E, Wooler KK. Ion Beam analysis of scalp hair as a monitor of occupational exposure. IEEE trans Nucl Sci 1983;NS30:1326–8. D'Haese PC, Landeghem GFV, Lamberts LV, Bekaert VA, Schrooten I, DeBroe ME. Measurement of strontium in serum, urine, bone, and soft tissues by Zeeman atomic absorption spectrometry. Clin Chem 1997;43:121–8. Dona EK, Dona A, Papoutsis J, et al. Cooper and Zinc concentrations in serum of healthy Greek adults. Sci Total Environ 2006;359:76–81. Esteban E, Rubin CH, Jones RL, Noonan G. Hair and blood as a substrates for screening children for lead poisoning. Arch Environ Health 1999;54:436–40. Frisch M, Schwartz BS. The pitfalls of hair analysis for toxicants in clinical practice: three case reports. Environ Health Perspect 2002;110(4):433–6. Goullé JP, Mahieu L, Castermant J, Neveu N, Bonneau L, Laine G, et al. Metal and metalloid multi-elementary ICP-MS validation in whole blood, plasma, urine and hair reference values. Forensic Sci Int 2005;153:39–44. Guentzel JL, Portilla E, Keith KM, Keith EO. Mercury transport and bioaccumulation in riverbank communities of the Alvarado Lagoon System, Veracruz State, Mexico. Sci Total Environ 2007;388:316–24. Harkins DK, Susten AS. Hair analysis: exploring the state of science. Environ Health Perspect 2003;111(4):576–8. Khalique A, Ahmad S, Anjum T, Jaffar M, Shah MH, Shaheen N, et al. A comparative study based on gender and age dependence of selected metals in scalp hair. Environ Monit Assess 2005;104:45–57. Kumar N, Butz JA, Burritt MF. Clinical significance of the laboratory determination of low serum copper in adults. Clin Chem Lab Med 2007;45:1402–10. Li P, Feng X, Qiu G, Shang L, Wang S. Mercury exposure in the population from wuchuan mercury mining area, Guizhou, China. Sci Total Environ 2008;395:72–9. Miekeley N, Dias Carneiro MTW, Porto da Silveira CL. How reliable are human hair reference intervals for trace elements? Sci Total Environ 1998;218:9–17. Mortada WI, Sobh MA, El-Defrawy MM, Farahat SE. Reference intervals of Cd, Pb, Hg in blood, urine, hair and nails among residents in Mansoura City, Nile Delta, Egypt. Environ Res 2002;90:104–10. Niculescu T, Dumitru R, Botha V, Alexandrescu R, Manolescu N. Relationship between the lead concentration in hair and occupational exposure. Br J Ind Med 1983;40:67–70. Nowak B, Chmielnicka J. Relationship of lead and cadmium to essential elements in hair, teeth, and nails of environmentally exposed people. Ecotoxicol Environ Saf 2000;46:265–74. Ohmori S. Hair as a monitor for environmental pollution, variations in element concentrations for different lead processing, workers and different washing methods of their hair. J Radioanal Nucl Chem 1984;84:451–9. Palmer CD, Lewis Jr ME, Geraghty CM, Barbosa F, Parsons PJ. Determination of lead, cadmium and mercury in blood for assessment of environmental exposure: a comparison between inductively coupled plasma–mass spectrometry and atomic absorption spectrometry. Spectrochim Acta Part B Atom Spectrosc 2006;61:980–90. Parsons PJ, Barbosa F. Atomic spectrometry and trends in clinical laboratory medicine. Spectrochim Acta Part B Atom Spectrosc 2007;62:992–1003. Paschal D, DiPietro ES, Phillips DL, Gunter EW. Age Dependence of metals in hair in a selected U.S. Population. Environ Res 1989;48:17–28. Pereira R, Ribeiro R, Gonçalves F. Scalp hair analysis as a tool in assessing human exposure to heavy metals (S. Domingos mine, Portugal). Sci Total Environ 2004;327:81–92. Rodrigues JL, Nunes JA, Batista BL, Souza SS, Barbosa F. A fast method for the determination of 16 elements in hair samples

376

SC IE N CE OF T H E TOT AL E N V I RO N ME N T 4 0 5 ( 2 00 8 ) 3 7 0–3 76

by inductively coupled plasma mass spectrometry (ICP-MS) with tetramethylammonium hydroxide solubilization at room temperature. J Anal Atom Spect 2008;23:992–6. Schuhmacher M, Domingo JL, Llobet JM, Corbella J. Lead in children's hair as related to exposure in Tarragona province, Spain. Sci Total Environ 1991;104:167–73. Schuhmacher M, Bélles M, Rico A, Domingo JL, Corbella J. Impact of reduction of lead in gasoline on the blood and hair lead levels in the population of Tarragona Province, Spain, 1990–1995. Sci Total Environ 1996;184:203–9. Seidel S, Kreutzer R, Smith D, McNeel S, Gilliss D. Assessment of commercial laboratories performing hair mineral analysis. JAMA 2001;285(1):67–72. Sen J, Chaudhuri ABD. Human hair lead and copper levels in three occupationally unexposed population groups in Calcutta. Bull Environ Contam Toxicol 1996;57:321–6.

Slotnick MJ, Nriagu JO. Validity of human nails as a biomarker of arsenic and selenium exposure: A review. Environ Res 2006;102:125–39. Stupar J, Dolinsek F, Erzen I. hair-Pb longitudinal profiles and blood–Pb in the population of young Slovenian males. Ecotoxicol Environ Saf 2007;68:134–43. Versieck J, Cornelis R. Trace elements in human plasma and serum. Boca Raton, Florida: CRC Press; 1988. Wilhelm M, Lombeck I, Ohnesorge FK. Cadmium, copper, lead and zinc concentrations in hair and toenails of young children and family members: a follow-up study. Sci Total Environ 1994;141:275–80. Wilhelm M, Pesch A, Rostek U, Begerow J, Shmitz N, Idel H, et al. Concentrations of lead in blood, hair and saliva of German children living in three different areas of traffic density. Sci Total Environ 2002;297:109–18.