What do the trace metal contents of urine and toenail samples from Qatar's farm workers bioindicate? $ Nora Kuiper a , Candace Rowell a , Jerome Nriagu b , Basem Shomar a,n a Qatar Environment and Energy Research Institute (QEERI), Qatar Foundation, P.O. Box 5825, Doha, Qatar b Department of Environmental Health Sciences, School of Public Health, University of Michigan,1415 Washington Heights, Ann Arbor, MI 48109, USA article info Article history: Received 14 January 2014 Received in revised form 12 February 2014 Accepted 17 February 2014 Available online 25 March 2014 Keywords: Biomonitoring Trace elements Farm workers Urine Toenails abstract Qatar's farm workers provide a unique population for exposure study: they are young, healthy males. This study combined trace element profiles in urine and toenail with survey information from 239 farm workers to assess the extent to which the biomarkers provide complementary exposure information. Urinary Mo levels (average ¼114 mg/L) were elevated; average urinary values (mg/L) for all other elements were: V (1.02), Cr (0.55), Mn (2.15), Fe (34.1), Co (0.47), Ni (2.95), Cu (15.0), As (47.8), Se (25.7), Cd (1.09), Ba (22.5), Pb (2.50) and U (0.15). Average toenail concentrations (mg/kg) were: Mn (2.48), Cu (4.43), As (0.26), Se (0.58), Mo (0.07), Cd (0.03), Ba (1.00), Pb (0.51) and U (0.02). No significant association was found between corresponding elements in urine and toenails. Elemental profiles suggest groundwater (with the exception of Mo) and soil–dust–crop exposure pathways cannot account for elemental variations. The main factors moderating trace element contents are related to depuration processes involving participants' trace element body burden prior to work in Qatar, and interactions of trace element metabolic cycles which over-ride the exposure footprint. Toenail and urine need to be carefully validated before reliable use as biomarkers of exposure in general populations for most elements in the study. & 2014 Elsevier Inc. All rights reserved. 1. Introduction Epidemiological studies of workers and populations exposed to high levels of some trace elements tend to show that when used congruently, urine and toenails may provide critical information on both short and long-term exposures, with urine providing an estimate of recent exposures, on the order of a few hours or days, while toenails reflect extended exposure from six to 12 months prior (Aguilera et al., 2008; Calafat, 2012; Karagas et al., 2001; Slotnick et al., 2005; Slotnick and Nriagu, 2006). The validity of this observation for all elements assessed under all population conditions has not been fully documen- ted (Heitland and Koster, 2006; Slotnick et al., 2005). Qatar has vast natural gas and oil reserves which drive the national economy and the gross domestic product (GDP) per capita to be one of the highest in the world (The World Bank, 2013). The recent rapid population increase in Qatar, due in part to the growing energy industry, has prompted the nation to expand its limited agricultural sector in order to reduce the dependence on food importation and vulnerability to changes in the global food market (QNFSP, 2013). Currently, farming in Qatar depends almost exclusively on foreign farm workers recruited mainly from countries in Asia and the Middle East. This study of Qatar's farm workers is the first of its kind in the country, offering unique perspectives on exposure and health in Qatar and the region. The recruitment process has given unique character- istics to the farm worker population in Qatar used for this exposure study: the population is young (controlling for age-related effects), healthy (controlling for pre-existing conditions) and only male (con- trolling for gender). This study examined the concentrations of essential and non- essential trace elements (As, Ba, Cd, Co, Cr, Cu, Fe, Mn, Mo, Ni, Pb, Se, U and V) in urine and toenail samples from 239 farm workers in Qatar to see the extent to which the elemental profiles of these biomarkers provide complementary exposure information. We were particularly interested in the influence of inter-element relationships on the concentrations of the elements in each biomarker after controlling for confounding variables including age, education level, duration of farm-work in Qatar, nationality and body mass index (BMI). 2. Materials and methods 2.1. Study area The study population consisted of immigrant farm workers from nine private and corporate farms in Qatar (Fig. 1). These farms constitute a representation of the various farms found throughout the country. Sampled farms ranged from small, family sites with one participant to large corporate farms with up to 72 participants, with the Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/envres Environmental Research http://dx.doi.org/10.1016/j.envres.2014.02.011 0013-9351/& 2014 Elsevier Inc. All rights reserved. ☆ This study received approval by the Qatar University Institutional Biosafety Committee. The research approval number is QU-IBC 08/11. n Corresponding author. E-mail address: [email protected](B. Shomar). Environmental Research 131 (2014) 86–94
Transcript
What do the trace metal contents of urine and toenail samples fromQatar's farm workers bioindicate?$
Nora Kuiper a, Candace Rowell a, Jerome Nriagu b, Basem Shomar a,n
a Qatar Environment and Energy Research Institute (QEERI), Qatar Foundation, P.O. Box 5825, Doha, Qatarb Department of Environmental Health Sciences, School of Public Health, University of Michigan, 1415 Washington Heights, Ann Arbor, MI 48109, USA
a r t i c l e i n f o
Article history:Received 14 January 2014Received in revised form12 February 2014Accepted 17 February 2014Available online 25 March 2014
Qatar's farm workers provide a unique population for exposure study: they are young, healthy males. Thisstudy combined trace element profiles in urine and toenail with survey information from 239 farmworkers toassess the extent to which the biomarkers provide complementary exposure information. Urinary Mo levels(average¼114 mg/L) were elevated; average urinary values (mg/L) for all other elements were: V (1.02), Cr(0.55), Mn (2.15), Fe (34.1), Co (0.47), Ni (2.95), Cu (15.0), As (47.8), Se (25.7), Cd (1.09), Ba (22.5), Pb (2.50) and U(0.15). Average toenail concentrations (mg/kg) were: Mn (2.48), Cu (4.43), As (0.26), Se (0.58), Mo (0.07), Cd(0.03), Ba (1.00), Pb (0.51) and U (0.02). No significant association was found between corresponding elementsin urine and toenails. Elemental profiles suggest groundwater (with the exception of Mo) and soil–dust–cropexposure pathways cannot account for elemental variations. The main factors moderating trace elementcontents are related to depuration processes involving participants' trace element body burden prior to work inQatar, and interactions of trace element metabolic cycles which over-ride the exposure footprint. Toenail andurine need to be carefully validated before reliable use as biomarkers of exposure in general populations formost elements in the study.
& 2014 Elsevier Inc. All rights reserved.
1. Introduction
Epidemiological studies of workers and populations exposed tohigh levels of some trace elements tend to show that when usedcongruently, urine and toenails may provide critical information onboth short and long-term exposures, with urine providing an estimateof recent exposures, on the order of a few hours or days, while toenailsreflect extended exposure from six to 12 months prior (Aguilera et al.,2008; Calafat, 2012; Karagas et al., 2001; Slotnick et al., 2005; Slotnickand Nriagu, 2006). The validity of this observation for all elementsassessed under all population conditions has not been fully documen-ted (Heitland and Koster, 2006; Slotnick et al., 2005).
Qatar has vast natural gas and oil reserves which drive the nationaleconomy and the gross domestic product (GDP) per capita to be one ofthe highest in the world (The World Bank, 2013). The recent rapidpopulation increase in Qatar, due in part to the growing energyindustry, has prompted the nation to expand its limited agriculturalsector in order to reduce the dependence on food importation andvulnerability to changes in the global food market (QNFSP, 2013).Currently, farming in Qatar depends almost exclusively on foreign
farm workers recruited mainly from countries in Asia and the MiddleEast. This study of Qatar's farm workers is the first of its kind in thecountry, offering unique perspectives on exposure and health in Qatarand the region. The recruitment process has given unique character-istics to the farm worker population in Qatar used for this exposurestudy: the population is young (controlling for age-related effects),healthy (controlling for pre-existing conditions) and only male (con-trolling for gender).
This study examined the concentrations of essential and non-essential trace elements (As, Ba, Cd, Co, Cr, Cu, Fe, Mn, Mo, Ni, Pb, Se, Uand V) in urine and toenail samples from 239 farmworkers in Qatar tosee the extent to which the elemental profiles of these biomarkersprovide complementary exposure information. We were particularlyinterested in the influence of inter-element relationships on theconcentrations of the elements in each biomarker after controllingfor confounding variables including age, education level, duration offarm-work in Qatar, nationality and body mass index (BMI).
2. Materials and methods
2.1. Study area
The study population consisted of immigrant farm workers from nine private andcorporate farms in Qatar (Fig. 1). These farms constitute a representation of the variousfarms found throughout the country. Sampled farms ranged from small, family siteswith one participant to large corporate farms with up to 72 participants, with the
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/envres
Environmental Research
http://dx.doi.org/10.1016/j.envres.2014.02.0110013-9351/& 2014 Elsevier Inc. All rights reserved.
☆This study received approval by the Qatar University Institutional BiosafetyCommittee. The research approval number is QU-IBC 08/11.
n Corresponding author.E-mail address: [email protected] (B. Shomar).
average being 26 workers per farm. Farms are concentrated in the north and northeastern regions of Qatar constrained by the Northern Groundwater Basin which coversapproximately 19% of the country's land area and has acceptable salinity ranges foragricultural irrigation (500–3000 mg/L). Groundwater in the southern regions of thecountry is highly saline, reaching 6000 mg/L in some areas (Darwish and Mohtar, 2012).Additionally, dune sands in the southern and western regions render the land unfit forfarming.
This study received approval from the Institutional Biosafety Committee atQatar University, Doha, Qatar. Informed consent was obtained from all participantsprior to collection of biological specimens and dissemination of the survey tool.
2.2. Sample collection and survey administration
Urine (u) and toenail (t) samples were collected between December 2012 andMarch 2013. Spot urine samples were collected on-site in polyethylene containersand stored in coolers with ice packs for transportation to laboratory facilities withsubsequent storage at 4 1C. Nail samples collected from all 10 toes with stainlesssteel clippers were stored in polyethylene vials at room temperature until analysis.
A brief questionnaire (approximately 10–15min) was administered in a face-to-faceinterview to each participant in their language of preference. The survey was used to
Fig. 1. Map of sampling locations in Qatar.
N. Kuiper et al. / Environmental Research 131 (2014) 86–94 87
obtain information regarding age, marital status, weight, height, nationality, educationlevel, crop consumption, length of time worked on farm, pesticide application methodsand duration, use of personal protective equipment, training status and health. Thesurvey instrument collected information on 29 symptoms typically associated with tracemetal poisoning and deficiency (Nriagu et al., 2011).
2.3. Sample preparation and analysis
Urine samples were diluted five-fold with 1% HNO3 (Optima, Fisher Chemical,Hampton, New Hampshire) to minimize differences in sample densities and thenanalyzed for total As, Ba, Cd, Co, Cr, Cu, Fe, Mn, Mo, Ni, Pb, Se, U and V usingFinnigan Element high resolution inductively-coupled plasma mass spectrometry(HR-ICP-MS) at the Keck Elemental Geochemistry Laboratory, Department of Earthand Environmental Sciences, University of Michigan, Ann Arbor, Michigan, USA.
Toenail samples were pre-washed and digested according to the proceduredescribed by Slotnick et al. (2005), with minor adjustments. Prior to digestion,toenails were cleaned by sonication and rinsed with acetone and Millipore water.Samples were oven dried overnight at 45 1C. Approximately 5–20 mg of samplewas predigested in 1 mL HNO3 (Optima, Fisher Chemical, Hampton, New Hamp-shire) for 5 min at room temperature and then gradually heated to 90 1C forapproximately 30 min. Solutions were cooled to room temperature and 1 mL H2O2
(Optima, Fisher Chemical, Hampton, New Hampshire) was added. The solution wasfurther heated (90 1C) until the sample was completely digested, approximately1.5 h. The final solution was diluted to 4 mL with Millipore water. Except for thepre-wash, the method was conducted under closed reflux conditions in a certifiedcleanroom.
For quality control, laboratory blanks and standard reference materials (SRM)were included in the analysis and digestion procedures. Each toenail digestionbatch included NIST 1400 (Bone Ash) and NIST 1515 (Apple Leaves) along with twolaboratory blanks. As validation of instrumental methods during the urine analysis,NIST 1640a (Trace Elements in Natural Water) diluted 20-fold was included inevery run batch of 15 samples.
2.4. Data analysis
All data analyses were performed using IBM SPSS Statistical Package 21. Linearregression, ANOVA, Student t-test and Pearson Correlations were performed on thedata set. Linear regressions were utilized to determine the effect of cofactors onexposure and health information. ANOVA and unpaired student t-tests wereperformed to determine if differences in elemental content of farm workersdiffered among farm sites, country of origin, age and burden of disease symptoms.Pearson correlations were used to determine correlations for inter-elementalanalyses. Cluster analysis was used when investigating health symptom correla-tions. Due to a lack of normal distribution of the data, all metal concentrations werelog-transformed prior to analysis. For establishing correlations between surveyresponses and elemental concentrations all non-responses were excluded fromanalysis.
3. Results and discussion
3.1. Distribution of trace elements in urine and toenail samples
The recovery of all reported elements in the SRMs along withinstrumental detection limits (DL) is shown in Table 1. Analysis ofNIST 1640a showed good sample recovery with data for all theelements being within 10% of the certified values. NIST 1400 waswithin 10% of the reference values for all elements with theexception of Cu and Fe. NIST 1515 was within 20% for As, Ba, Cd,Cu, Mn, Mo, Pb, Se and U. The poor recovery of the remainingelements (Co, Cr, Fe, Ni and V) during analysis of toenail samplesled to their exclusion in the analyses of the data for this biomarker.Less than 5% of the elements in urine samples were below the DL(oDL) with the exception of u-V (25.6% of samples were below)and u-U (67.0% of samples were below). For quality assurance, itneeds to be noted that all outliers were re-analyzed and thereported concentrations were confirmed. Additionally, low labora-tory blank values indicate that the observed values are not biasedby sample contamination. Concentrations oDL were given anarbitrary value of 50% DL during statistical analysis.
This study included farm workers who are currently workingand residing in farms in Qatar; no attempt was made to ascertainwhether they were farmers before coming to the country. The
demographic characteristics of the study participants are shown inTable 2. The farm workers were recruited primarily from eightcountries; the largest numbers were of Bangladeshi (48.1%) andNepalese (23.8%) origin. Participants were employed at the farmsites from two months to 30 years with the median length of thistype of work in Qatar being 4.2 years.
Of the population group sampled, more than 56% of allparticipants were between the ages of 21–35 years and the greatmajority (about 90%) were in the 21–50 age category (Table 2). TheBMI of 63% of participants were within the normal BMI range, withonly 3.4% being obese and 17% underweight (Table 2) (WHO,2013). The participants self-reported an average of 3 healthsymptoms with 33% reporting one or no symptoms at all(Table 2). The highest number of symptoms reported was 16 outof a possible maximum of 29. These data suggest that the farmworkers that participated in the study were remarkably healthy,but the results should be viewed circumspectively. Farming inQatar depends on recruiting young and healthy farmworkers fromneighboring countries who return to their countries of origin ifhealth or age affects their ability to perform the necessaryjob tasks.
The concentrations of the essential nutrients (Co, Cu, Fe, Mn,Mo and Se), and non-essential elements (As, Ba, Cd, Ni, Pb, U andV) in urine and toenails are shown in Tables 3 and 4, respectively.These tables also include typical ranges in concentrations reportedfrom studies in other parts of the world (urine) and pooled meanvalues for such studies (toenail). The results in these tables aremarked by large ranges in element concentrations in the urine andtoenail samples. Two participants were found to contain exceed-ingly high levels of u-As (783 mg/L and 1963 mg/L); surveyresponses did not indicate high exposure risk behavior for thesetwo participants and the u-As levels are likely due to non-occupational exposures such as recent dietary intake (Wanget al., 2013). The maximum values measured for u-Cr (1067 mg/L), u-Fe (4071 mg/L) and u-Ni (557 mg/L) were all from oneparticipant. Survey responses did not indicate any high riskexposure factors unique to this individual; we therefore suspectthat this worker was involved in the repair of farm machines andequipment made of stainless steel. These outliers were removedfrom the dataset prior to statistical analysis.
Urinary concentrations of V, Cd and Mo of Qatar's farmworkersare elevated relative to the typical ranges that have been reportedin studies in other parts of the world (Table 3). Analysis of soilsamples collected from each farm site did not indicate anyelevation in levels of V (average¼1.02 mg/kg), Cd (average¼0.17 mg/kg), or Mo (0.82 mg/kg) when compared to acceptedworldwide background levels (Shomar, unpublished data). Withthe exception of Mo (average¼18 mg/L), the concentrations of V(average¼23.5 mg/L), and Cd (average¼0.14 mg/L) in groundwaterused for agricultural irrigation and drinking purposes on the farmsites were also not elevated (Shomar, unpublished data). We cantherefore attribute the elevated u-Mo concentration in the farmworkers to ingestion of Mo-contaminated groundwater and cropsirrigated with Mo-contaminated groundwater. These data suggestthat the water and dust–soil–crop pathways may not be critical indetermining the levels of the trace elements in the biomarkers atthe exposure doses for farm workers in Qatar.
Urinary Fe concentrations indicate that a portion of Qatar'sfarm worker population has low iron levels compared to generalpopulations. The lowest 10th percentile values for u-Fe in farmworkers (6.97 mg/L) are extremely low when compared to therange (16.29–114.95 mg/L) reported as normal by Rodriguez andDiaz (1995). The health implications of such low urinary iron loadfor some of Qatar’s farm workers deserve more attention.
Toenails likewise accumulate both toxic and essential elements(Table 4). The study population has low mean values of t-Cu and t-
N. Kuiper et al. / Environmental Research 131 (2014) 86–9488
Se compared to the general population (GP) values imputed bySlotnick and Nriagu (2006) suggesting that some of Qatar's farmworkers may be deficient in these elements. On the other hand,the concentrations of t-As, t-Cd, t-Pb and t-Ba, which are non-essential, are all below the imputed general population values andwell below the values reported for highly exposed populations(HEP) in other countries (Table 4). These results show that the
expected exposures from the participants' agricultural workand/or legacy exposures from their home country are not beingreflected in the trace element contents of toenails. Table 4also includes the concentrations of trace elements in toenailsamples from un-exposed Arab Americans in the metro-Detroit(Michigan) area (Slotnick et al., 2005). Except for t-Mn and t-Ba,the trace metal concentrations are surprisingly similar in the twopopulations.
Table 1Instrumental detection limits and standard reference material recovery values.
Element Detection limit (mg/L) NIST 1640a (mg/L) with 20-fold dilution NIST 1400 (mg/kg) NIST 1515 (mg/kg)
Expected value Measured value Expected value Measured value Expected value Measured value
a Less than instrumental detection limit.b N/A¼not available.c GP, HEP, and AAD¼General unexposed population, High Exposure Population,
and Arab Americans in Detroit Michigan (Slotnick et al., 2005).
N. Kuiper et al. / Environmental Research 131 (2014) 86–94 89
3.2. Correlations between trace metals in urine and toenails
Analysis of the data by means of Student's t-test showed nosignificant associations between any of the trace elements in urineand toenails. Our results are contrary to those of previous studies thathave reported significant correlations between several correspondingmetals in the two biomarkers. For arsenic, in fact, the concentrations inboth urine and toenail are generally used as predictors of arsenicexposure. Rivera-Nunez et al. (2012) reported strong association(r¼0.54; po0.0001) between As in urine and toenails in a group inMichigan that was exposed to slightly elevated levels of arsenic indrinking water. Similar results have also been reported by Karagaset al. (2001), Slotnick et al. (2007) and many others (reviewed byMarchiset-Ferlay et al. (2012)). The observation of toenail As as apredictor of urinary As has been used to support the claim that urinaryarsenic concentrations remain constant over long periods of time inpopulations with no changes in their drinking water supplies oractivity patterns (Navas-Acien et al., 2009). While this observationmay be valid for populations exposed to elevated levels of arsenic, ourresults for Qatar's farm workers show that this conclusions is notuniversally applicable and may not hold for people exposed to lowlevels of arsenic. The relationships of other metals in urine and toenailare less well researched than for arsenic and the results have generallybeen contradictory (Adair et al., 2006; Laohaudomchok et al., 2011;Slotnick et al., 2005).
3.3. Confounding variables and health status
The associations of urinary and toenail contents with potentialcofactors including age, educational attainment, nationality, dura-tion of farmworking in Qatar and BMI were assessed using partial-linear regression models (Table 5). Age was strongly associatedwith how long a participant had been a farm worker in Qatar(r¼0.660; pr0.001), and negatively with level of education(r¼�0.235; p r0.001). Except for level of education and workduration in Qatar (r¼�0.215; p¼0.001), no other associationswere found between these confounding variables.
Trace metals in urine and toenail were found to be poorpredictors of BMI and BDS (Table 5). The strongest associationwas between u-Cu and BMI. Recent studies have drawn attention tothe role of copper deficiency/excess on metabolic diseases. Limaet al. (2006) reported that the copper concentrations were sig-nificantly higher in the plasma of overweight and obese partici-pants of their study compared to the control group. Hatano et al.(1982) have suggested that excess weight associated with lipidmetabolism disorders might predispose individuals to changes inCu concentrations in plasma. Some studies, however, have notfound any association between copper in urine or blood and BMI(Błażewicz et al., 2013). Prenatal exposure to lead has been linked toobesity in males (Fox, 2008), and blood lead levels have beenassociated with obesity in infants, adolescents and adults(Scinicariello et al., 2013). Błażewicz et al. (2013) reported a strongnegative association between u-Fe and BMI (r¼0.79), but no suchcorrelation was found in this study.
We are not aware of previous efforts that have specificallyrelated the burden of disease symptoms (BDS) to metals in urineor toenails. Both u-Mn (2.2 mg/L) and u-Cd (1.1 mg/L) are in the highend of the typical range reported in other studies which mayaccount for their significant associations with the burden of diseasesymptoms (Table 5). The negative association between u-Se (anessential element) and BDS is inexplicable in view of the fact thatthe urinary Se levels do not appear to be abnormally low; it maybe a reflection of cross-talk in the metabolic cycle of Se. AlthoughMo does not play an important metabolic role in the human body, itis somewhat surprising that the high urinary Mo levels are notassociated with BMI or BDS.
Age is negatively associated with u-Pb, t-As and t-Cd (Table 5). Weinterpret these associations to mean that the farm workers have beenshedding their body burdens of these metals since beginning work inQatar. This inference is consistent with the negative associationbetween duration of work in Qatar and both t-As and t-Cd, as wellas between duration of work in Qatar and u-Pb (Table 5). Theassociation of u-Pb with age is most likely related to the fact thatmost of the farmworkers came from countries where lead in gasolinewas banned recently and the older workers had acquired higher bodyburdens of lead which are leaching into their urine. This is confirmedby the negative association between u-Pb and duration of farm-working in Qatar (Table 5). The weak negative association betweenduration of work and t-U is notable in the sense that we know littleabout human exposure to U in the environment.
Nationality was a predictor for several metals in the urine samplesincluding u-As (r¼�0.012; p¼0.069), u-Cd (r¼�0.182; p¼0.006),and u-Mo (r¼�0.163; p¼0.014) as well as with t-As (r¼�0.209;p¼0.001). These associations were rather interesting and we decidedto further explore the role of the country of origin as a predictor of thelevels of trace elements in urine and toenails of Qatar's farm workers(Table 6). We found that Indian farmworkers had approximately 650%higher u-As while Bangladeshi farm workers had approximately 256%higher u-As and 269% higher t-As than several other nationalitiesrepresented. Elevated As levels among these nationalities wereobserved across the farm sites. The fact that workers from India andBangladesh had the highest u-As and t-As reflects exposures asso-ciated with ingestion of food and water contaminated with arsenic intheir countries of origin (Halder et al., 2012).
3.4. Inter-element relationships in urine and toenails
Bivariate models were used to determine significant inter-element correlations in urine and toenail samples controlling forthe confounding variables, namely age, nationality, level of educa-tion and duration of farm-working in Qatar (Tables 7 and 8). Theinter-element associations that are significant are quite large.Element pairs in urine samples with correlation coefficients(r-values) above 0.4 include V–Cr, V–Mn, V–Fe, V–U, Fe–U, Co–Ni,Mo–Cd, Cd–Pb (Table 7). By contrast, the r-values for toenailsamples were above 0.4 only for three element pairs, Mn–U,Mn–Ba, and Cu–Pb (Table 8). The multiplicity of inter-elementrelationships suggests extensive interactions in the metabolic cyclesof the elements and are believed to be an important determinant oftheir concentrations in urine, but much less so in toenails.
Epidemiological studies now use urine and toenail very frequentlyas biomarkers because they are a less invasive and easy sample toobtain in large populations. Most often, urinary trace elements areused as an indicator of recent exposure (via ingestion or inhalation)
Table 5Significant results (po0.05) of elemental content and the relationship to BMI,burden of disease symptoms (BDS), age and duration of work in Qatar.
Parameter Factor R p-value
u-Cu BMI 0.169 0.010u-Pb BMI 0.142 0.032u-Cd BDS 0.164 0.013u-Mn BDS 0.145 0.028u-Pb BDS �0.142 0.031u-Se BDS �0.172 0.009t-As Age �0.154 0.018t-Cd Age �0.144 0.028u-Pb Age �0.186 0.004t-As Duration of work �0.162 0.013t-Cd Duration of work �0.141 0.031u-Pb Duration of work �0.195 0.003t-U Duration of work �0.127 0.053
N. Kuiper et al. / Environmental Research 131 (2014) 86–9490
because urine is presumed to be the main route of excretion of mosttrace metal species. Studies of industrial workers and populationsexposed to high levels have shown that there is a significantcorrelation between some urinary trace elements and measures oftheir exposure via ingestion, inhalation or dermal contact (see recentreview for arsenic by Marchiset-Ferlay et al. (2012)). This observationhas now been extended to low-level exposure of the general popula-tion, primarily for arsenic and lead; an extrapolation based on theassumption that the relationship between exposure dose and urinaryconcentration holds regardless of the route, level, and duration ofexposure. This assumption has not been properly validated for manytrace elements and the results of this study call such presumption intoquestion. Our study participants fall within a narrow age range(controlling for age effects) and are healthy (controlling for existinghealth conditions). Although the participants perform different taskson the farm, which may predispose them to different exposure risks,the levels of trace elements in their urine and toenails generally fall inthe range for un-exposed populations in other parts of the world(Tables 3 and 4). Since participants lived in the farms where theyworked, they were likely to consume the same foods and drink fromthe same water sources. One would therefore expect to see a narrow
spread in the urinary trace element levels in any given farm, whichwas clearly not the case. As noted previously, the 50th percentileparticipant had lived and worked on the same farm for 4.2 years. Sincethe toenails used in the study contain no more than two years ofexposure (internal and external), one would expect to see somecomplementary relationships between the trace metal contents ofparticipants' urine and toenail. No such thing was found.
Tables 9 and 10 show the respective distributions of traceelements in urine and toenail samples from the individual farms.The dispersion in the levels of trace elements in each biomarker isquite large in any given farm, as are the differences between thefarms. As noted previously, the associations between correspondingmetals in the urine and toenail samples are mostly insignificant.
4. Conclusions
The current low levels of exposures may not be the criticaldeterminant of the concentrations of trace metals in the urine andtoenail samples. The main factors appear to be (a) depuration pro-cesses involving the body burden of trace elements that the
Table 6Correlations between urinary and toenail As content and nationality.
N. Kuiper et al. / Environmental Research 131 (2014) 86–9492
participants brought with them to Qatar and (b) the metabolicprocesses and cross-talks between the metabolic cycles of the traceelements which seem to over-ride the exposure footprint. Weconclude that the concentrations and variations of trace elementsfound in toenail (especially) and urine samples need to be carefullyvalidated before they can be used as biomarkers of exposure in thegeneral population for many trace elements.
Acknowledgments
We would like to acknowledge Qatar University for theircontribution to sample collection and survey administration. Wespecifically thank Islam A. Qunnaby and Mohammad Fazle Rakibfor their involvement in each aspect of the project.
References
Adair, B.M., Hudgens, E.E., Schmitt, M.T., Calderon, R.L., Thomas, D.J., 2006. Totalarsenic concentrations in toenails quantified by two techniques provide auseful biomarker of chronic arsenic exposure in drinking water. Environ. Res.101, 213–220.
Aguilera, I., Daponte, A., Gil, F., Hernandez, A.F., Godoy, P., Pla, A., Ramos, J.L., 2008.Biomonitoring of urinary metals in a population living in the vicinity ofindustrial sources: a comparison with the general population of Andalusia,Spain. Sci. Total Environ. 407, 669–678.
Błażewicz, A., Klatka, M., Astel, A., Partyka, M., Kocjan, R., 2013. Differences in tracemetal concentrations (Co, Cu, Fe, Mn, Zn, Cd, and Ni) in whole blood, plasma,and urine of obese and nonobese children. Biol. Trace Elem. Res. 155, 190–200.
Calafat, A.M., 2012. The U.S. National Health and Nutrition Examination Survey andhuman exposure to environmental chemicals. Int. J. Hyg. Environ. Health 215,99–101.
Darwish, M.A., Mohtar, R., 2012. Qatar water challenges. Desalination Water Treat.52, 1–12, http://dx.doi.org/10.1080/19443994.2012.693582.
Fox, N., 2008. Prenatal exposure to lead linked to obesity in males. Science Daily. From⟨http://www.sciencedaily.com/releases/2008/02/080220141206.htm⟩ (accessedOctober 2013).
Halder, D., Bhowmich, S., Biswas, A., Mandal, U., Nriagu, J., Mzumdar, D.N.G.,Chatterjee, D., Bhattacharya, P., 2012. Consumption of brown rice: a potentialpathway for arsenic exposure in rural Bengal. Environ. Sci. Technol. 46, 4142–4148,http://dx.doi.org/10.1021/es204298a.
Hatano, S., Nishi, Y., Usui, T., 1982. Copper levels in plasma and erythrocytes inhealthy Japanese children and adults. Am. J. Clin. Nutr. 35, 120–126.
Heitland, P., Koster, H.D., 2006. Biomonitoring of 30 trace elements in urinechildren and adults by ICP-MS. Clin. Chim. Acta 365, 310–318.
Karagas, M.R., Le, C.X., Morris, S., Blum, J., Lu, X., Spate, V., Carey, M., Stannard, V.,Klaue, B., Tosteson, T.D., 2001. Markers of low level arsenic exposure forevaluating human cancer risks in a US population. Int. J. Occup. Med. Environ.Health 14, 171–175.
Laohaudomchok, W., Lin, X., Herrick, R.F., Fang, S.C., Cavallari, J.M., Christiani, D.C.,Weisskopf, M.G., 2011. Toenail, blood and urine as biomarkers of manganeseexposure. J. Occup. Environ. Med. 53, 506–510.
Lima, S.C., Arrais, R.F., Sales, C.H., Almeida, M.G., de Sena, K.C., Oliveira, V.T., deAndrade, A.S., Pedrosa, L.F., 2006. Assessment of copper and lipid profile inobese children and adolescents. Biol. Trace Elem. Res. 114, 19–29.
Marchiset-Ferlay, N., Savanovitch, C., Sauvant-Rochat, M.-P., 2012. What is the bestbiomarker to assess arsenic exposure via drinking water? Environ. Int. 39,150–171.
Table 10Trace metals in toenail samples from individual farms.
Navas-Acien, A., Umans, J.G., Howard, B.V., Goessler, W., Franscesconi, K.A.,Crainiceanu, C.M., Silbergeld, E., 2009. Urine arsenic concentrations and speciesexcretion patterns in American Indian communities over 10-year period: theStrong Heart Study. Environ. Health Perspect. 117, 1428–1433.
Nriagu, J.O., Smith, P., Socier, D., 2011. A rating scale for housing-based healthhazards. Sci. Total Environ. 409, 5423–5431.
Qatar National Food Security Programme (QNFSP) – The Issue. ⟨http://www.qnfsp.gov.qa/issue⟩ (accessed 6 September 2013).
Rivera-Nunez, Z., Meliker, J.R., Meeker, J.D., Slotnick, M.J., Nriagu, J.O., 2012. Urinaryarsenic species, toenail arsenic, and arsenic intake estimates in a Michiganpopulation with low levels of arsenic in drinking water. J. Expo. Sci. Env.Epidemiol. 22, 182–190.
Rodriguez, E., Diaz, R., 1995. Iron, copper and zinc levels in urine: relationships tovarious individual factors. J. Trace Elem. Med. Biol. 9, 200–209.
Scinicariello, F., Buser, M.C., Mevissen, M., Portier, C.J., 2013. Blood lead levelassociation with lower body weight in NHANES 1999–2006. Toxicol. Appl.Pharmacol. (Epub ahead of print, 2013 October 4. pii: S0041-008X(13)00422-5)
Slotnick, M.J., Meliker, J., AvRuskin, G.A., Ghosh, D., Nriagu, J.O., 2007. Toenails as abiomarker of inorganic arsenic intake from drinking water and foods. J. Toxicol.Environ. Health A 70, 148–158.
Slotnick, M.J., Nriagu, J.O., 2006. Validity of human nails as a biomarker of arsenicand selenium exposure: a review. Environ. Res. 102, 125–139.
Slotnick, M.J., Nriagu, J.O., Johnson, M.M., Linder, A.M., Savoie, K.L., Jamil, H.J.,Hammad, A.S., 2005. Profiles of trace elements in toenails of Arab-Americans inthe Detroit Area, Michigan. Biol. Trace Elem. Res. 107, 113–126.
The World Bank, 2013. Data – GDP Per Capita. ⟨http://data.worldbank.org/indicator/NY.GDP.PCAP.CD/countries⟩ (accessed 8 September 2013).
Wang, H.S., Sthiannopkao, S., Chen, Z.J., Man, Y.B., Du, J., Xing, G.H., Kim, K.W., Yasin,M.S.M., Hashin, J.H., Wong, M.H., 2013. Arsenic concentration in rice, fish, meatand vegetables in Cambodia: a preliminary risk assessment. Environ. Geochem.Health 35, 745–755.
WHO, 2013. World Health Organization (BMI Classification). ⟨http://apps.who.int/bmi/index.jsp?introPage=intro_3.html⟩ (accessed November 4, 2013).
N. Kuiper et al. / Environmental Research 131 (2014) 86–9494
