ARTICLE IN PRESS

Environmental Research 101 (2006) 213–220

www.elsevier.com/locate/envres

Total arsenic concentrations in toenails quantified by two techniques

provide a useful biomarker of chronic arsenic
exposure in drinking water$
Blakely M. Adaira,, Edward E. Hudgensb, Michael T. Schmittb,
Rebecca L. Calderonb, David J. Thomasa
a
Experimental Toxicology Division, National Health and Environmental Effects Research Laboratory, Office of Research and Development,
United States Environmental Protection Agency, Research Triangle Park, North Carolina, USA
b
Human Studies Division, National Health and Environmental Effects Research Laboratory, Office of Research and Development,
United States Environmental Protection Agency, Research Triangle Park, North Carolina, USA

Received 11 April 2005; received in revised form 2 August 2005; accepted 4 August 2005
Available online 26 September 2005

Abstract

Accurate quantitation of any contaminant of interest is critical for exposure assessment and metabolism studies that support risk
assessment. A preliminary step in an arsenic exposure assessment study in Nevada quantified total arsenic (TAs) concentrations in tissues
as biomarkers of exposure. Participants in this study (n ¼ 95) were at least 45 years old, had lived in the area for more than 20 years, and
were exposed to a wide range of arsenic concentrations in drinking water (3–2100 ppb). Concentrations of TAs in blood, urine, and
toenails determined by hydride generation-atomic fluorescence spectrometry (HG-AFS) ranged from below detection to 0.03, 0.76, and
12 ppm, respectively; TAs in blood rarely exceeded the limit of detection. For comparison, TAs in toenails determined by neutron
activation analysis (NAA) ranged from below detection to 16 ppm. Significant (Po0:0001) positive regressions were seen between the
TAs concentration in toenails and in drinking water (adjusted r2 ¼ 0:3557 HG-AFS, adjusted r2 ¼ 0:3922 NAA); TAs concentrations in
urine were not described by drinking water As (adjusted r2 ¼ 0:0170, P ¼ 0:1369). Analyses of TAs in toenails by HGAFS and NAA
yielded highly concordant estimates (r ¼ 0:7977, Po0:0001). These results suggest that toenails are a better biomarker of chronic As
exposure than urine in the current study, because the sequestration of As in toenails provides an integration of exposure over time that
does not occur in urine.
r 2005 Elsevier Inc. All rights reserved.

Keywords: Arsenic; Biomarker; Neutron activation analysis; Toenail; Hydride generation-atomic fluorescence spectroscopy

1. Introduction

$
This research was performed with intramural funding from the US
Environmental Protection Agency. Authors regarding the article sub-
mitted noted no potential for competing financial interests. The authors
agree with the contents and feel that the manuscript is ready for
publication. This study was reviewed and approved by the University of
North Carolina-Chapel Hill Institutional Review Board and EPA Human
Studies Official, Peter Preuss. Human subjects participating in the study
did so with informed consent.
Corresponding author. US Environmental Protection Agency, Phar-
macokinetics Branch—Mail Drop B 143-01, 109 Alexander Drive,
Research Triangle Park, North Carolina 27711, USA.
Fax: +1 919 541 1937.
E-mail address: adair.blakely@epa.gov (B.M. Adair).
Adverse health effects associated with chronic ingestion
of inorganic arsenic (iAs) in drinking water create
significant public health problems worldwide (Calderon et
al., 1999; Karim, 2000; Lin et al., 1998; Xia and Liu, 2004).
The origin of iAs in surface and ground water is usually
geological (Anawar et al., 2002; Meza et al., 2004).
Although some foods, such as rice, may be a source of
iAs exposure (Chakraborti et al., 2004), drinking water is
typically the main source. Organic arsenic species from
other food sources may contribute to total arsenic (TAs)
concentrations in biological samples such as urine (Ma and

0013-9351/$ - see front matter r 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.envres.2005.08.004
 ARTICLE IN PRESS
214 B.M. Adair et al. / Environmental Research 101 (2006) 213–220
Le, 1998; Sekhar et al., 2003). Estimating total exposure to
arsenic requires quantifying intake from all sources, such
as food and water. It has been suggested that total As
concentrations (TAs) in biological matrices (e.g., urine and
toenails) can be defined as biomarkers of exposure that
represent exposure from all sources and should accurately
and quantitatively reflect these exposures (Karagas et al.,
2001b).
Ideally biomarkers of exposure are alternatives to the
cumbersome and often inaccurate estimation of intake
from multiple sources. Biomarkers of exposure can also be
used to validate other methods of exposure assessment
(Garland et al., 1993). Several studies have evaluated
biomarkers of As exposure and have examined relation-
ships between these biomarkers and TAs in drinking water
(Calderon et al., 1999; Karagas et al., 2001a; Mandal et al.,
2003, 2004). Biomarkers of As exposure have been used in
several case-control studies (Beane Freeman et al., 2004;
Garland et al., 1993; Karagas et al., 2004).
Choosing an appropriate biomarker of exposure can be
difficult. Concentrations of TAs in urine or blood have
been used as biomarkers of As exposure from drinking
water (Calderon et al., 1999; Mandal et al., 2004; Meza et
al., 2004; Sekhar et al., 2003). However, the short residence
times of As in urine (3–4 days) and blood (2–3 h) ensure
that As concentrations in these biological fluids will be
strongly influenced by recent exposures (Karim, 2000;
Mandal et al., 2004). In contrast, As is sequestered in nails,
hair, and skin during their formation; therefore, the TAs
concentration in samples of the biological matrices
represents a time frame of exposure that may minimize
fluctuations from occasional dietary As sources (Anawar et
al., 2002; Garland et al., 1993; Hinwood et al., 2003). Nails,
hair, and skin can be contaminated through contact with
As-contaminated water or particles (Hindmarsh, 2002;
Hinwood et al., 2003; Mandal et al., 2003). Chemical hair
treatments, medications, and general health status can alter
growth rates and As distribution in nail, hair, and skin
(Daniel et al., 2004; Hindmarsh, 2002). Toenails may be
preferred for use in exposure assessment studies, because
they are exposed less extensively to outdoor air or water
than are fingernails, skin, and hair. This lower exposure
may reduce the risk of external As contamination (Das et
al., 1995; Karagas et al., 2000). Toenail bed lengths and
growth rates vary slightly among toes so that a pooled
sample of clippings from all nails on a foot reflects As
exposure from 2 to 12 months before collection (Garland et
al., 1993; Samanta et al., 2004). Appropriate cleaning,
preparation, and analysis techniques can minimize the risk
of altering TAs concentrations in toenails during proces-
sing (Agahian et al., 1990; Hindmarsh, 2002). With
minimal sample preparation and relatively low instrument
detection limits, neutron activation analysis (NAA) has
shown promise as an accurate method for TAs quantifica-
tion in nails (Henke et al., 1982; Nichols et al., 1998). Total
As concentrations in toenails have been consistent among
samples collected years apart, suggesting that toenails act
as long-term integrators of As exposure (Garland et al.,
1993; Karagas et al., 2001a). These characteristics indicate
that toenails may be an ideal biomarker for As exposure in
individuals chronically exposed to iAs.
This preliminary study examines the usefulness of
toenails as a biomarker of chronic exposure to iAs in
individuals who ingest drinking water contaminated with a
wide range of As concentrations. The TAs concentrations
in toenails, urine, blood, and water were compared to
determine the best indicator of exposure in the study
population. Preparation and analysis methods were opti-
mized for toenail analysis using hydride generation-atomic
fluorescence spectroscopy (HG-AFS), and the results from
HG-AFS were compared to analytical results from NAA
to determine the comparability and reliability of the two
techniques.
2. Materials and methods
2.1. Reagents
All reagents were of analytical grade or higher (EMD
Chemicals, Gibbstown, NJ). Samples and standards were
diluted with house de-ionized water. Analysis solutions
were made up in distilled deionized water (Glass Still,
Barnstead FI-streem, Dubuque, Iowa). Standards for
arsenic quantitation were dilutions of 1000 mg/mL arsenic
standard (Spex CertiPrep, Metuchen, NJ). Several certified
standard reference materials were used for QA/QC;
DORM-2, DOLT-3 (Institute for National Measurement
Standards, National Research Council, Canada), RM-50,
and SRM-2670 (National Institute of Standards and
Technology, USA). An internal standard of homogenized
nails fortified with As was also used. All glassware was
soaked in a 10% nitric acid bath overnight and rinsed
multiple times with de-ionized water.
2.2. Study design
Participants who provided samples resided in Churchill
County, Nevada, where the primary source of drinking
water was drawn from the Carson Sink alluvial aquifer.
Most of the 25,000 residents of the county lived within 20
miles of Fallon, the county seat. Participants using Fallon’s
municipal water supply were assigned a mean TAs
concentration in drinking water of 89 ppb based on
analysis of 50 samples collected by the Nevada State
Health Lab throughout the distribution system (8976 ppb,
mean7standard deviation). Water samples collected from
participants using well water had TAs concentrations
ranging from o3–2100 ppb. Participants were at least 45
years old and had lived in Churchill County for 20 or more
years. This study was reviewed and approved by the
University of North Carolina-Chapel Hill Institutional
Review Board. Consenting participants completed ques-
tionnaires to provide information on health history and
consumption rates of water and food. Urine, blood, and
 ARTICLE IN PRESS
B.M. Adair et al. / Environmental Research 101 (2006) 213–220
toenail samples from 95 participants were analyzed for TAs
using HG-AFS. A second toenail aliquot from each
participant was analyzed using NAA.
2.3. Sample collection, storage, and cleaning
Urine, blood, and participant information were collected
in the field clinic on the first visit. Due to time and staffing
constraints and for donor convenience donors collected
water and toenail samples at home for delivery the next day
at the follow-up visit to the field clinic. Field staff gave all
participants who did not use water from the Fallon
municipal water supply detailed instructions on collection
and storage along with 500 mL acid washed bottles that
were supplied by the Nevada State Health Laboratory. The
water filled bottles were acidified to pHo2 and stored at
room temperature for analysis using inductively coupled
plasma-mass spectroscopy (ICP-MS) or graphite furnace-
atomic absorption spectroscopy (GF-AAS) depending on
sample turbidity at the Nevada State Health Laboratory.
Spot urine samples were collected in polyethylene
centrifuge tubes, initially stored at 4 1C, and then frozen
to
80 1C until analysis. Blood was collected in 7 mL royal
blue top tubes with sodium citrate anticoagulant (BD
Biosciences, Palo Alto, CA) and stored at 4 1C until
analysis. Using new clippers cleaned with 70% ethanol,
1–3 mm of nail from each toe on one foot were cut and
stored at room temperature in Ziplocs bags until cleaning.
The toenail cleaning technique has been described (Das et
al., 1995). Briefly, toenails were sonicated for 10 min in
fresh distilled/de-ionized water two times. After an acetone
rinse, toenails were soaked in acetone for 5 min and dried
overnight at room temperature. One aliquot of toenails
(X0.025 g) was submitted for NAA, and the other aliquot
(0.01–0.1 g) was digested for HG-AFS analysis.
2.4. Total arsenic: ICP-MS or GF-AAS
Water samples with nephelometric turbidity unit
(NTU)p1 were processed according to Environmental
Protection Agency Method 200.8 (Creed et al., 1994) using
ICP-MS for analysis. Samples with high turbidity
(NTU41) were prepared using ASTM D2972 (ASTM
International, 1997), which incorporated GF-AAS detec-
tion to minimize interferences. The Nevada State Health
Laboratory is inspected and accredited by the US EPA
Region IX under the Safe Drinking Water Act (SDWA)
program and reports water analysis results for all commu-
nity systems in Nevada. Therefore, QA/QC procedures for
water analysis met the requirements of this study under
SDWA guidelines.
2.5. Total arsenic: HG-AFS
Samples were prepared for analysis with HG-AFS using
a modified method developed by Cox (1980). Concentrated
nitric acid (2–3 mL) and perchloric acid (0.5–1 mL) were
215
added to covered beakers (25 mL) with urine (1 mL), blood
(2 mL), or toenail (0.01–0.1 g) aliquots. After predigesting
at room temperature to dissolve most tissue, samples were
heated for 1 h (80 1C hotplate). Nitric acid (2 mL) was
added to cooled beakers, samples were heated at 115 1C to
reduce the volume to 1 mL. After sulfuric acid (2 mL) was
added, closed beakers were slowly heated from 80 to 280 1C
(with chartreuse color indicator). Beakers were removed
from the hotplate 20 min after the solution changed from
chartreuse to clear. Samples were quantitatively transferred
to polypropylene centrifuge tubes and brought to a final
volume of 10 mL with HCl (2.1 M). Digests were stored at
4 1C for analysis the following day.
The concentration of TAs in each matrix was determined
using a Varian (Sugarland, TX) inert quaternary pump
(9012) with auto sampler (Prostar 410), which was
connected to PSAnalytical Millennium Excalibur HG-
AFS (PSAnalytical, Deerfield Beach, FL). Briefly, 100 mL
of sample was injected into a water stream (3 mL/min) that
mixed with HCl (1.9 mL/min, 5% v/v), then NaBH4
(1.9 mL/min 1.4% w/v in 0.1 M NaOH) and was pumped
into a gas/liquid separator. Arsines generated by the
NaBH4 reaction are flushed from the gas/liquid separator
in an Ar carrier stream (230 mL/min) to the fluorescence
detector. The TAs recovery of QA/QC samples analyzed
with sample batches using the optimized digestion and
analysis procedure was 92718% for fortified samples
(n ¼ 19, mean7standard deviation) and 108723% for
SRM 2670 (n ¼ 9).
2.6. Total arsenic: NAA
The TAs concentration in toenails was determined by
NAA as described previously (Heydorn, 1984) at the
Nuclear Services Facility, Department of Nuclear Engi-
neering, North Carolina State University. Briefly, toenail
samples were bombarded with g-rays at a maximum flux of
1  1013 N/cm2 s with a 1 MW Pulstar Nuclear Reactor.
The decay of the nuclide produced was measured at
559.1 KeV using a germanium detector with an efficiency of
38–42%. The data were processed using Genie 2000
spectroscopy software (Canberra, USA). The percentage
of As recovered from three standard reference materials
(DORM-2, DOLT-3, and RM-50) analyzed as QA/QC
controls was 97.373% (n ¼ 12).
2.7. Statistical analysis
Samples with TAs concentrations that were below the
method detection limit (MDL, Table 1) were not used in
statistical analyses. TAs concentrations in biological
matrices were log-transformed for regression and correla-
tion analyses to meet equal variance and normality
assumptions. All linear regressions were performed using
Sigma Plot (v 8.0, SPSS Inc., Chicago, IL), and r2 values
were adjusted for the number of samples. All correlations
were performed in Matlab (v5.2, The Mathworks Inc.,
 ARTICLE IN PRESS
216 B.M. Adair et al. / Environmental Research 101 (2006) 213–220
Table 1
Summary of total As detection limits and distribution in samples collected from residents in Churchill Co., NV
Sample matrix Sample number Analysis methoda
Urine 95 AFS
Blood 25 AFS
Nail 95 AFS
Nail 95 NAA
Water 95 AAS
a
AFS—atomic fluorescence spectroscopy, NAA—neutron activation analysis.
b
MDL—method detection limit in ppm as calculated with instrument detection limit and the sample amount in parentheses.
Natick, MA). Linear regression analysis of TAs in water
vs. TAs in biological matrices was performed using the
detectable data collected from participants who consumed
untreated plain tap water (np75). Correlations among TAs
concentrations in biomarkers were performed with data
collected from all participants (np95) regardless of water
treatment.
Results of the food consumption survey indicated that 31
participants consumed fish or seafood within 48 h of
donating urine samples. The impact of fish or seafood
consumption on regressions of TAs in water vs. TAs in
biological matrices was tested by removing data from the
26 participants who drank tap water and consumed fish or
seafood within 48 h of biological sample collection. The
same influence on correlations among TAs in biological
matrices was tested using data from the 64 participants
who did not consume fish or seafood within 48 h of
biological sample collection.
3. Results
For each sample matrix, the MDLs, concentration
ranges of TAs detected, and the numbers of samples with
TAs concentrations greater than the MDL are presented in
Table 1. Arsenic was detected in 93.7% of urine samples at
concentrations ranging from 0.01 to 0.76 ppm. The TAs
concentrations in blood (0.01–0.03 ppm) were lower than
those in urine, with TAs in 80% of blood samples below
the MDL. Therefore, only a subset of blood samples was
analyzed and no correlations were performed. There were
similar TAs concentration ranges between toenails ana-
lyzed with HG-AFS (0.3–11.6 ppm) and NAA
(0.09–16.3 ppm). However, As was detected in fewer
toenail samples analyzed with HG-AFS (81.1%) than with
NAA (99.0%). As expected, most water samples (96.8%)
contained detectable arsenic concentrations.
The utility of TAs in urine and toenails as biomarkers of
exposure from drinking water was tested using linear
regression analysis. As shown in Fig. 1A, there was no
significant correlation between TAs in urine and TAs in
water (adj r2 ¼ 0:0170, P ¼ 0:1369). Neither normalizing
the TAs concentration in urine for creatinine concentration
(data not shown), nor omitting data from participants who
Representative MDL (ppm) Samples above MDL n Range of detectable
(sample amount)b (%) As (ppm)
0.012 (1 mL) 89 (93.7) 0.0112–0.759
0.006 (2 mL) 5 (20.0) 0.005–0.032
0.500 (0.025 g) 77 (81.1) 0.302–11.57
0.08 (0.025 g) 94 (98.9) 0.0868–16.3
0.003 (1 mL) 92 (96.8) 0.009–2.1
consumed fish or seafood within 48 h of sample collection
improved the correlation between TAs in drinking water
and urine (adj r2 ¼ 0:049, P ¼ 0:072). TAs in toenails
determined by HG-AFS or NAA (Fig. 1B and C) had
better correlations with TAs in water (Po0:0001, adj r2 ¼
0:3557 or 0.3922, respectively) than was observed between
TAs in urine and TAs in water.
Correlations were performed to compare TAs concen-
trations in urine and toenails, while also examining the
precision of the analytical techniques. For analyses
performed by HG-AFS, there was no correlation between
TAs in toenails and urinary TAs (Fig. 2A, r ¼ 0:1377,
P ¼ 0:25). However, TAs in toenails from NAA did
correlate with TAs in urine analyzed with HG-AFS (Fig.
2B, r ¼ 0:3006, Po0:0001). There was no significant
difference in correlations between TAs in urine and TAs
in toenails upon removal of data from participants
consuming seafood or fish within 48 h of sampling from
the correlations (Figs. 2A and B). Toenail TAs concentra-
tions detected using HG-AFS and NAA were in con-
cordance (Fig. 2C, r ¼ 0:7977, Po0:0001).
4. Discussion
The present study examined biomarkers of TAs exposure
and correlated TAs concentrations in these biomarkers
with the concentration of As in drinking water, the
presumptive main source of exposure to this metalloid. In
this sample from Churchill County, Nevada, we did not
find a statistically significant correlation between the TAs
concentrations in urine and drinking water. This lack of
correlation stands in contrast to the findings of other
studies where TAs concentrations in urine and water were
correlated (Calderon et al., 1999; Das et al., 1995; Meza et
al., 2004; Sekhar et al., 2003). Several factors could account
for the lack of correlation between the variables in the
present study. First, the older population (445 years old)
studied and underlying health issues that apply to older
populations could affect the relationship between TAs in
urine and As in drinking water. Although one study in
Utah which found a correlation between these variables
(Calderon et al., 1999) from individuals ranging in age
from 6 to 83 years, the age distribution was clustered in the
 ARTICLE IN PRESS
B.M. Adair et al. / Environmental Research 101 (2006) 213–220 217

1 10

Toenail ppm As, AFS

0.1
Urine ppm As

0.01
n = 47 n = 40
adj r 2 = 0.049, p = 0.0725 n = 47 adj r 2 = 0.4994, p < 0.0001 n = 40
n = 26 n = 21
adj r 2 = 0.0170, p = 0.1369 n = 73 adj r 2 = 0.3557, p < 0.0001 n = 61
0.001 0.1
0.0 0.2 0.4 0.6 0.8 0.0 0.2 0.4 0.6 0.8
(A) Water ppm As (B) Water ppm As

100
Toenail ppm As, NAA

10

n = 49
0.1
adj r 2 = 0.5222, p < 0.0001 n = 49
n = 26
adj r 2 = 0.3922, p < 0.0001 n = 75

0.01
0.0 0.2 0.4 0.6 0.8

(C) Water ppm As

Fig. 1. Regression in log scale of total As concentrations (TAs) in water vs. TAs in urine (A), nails analyzed with AFS (B), and nails analyzed with NAA
(C) collected from adults (445 years old) inhabiting Churchill Co., Nevada for 420 years and drinking untreated plain tap water. Symbols: open circles
represent participants who did not consume fish or seafood within 48 h of sample collection. Dotted line represents the line of best fit for open circle
participants. Solid circles represent participants who consumed fish or seafood within 48 h of sample collection. Solid line represents the line of best fit for
all participants with detectable TAs concentrations.

middle with few senior citizens to impact the correlations. is more indicative of acute recent exposures than long-term
Age may have influenced the results of creatinine corrected exposures (Hindmarsh, 2002). Like TAs in urine, TAs in
data. A continuous age-dependent decline in urinary blood also proved to be a poor biomarker of iAs exposure
creatinine clearance begins by 30 years of age in about in the current study. Because the TAs concentration in
two-thirds of the population; thus, dividing urinary TAs blood was about an order of magnitude less than that in
concentration by creatinine concentration may not normal- urine, there were too few samples with detectable
ize for the use of spot urine samples in older people (Beers, concentrations of TAs in blood for correlation analyses.
2004). Second, studies have shown that nontoxic orga- In contrast, blood collected from people exposed to As
noarsenicals from seafood consumption may contribute to from industrial contamination in India had TAs concen-
the elevated TAs concentrations in urine that would not be trations that were 100 times higher than the Churchill
accounted for by variation in the As concentrations in County residents with similar urinary TAs concentrations
water supplies (Ma and Le, 1998). However, in the current (Sekhar et al., 2003). The higher ratio of TAs in blood to
study, urinary TAs from participants who consumed fish or urine in India could reflect higher exposure to As among
seafood before urine collection was not different from Indian subjects or a consequence of other environmental
urinary TAs in other patients as demonstrated by similar factors (e.g., poor nutritional status).
regressions with and without data from fish/seafood eaters. The relationship between the TAs concentrations in
Blood is another common matrix for a biomarker of toenails and drinking water in this study was more
exposure for many contaminants of concern including As informative than for other biomarkers examined. Total
(Gerhardsson and Skerfving, 1996). Arsenic is cleared from As in toenails was positively and significantly correlated
human blood in less than 24 hours; therefore, TAs in blood with TAs in drinking water. Other studies have found
 ARTICLE IN PRESS
218 B.M. Adair et al. / Environmental Research 101 (2006) 213–220

1 1
Urine ppm As

0.1 0.1

Urine ppm As
0.01 0.01

n = 48 n = 57
r = 0.1098, p = 0.4576 n = 48 r = 0.3634, p = 0.0055 n = 57
n = 25 n = 30
r = 0.1377, p = 0.2500 n = 73 r = 0.3006, p < 0.0001 n = 87
0.001 0.001
0.01 0.1 1 10 100 0.01 0.1 1 10 100

(A) Nail ppm As, AFS (B) Nail ppm As, NAA

100

10
Nail ppm As, AFS

0.1 n = 51
r = 0.8429, p < 0.0001 n = 51
n = 25
r = 0.7977, p < 0.0001 n = 76
0.01
0.01 0.1 1 10 100

(C) Nail ppm As, NAA

Fig. 2. Correlations in log scale among TAs biomarkers collected from adults (445 years old) inhabiting Churchill Co., Nevada for 420 years and
drinking treated and untreated tap water. (A) Nails analyzed with AFS vs. urine, (B) nails analyzed with NAA vs. urine, and (C) nails analyzed with AFS
vs. NAA. Symbols and lines defined in Fig. 1 legend.

similar relationships between TAs concentrations in nails concentrations precluded the use of this data to confirm the
and in water (Hinwood et al., 2003; Karagas et al., 2000; accuracy of toenail TAs concentrations. Correlations
Mandal et al., 2003, 2004; Sekhar et al., 2003). Since among toenail and urinary TAs concentrations were poor.
pooled nail samples in the present Fallon study represent The difference in exposure dates represented by TAs in
1–2 months of growth that started up to a year ago (Daniel urine and toenails could also contribute poor correlations
and Scher, 1985; Henke et al., 1982), there was no impact between the two biomarkers. But the results implicate
from seafood or fish consumption within 48 h of sampling urinary TAs concentrations as a less informative biomar-
on correlations between TAs in toenails and water. One ker in the current study, because it did not correlate with
study demonstrated that TAs concentrations of 2 mm TAs in water or toenails.
toenail segments collected from proximal to distal end of a In the present study, other facets of toenails as a matrix
big toenail would produced a trend of TAs accumulation, for biomarkers of exposure were also examined, including
but in smaller segments (0.5 mm) fluctuations of TAs were sample collection, processing, and sensitivity of TAs
seen that correlated with hospital visits for arsenical analysis by HG-AFS and NAA. External contamination
poisoning symptoms (Henke et al., 1982). Similarly, TAs from particulate deposition and absorption must be
concentrations in toenails from Fallon residents would not considered in toenail sample collection and processing.
be impacted by periodic seafood consumption, thereby Studies have demonstrated that particulate matter can be
producing significant correlations between TAs in toenails removed from toenails with dilute detergent or weak
and water. solvents without removing bound As (Chen et al., 1999;
The accuracy of new biomarkers, like toenail TAs, can Hindmarsh, 2002). Therefore, toenails used in this study
be tested with correlations between new and validated were subjected to sonication with water and acetone and
biomarkers (Garland et al., 1993). Unfortunately, low TAs drying at room temperature before analysis by either HG-
concentrations in blood and high variability in urinary TAs AFS or NAA. Differentiation between exogenous As
 ARTICLE IN PRESS
B.M. Adair et al. / Environmental Research 101 (2006) 213–220
bound to hair or nails and depurated As in hair and nails is
a persistent analytical issue, but the results of in vitro
studies demonstrated that the absorption of As by nails
(o3% TAs absorbed) is lower than that by hair (9–16%
iAs absorbed) when soaked in As fortified solutions
(Karagas et al., 2000; Mandal et al., 2004). Because nails
absorb smaller amounts of As than hair, and toenails are
exposed to fewer sources of contamination than are
fingernails, the risk of exogenous interference in toenail
analysis is probably minimal.
Cleaned toenail samples were used to compare the utility
of HG-AFS and NAA for quantifying TAs in toenails.
Although both analytical techniques were satisfactory,
NAA appeared to be the most useful method for analysis
of As in toenail samples. For example, TAs determination
by AFS, AAS, or ICP-MS requires a multi-step acid
digestion of samples that can take days to complete (Beane
Freeman et al., 2004; Chen et al., 1999; Mandal et al., 2003;
Samanta et al., 2004). There is also a risk of As
contamination from reagents and volatilization of As by
excessive heat with acid digestion. Matrix interferences
from toenails and digestion reagents may affect spectro-
scopic techniques, particularly for MS (Chen et al., 1999).
In contrast, sample preparation for NAA involves little
manipulation beyond cleaning which minimizes the risk of
As contamination or loss (Chen et al., 1999; Garland et al.,
1993; Nichols et al., 1998). In the current study, the
detection limit of NAA (0.08 mg As/g nail) was lower than
that of HG-AFS (0.305 mg As/g nail). By comparison,
other studies have reported detection limits of 0.001 mg/g
nail—0.005 mg/g for 25 mg nail samples with NAA
(Garland et al., 1993; Karagas et al., 2000) and of
0.007–2 mg/g for 20 mg nail samples with other spectro-
scopy methods (Chen et al., 1999; Hinwood et al., 2003).
Comparison of quality control and assurance results for
HG-AFS in the current study and other spectroscopic
techniques found HG-AFS analysis to be slightly more
variable than other methods. Duplicate analyses of nail
samples were within 15%, and other spectroscopic methods
showed CVs less than the suggested limit of 20% (Chen et
al., 1999; Hinwood et al., 2003). In the present study, the
CV for NAA was 3%, which compared favorably to the
5–8% CVs reported in other NAA studies (Das et al., 1995;
Karagas et al., 2000). The higher accuracy and precision in
toenail TAs concentrations from NAA compared to HG-
AFS could account for the slightly stronger correlations
with TAs concentrations in water and the significant
correlation with TAs in urine.
The results of the present study and other studies suggest
that NAA is the optimal method for quantitation of TAs in
nails. There are some limitations to the use of NAA in
arsenic studies. As with all analytical methods, sample size
must be sufficient to account for As variability within and
between nails or hairs (Hindmarsh, 2002; Nichols et al.,
1998). Despite its high sensitivity neither the chemical
species of As nor the oxidation state of As present in
samples can be identified by NAA. Finally, facilities for
219
NAA are rare, potentially limiting access to the analytical
method. Similarities among methods of toenail TAs
concentrations and of regressions between TAs concentra-
tions in toenails and water, demonstrated that reliable
results could be obtained with HG-AFS if NAA is
unavailable.
In conclusion, TAs in toenails is a reliable biomarker of
As exposure, which correlates with As concentrations in
drinking water. Toenails are easier to collect and store than
other biomarkers, and there is a lower risk of external
contamination with toenails than with hair and fingernails.
Two studies also found good reproducibility between
toenail TAs concentrations collected from patients 3–6
years apart suggesting that TAs in nails may be represen-
tative of long-term exposure (Garland et al., 1993; Karagas
et al., 2001a). A comparison of TAs determination by
NAA and HG-AFS in the current study indicates NAA is
the preferred analytical technique, although useful data can
be obtained by HG-AFS. Further studies of TAs in
toenails using NAA should be evaluated with different
exposure scenarios to verify the accuracy of TAs in toenails
as a biomarker of exposure in such circumstances.
Acknowledgments
We thank Karen Herbin-Davis, field team, and Scott
Rhoney (US EPA), S. Lassell (North Carolina State
University Nuclear Facility), Nevada State Health Lab
and Guarding Our Local Drinking Water Recruiters
(University of Nevada, Reno: Cooperative Extensions),
Churchill Community Hospital, Fallon, NV, and study
participants for their efforts.
Disclaimer: This article has been reviewed and approved
for publication in accordance with National Health and
Environmental Effects Research Laboratory, US Environ-
mental Protection Agency. Approval does not signify that
contents reflect agency views. Mention of trade names or
commercial products does not constitute endorsement or
recommendation for use.
References
Agahian, B., Lee, J.S., Nelson, J.H., Johns, R.E., 1990. Arsenic levels in
fingernails as a biological indicator of exposure to arsenic. Am. Ind.
Hyg. Assoc. J. 51, 646–651.
Anawar, H.M., Akai, J., Mostofa, K.M., Safiullah, S., Tareq, S.M., 2002.
Arsenic poisoning in groundwater: health risk and geochemical sources
in Bangladesh. Environ. Int. 27, 597–604.
ASTM International, 1997. ASTM D 2972, Standard Test Methods for
Arsenic in Water.
Beane Freeman, L.E., Dennis, L.K., Lynch, C.F., Thorne, P.S., Just, C.L.,
2004. Toenail arsenic content and cutaneous melanoma in Iowa. Am.
J. Epidemiol. 160, 679–687.
Beers, M.H. (Ed.), 2004. The Merck Manual of Geriatrics: Chapter 6,
Clinical Pharmacology, Merck and Co., Inc., Medical Services,
USMEDSA, USHH, City, http://www.merck.com/mrkshared/
mm_geriatrics/home.jsp, (accessed January 2005).
Calderon, R.L., Hudgens, E.E., Le, C.X., Schreinemachers, D., Thomas,
D.J., 1999. Excretion of arsenic in urine as a function of exposure to
arsenic in drinking water. Environ. Health Perspect. 107, 663–667.
 ARTICLE IN PRESS
220 B.M. Adair et al. / Environmental Research 101 (2006) 213–220
Chakraborti, D., Sengupta, M.K., Rahman, M.M., Ahamed, S.,
Chowdhury, U.K., Hossain, M.A., 2004. Groundwater arsenic
contamination and its health effects in the Ganga-Meghna-Brahma-
putra Plain. J. Environ. Monit. 6, 74N–83N.
Chen, K.L., Amarasiriwardena, C.J., Christiani, D.C., 1999. Determina-
tion of total arsenic concentrations in nails by inductively coupled
plasma mass spectrometry. Biol. Trace Elem. Res. 67, 109–125.
Cox, D.H., 1980. Arsine evolution-electrochemical atomic absorption
method for the determination of nanogram levels of total arsenic in
urine and water. J. Anal. Toxicol. 4, 207–211.
Creed, J.T., Brockoff, C.A., Martin, T.D., 1994. Methods for the
determination of metals in environmental samples: supplement 1
(EPA/600/R-94/111,200.8). US Environmental Protection Agency,
Washington, DC, pp. 1–58.
Daniel 3rd, C.R., Scher, R.K., 1985. Nail changes caused by systemic
drugs or ingestants. Dermat. Clin. 3, 491–500.
Daniel 3rd, C.R., Piraccini, B.M., Tosti, A., 2004. The nail and hair in
forensic science. J. Am. Acad. Dermatol. 50, 258–261.
Das, D., Chatterjee, A., Mandal, B.K., Samanta, G., Chakraborti, D.,
Chanda, B., 1995. Arsenic in ground water in six districts of West
bengal, India: the biggest arsenic calamity in the world. Part 2. Arsenic
concentration in drinking water, hair, nails, urine, skin-scale and liver
tissue (biopsy) of the affected people. Analyst 120, 917–924.
Garland, M., Morris, J.S., Rosner, B.A., Stampfer, M.J., Spate, V.L.,
Baskett, C.J., et al., 1993. Toenail trace element levels as biomarkers:
reproducibility over a 6-year period. Cancer Epidemiol. Biomarkers
Prev. 2, 493–497.
Gerhardsson, L., Skerfving, S., 1996. Concepts on biological markers and
biomonitoring for metal toxicity. In: Chang, L.W. (Ed.), Toxicology of
Metals. CRC Press/Lewis Publishers, Boca Raton, New York,
London, Tokyo, pp. 81–110.
Henke, G., Nucci, A., Queiroz, L.S., 1982. Detection of repeated arsenical
poisoning by neutron activation analysis of foot nail segments. Arch.
Toxicol. 50, 125–131.
Heydorn, K., 1984. Neutron Activation Analysis for Clinical Trace
Element Research. CRC Press, Boca Raton, FL.
Hindmarsh, J.T., 2002. Caveats in hair analysis in chronic arsenic
poisoning. Clin. Biochem. 35, 1–11.
Hinwood, A.L., Sim, M.R., Jolley, D., de Klerk, N., Bastone, E.B.,
Gerostamoulos, J., et al., 2003. Hair and toenail arsenic concentrations
of residents living in areas with high environmental arsenic concentra-
tions. Environ. Health Perspect. 111, 187–193.
Karagas, M.R., Tosteson, T.D., Blum, J., Klaue, B., Weiss, J.E.,
Stannard, V., et al., 2000. Measurement of low levels of arsenic
exposure: a comparison of water and toenail concentrations. Am. J.
Epidemiol. 152, 84–90.
Karagas, M.R., Le, C.X., Morris, S., Blum, J., Lu, X., Spate, V., et al.,
2001a. Markers of low level arsenic exposure for evaluating human
cancer risks in a US population. Int. J. Occup. Med. Environ. Health
14, 171–175.
Karagas, M.R., Stukel, T.A., Morris, J.S., Tosteson, T.D., Weiss, J.E.,
Spencer, S.K., et al., 2001b. Skin cancer risk in relation to toenail
arsenic concentrations in a US population-based case-control study.
Am. J. Epidemiol. 153, 559–565.
Karagas, M.R., Tosteson, T.D., Morris, S.J., Demidenko, E., Mott, L.A.,
Heaney, J., et al., 2004. Incidence of transitional cell carcinoma of the
bladder and arsenic exposure in New Hampshire. Cancer Cause
Control 15, 465–472.
Karim, M.M., 2000. Arsenic in groundwater and health problems in
Bangladesh. Water Res. 34, 304–310.
Lin, T.H., Huang, Y.L., Wang, M.Y., 1998. Arsenic species in drinking
water, hair, fingernails, and urine of patients with blackfoot disease.
J. Toxicol. Environ. Health A 53, 85–93.
Ma, M., Le, C.X., 1998. Effect of arsenosugar ingestion on urinary arsenic
speciation. Clin. Chem. 44, 539–550.
Mandal, B.K., Ogra, Y., Suzuki, K.T., 2003. Speciation of arsenic in
human nail and hair from arsenic-affected area by HPLC-inductively
coupled argon plasma mass spectrometry. Toxicol. Appl. Pharmacol.
189, 73–83.
Mandal, B.K., Ogra, Y., Anzai, K., Suzuki, K.T., 2004. Speciation
of arsenic in biological samples. Toxicol. Appl. Pharmacol. 198,
307–318.
Meza, M.M., Kopplin, M.J., Burgess, J.L., Gandolfi, A.J., 2004. Arsenic
drinking water exposure and urinary excretion among adults in the
Yaqui Valley, Sonora, Mexico. Environ. Res. 96, 119–126.
Nichols, T.A., Morris, J.S., Mason, M.M., Spate, V.L., Baskett, C.K.,
Cheng, T.P., et al., 1998. The study of human nails as an intake
monitor for arsenic using neutron activation analysis. J. Radioanal.
Nucl. Chem. 236, 51–56.
Samanta, G., Sharma, R., Roychowdhury, T., Chakraborti, D., 2004.
Arsenic and other elements in hair, nails, and skin-scales of arsenic
victims in West Bengal, India. Sci. Tot. Environ. 326, 33–47.
Sekhar, K.C., Chary, N.S., Kamala, C.T., Rao, J.V., Balaram, V.,
Anjaneyulu, Y., 2003. Risk assessment and pathway study of arsenic in
industrially contaminated sites of Hyderabad: a case study. Environ.
Int. 29, 601–611.
Xia, Y., Liu, J., 2004. An overview on chronic arsenism via drinking water
in PR China. Toxicology 198, 25–29.