Page 1
IJOMEH 2010;23(1) 21
O R I G I N A L P A P E R S
International Journal of Occupational Medicine and Environmental Health 2010;23(1):21 – 26
DOI 10.2478/v10001-010-0003-x
UNMETABOLIZED VOCs IN URINE AS BIOMARKERS
OF LOW LEVEL OCCUPATIONAL EXPOSURE
BEATA JANASIK, MAREK JAKUBOWSKI, WIKTOR WESOŁOWSKI, and MAŁGORZATA KUCHARSKA
Nofer Institute of Occupational Medicine, Łódź, Poland
Department of Chemical Hazard
Abstract
Objectives: To compare the usefulness of determining unchanged forms of volatile organic compounds (VOCs), namely
toluene (TOL), ethylbenzene (EB) and xylene (XYL), in urine with the effectiveness of the already used biomarkers of
occupational exposure. Materials and Methods: Surveys were conducted in two workplaces (paint factory and footwear
factory). In total, 65 subjects participated in the study. Air samples were collected using individual samplers during work
shift. Urine and blood samples were collected at the end of work shift. Urine samples were analyzed for unchanged com-
pounds and selected metabolites, while blood samples were tested for unchanged compounds. VOCs in blood and urine
were determined by solid phase microextraction gas chromatography (SPME-GC-MS). Results: In the paint factory, the
geometric mean (GM) concentrations of VOCs in the air ranged as follows: 0.2–4.7 mg/m3 for TOL, 0.4–40.9 mg/m3 for EB
and 0.1–122.6 mg/m3 for XYL. In the footwear factory, the GM concentration of TOL in the air amounted to 105.4 mg/m3.
A significant correlation (p < 0.05) was observed between VOCs in blood, urine and air. The regression analyses performed
for paint factory workers showed that TOL-U and TOL-B were better biomarkers of exposure (r = 0.72 and r = 0.81) than
benzoic acid (r = 0.12) or o-cresol (r = 0.55). Conclusion: The findings of the study point out that the concentration of
unchanged VOCs in urine can be a reliable biological indicator of low level occupational exposure to these compounds.
Key words:
Volatile organic compounds, Urinanalysis, Blood analysis, Biological monitoring, Occupational exposure
Received: November 23, 2009. Accepted: February 2, 2010.
Address reprint request to B. Janasik, Department of Biological Monitoring, Nofer Institute of Occupational Medicine, św. Teresy 8, 91-348 Łódź, Poland
(e-mail: beatajan@imp.lodz.pl).
INTRODUCTION
Volatile organic compounds such as toluene (TOL),
ethylbenzene (EB) or xylene (XYL) are the popular
components of organic solvents that are widely used in
industries. For many years, biological monitoring of oc-
cupational exposure to volatile organic compounds was
based on the determination of their specific metabolites
in urine. However, in some cases, the normal physi-
ological processes or digestion of food additives may
modify urinary excretion of the metabolites and there-
by affect the specificity and sensitivity of this method.
Furthermore, the determination of urinary metabolites
is reliable in the cases involving exposure to a single
compound, and VOCs are almost as a rule present as
mixtures in the occupational setting. The method of si-
multaneous determination of different VOCs in blood
or exhaled air, which was proposed in the past, has not
gained wide acceptance because it was invasive and the
sampling was difficult. Attempts to apply the determina-
tion of unchanged VOCs in urine to assess occupational
exposure began over twenty years ago, but this approach
was rather limited due to the low rates of urinary ex-
cretion of VOCs and the low sensitivity of the deter-
mination methods. With the improvement of analytical
methods, the determination of unmetabolized VOCs in
urine has gained interest anew. Moreover, it has been
assumed that the influence of the kinetics of VOCs elim-
ination in urine on the results will be much lower than
for VOCs determinations in blood, where the half-life
during phase I of elimination is as short as 3–5 min [1].
The kinetics of urinary VOCs elimination complies with
an open two-compartment model. The half-time values
Page 2
O R I G I N A L P A P E R S B. JANASIK ET AL.
IJOMEH 2010;23(1)22
Analytical methods
After desorption from coconut shell charcoal, airborne
VOCs were determined by GC-MS analysis. The analy-
sis was performed with 6890N gas chromatograph (Agi-
lent Technologies) equipped with HP 5973 mass detec-
tor, split-splitless injector and HP-PONA column (50 m
length, 0.2 mm ID, 0.5 μm film thickness).
VOCs in blood and urine were determined by solid phase
microextraction gas chromatography (SPME-GC-MS).
The analyses were carried out on HP 6890 gas chromato-
graph with HP 5973 mass detector, split injector, and cap-
illary column (HP-INNOWAX) using the method previ-
ously described by Fustinioni et al. [12].
Metabolites in urine were determined by gas chromatog-
raphy (Hewlett-Packard 5890 Series II Plus, GC column
HP-5, 50 m × 0.32 mm × 1.05 μm) using the method de-
scribed by Janasik et al. [2].
Statistical analysis
Statistical analysis was performed using Statistica
StatSoft®Polska software package. Linear regression anal-
ysis was used to estimate the slope and intercepts of the
relationship between variables. The P value of 0.05 was
considered statistically significant. Air and urinary con-
centrations below the level of quantification were replaced
with the values equal to the level of quantification divided
by two.
RESULTS
The geometric mean concentration for toluene, xylene
and ethylbenzene in the air in the paint and footwear
factories are presented in Table 1. The concentrations in
occupational setting were generally low, with the excep-
tion of toluene concentrations in the footwear factory that
were close to the value of occupational exposure limit
valid currently in Poland (100 mg/m3). The results of de-
terminations of unchanged VOCs in urine and blood and
their metabolites in urine collected from workers at both
workplaces are summarized in Table 2. The data on the
relationship between VOC concentrations in the air and
concentrations of unchanged compounds in blood and
for toluene, ethylbenzene, xylene and mesitylene varied
from 0.45 h to 0.88 h for phase I and from 6.7 h to 19.2 h
for phase II [2].
The data reported recently for exposure in occupational
setting show that the determination of unmetabolized sol-
vents in urine provides a highly sensitive and specific index
of exposure to VOCs [3–11]. The purpose of the present
study was to compare the validity of various biomarkers of
exposure to toluene, ethylbenzene and xylene at low levels
of occupational exposure and to find out whether the de-
termination of unchanged VOCs in urine would be useful
for the assessment of exposure around and below the cur-
rent occupational exposure limit.
MATERIALS AND METHODS
Ethical issues
The local Bioethical Committee approved the study pro-
tocol. Each of the participants agreed to join the survey.
Study population and air sampling
Paint factory: Twenty eight workers (20 males and 8 fe-
males) exposed to volatile organic compounds participat-
ed in the study.
Footwear factory: Thirty three workers (9 males and 26 fe-
males) were recruited for the study.
In both workplaces, airborne VOCs were sampled in the
workers´ breathing zone, using a passive diffusive per-
sonal sampler (SKC, Gilian Air-350). The sampling was
performed during the work shift.
Collection of biological material
Urine samples were collected at the end of work shift.
Blood samples were collected 15 min after the work shift
was over. About 5 ml of venous blood from the cubical
vein was drawn using the venoject system.
Urine samples were collected into glass bottles. Imme-
diately after collection, 2 ml of urine was transferred
into 10 ml headspace vials containing 1 g NaCl. The vials
were sealed using caps with teflon membrane and stored
at 4–8°C until analysis.
Page 3
URINARY VOCS AS BIOMARKERS OF EXPOSURE O R I G I N A L P A P E R S
IJOMEH 2010;23(1) 23
Table 1. VOC concentrations in the air
Workplace VOC No. of subjects
Exposure
concentrations
(mg/m3)
GM±GSD
Min. Max.
Paint factory toluene 19 1.1±2.23 0.2 4.7
ethylbenzene 23 3.1±3.85 0.4 40.9
m,p-xylene 24 9.7±4.66 0.6 122.6
o-xylene 22 1.9±4.09 0.1 20.9
Footwear factory toluene 35 105.4±1.76 31.9 349.4
Table 2. Concentrations of biomarkers in blood and urine
Workplace VOC No. of subjects
Blood (μg/l)
GM±GSD
Urine (μg/l)”
GM±GSD
Urinary metabolites
GM±GSD
Paint factory toluene 19 5.36±2.03 2.01±1.73 benzoic acid
10.65±2.58 (mg/h)
ethylbenzene 23 16.47±3.54 12.38±2.52 mandelic acid
1.77±5.31 (mg/h)
m,p-xylene 24 35.14±5.26 2.04±2.77 m,p-methylhippuric acid
0.053±5.45(g/l)
o-xylene 22 8.05±4.73 1.98±2.43 o-methylhippuric acid
0.01±3.71 (g/l)
Footwear factory toluene 35 363.1±1.51 228.1±1.68 benzoic acid
35.5±1.86 (mg/h)
o-cresol
1.1±2.4 (mg/l)
Fig. 1. Relationship between toluene concentration in the air
and in blood samples collected from paint factory workers.
Fig. 2. Relationship between toluene concentration
in the urine samples collected from paint factory workers.
Page 4
O R I G I N A L P A P E R S B. JANASIK ET AL.
IJOMEH 2010;23(1)24
The correlation coefficients between the concentrations of
TOL, EB, o-XYL, m,p-XYL in the air, urine and blood in
the paint factory were high, r = 0.72, 0.71, 0.88, 0.93 and
r = 0.81, 0.72, 0.75, 0.72, respectively. The results indicate
that when the normal value of urinary metabolite is low,
as in the case of methylhippuric acid, the correlation be-
tween the air concentration of a given VOCs and of urine
concentration of its metabolite is high. However, in our
study, this finding did not refer to toluene for which, under
urine or of selected metabolites in urine samples collected
from workers at the paint and footwear factories are pre-
sented in Table 3.
Table 3. Relationship between VOCs concentrations in the air, blood and urine and VOCs metabolites in urine
VOCs in air (mg/m3)
GM±GSD
Biomarkers Regression equation
Correlation
coefficient
P values
Toluene1
N=19
1.1±2.2
toluene in blood (μg/l) Y = 3.96x+0.99 0.81 p < 0.05
toluene in urine (μg/l) Y = 0.69x+1.44 0.72
benzoic acid in urine (mg/h) Y = 0.59x+15.23 0.12
Toluene2
N = 35
105.4±1.76
toluene in blood (μg/l) Y = 2.24x+120.5 0.87 p < 0.05
toluene in urine (μg/l) Y = 0.71x–5.5 0.91
benzoic acid in urine (mg/h) Y = 0.33x+2.34 0.83
o-cresol in urine (mg/l) Y = 0.01x+0.30 0.52
Ethylbenzene N = 24
3.5±4.2
ethylbenzene in blood (μg/l) Y = 2.84x+10.7 0.72 p < 0.05
ethylbenzene in urine (μg/l) Y = 0.13x+1.02 0.71
mandelic acid in urine (mg/h) Y = 0.23x+0.17 0.88
o-Xylene
N = 24
1.9±3.9
o-xylene in blood (μg/l) Y = 3.50x+6.19 0.75 p < 0.05
o-xylene in urine (μg/l) Y = 0.799x–0.65 0.88
o-methylhippuric acid in urine (g/l) Y = 0.0046x+0.0062 0.91
m,p-Xylene
N = 24
9.8±4.7
m,p-xylene in blood (μg/l) Y = 2.84x+33.4 0.72 p < 0.05
m,p-xylene in urine (μg/l) Y = 0.592x–2.506 0.93
m,p-methylhippuric acid in urine (g/l) Y = 0.006x+0.026 0.92
1 Paint factory.
2 Footwear factory.
Table 4. Comparative evaluation of blood analysis and
urinanalysis as the means of biological monitoring
Compound
Blood
LSC
Urine
LSC
Urinary metabolites
compound LSC1 LSC2
Toluene 1.2 1.2 benzoic acid no regression
Ethylbenzene 7.8 11.5 mandelic acid 2.0 25.00
Xylene 3.8 6.5 methylhippuric
acids
2.5 1.75
LSC/LSC1 — concentration of VOC or its metabolites at which the
lower 95% confidence limit for the VOCs in blood or urine is equal
to the upper 95% confidence limit at 0 mg/m3. LSC2 — normal level of
metabolites in urine.
Fig. 3. Relationship between toluene concentration in the air
and the rate of urinary excretion of benzoic acid in the samples
collected from paint factory workers.
Page 5
URINARY VOCS AS BIOMARKERS OF EXPOSURE O R I G I N A L P A P E R S
IJOMEH 2010;23(1) 25
conditions of low level exposure, the correlation between
air concentration of toluene and urinary concentration of
benzoic acid was significantly lower (r = 0.12) than for
TOL-U or TOL-B.
Using the method described by Kawai [13], trials were
made to establish the lowest separation concentration
(LSC) at which the exposed subject can be distinguished
from the nonexposed one. The trials were performed with
the use of the biological exposure index. A graphic analy-
sis was carried out to determine the solvent concentration
at which the lower 95% confidence limit for the solvent in
blood or urine was equal to the upper 95% confidence lim-
it at 0 mg/m3. Examples of the graphic analysis for toluene
are presented in Figures 1–3. The results are summarized
in Table 4.
DISCUSSION
In our study, the relationship between VOCs concentra-
tions in the air and urine samples was linear within the low
study range of air VOCs concentrations. The high correla-
tions between airborne and urinary VOCs are consistent
with the previously reported data [14,4,6,7].
The comparative evaluation of toluene exposure
showed that in the low range of exposure concentra-
tions, TOL-U can be considered a better biomarker of
exposure than BA-U and o-cresol in urine. These results
are concordant with the findings reported by different
authors [15,16].
The results of the LSC analysis conducted to distinguish
between the exposed and non-exposed subjects showed
that VOCs in urine and blood can be considered equiva-
lent biomarkers. The determination of VOCs in urine is
noninvasive and the excretion of unchanged VOCs inte-
grates the first two rapid phases of elimination from blood,
which simplifies the sampling strategy [2]. Moreover, the
unchanged VOCs in urine are more specific biomarkers
than their metabolites and the method enables simulta-
neous determinations of VOCs concentrations in mix-
tures. The determination of unmetabolized form of tolu-
ene as a biomarker of occupational exposure was for the
first time proposed by ACGIH in the notice of intended
changes [17,18].
The present study confirms that the concentration of un-
changed volatile organic compounds in urine is a specific
and sensitive biomarker for assessment of VOCs expo-
sure.
The findings of the present study make us conclude that
the unmetabolized forms of volatile organic compounds
in urine can be regarded not only as biomarkers of occu-
pational, but also of environmental exposure.
CONCLUSIONS
Biological monitoring of exposures to VOCs has been
used for decades in the assessment of the internal expo-
sure of workers. However, the progress in the application
of biomonitoring in occupational health has been slow,
mainly because monitoring has been based mainly on the
separate determination of specific metabolites in urine.
The arguments for the use of measurement of unchanged
compounds in urine are: the non-invasive specimen col-
lection, the simultaneous quantification of VOCs mixture
in single urine sample, and the possibility of monitoring
at low exposure concentrations. This is of particular im-
portance as the biological monitoring has been increas-
ingly used for assessment of VOCs exposure in the gen-
eral population. For the practical implementation of the
method both in occupational setting and indoor exposure,
it would be necessary to conduct more detailed analyses of
the findings, which could be the subject of further research
on this topic.
REFERENCES
1. Kostrzewski P, Piotrowski JK. Toluene determination in capil-
lary blood as a biological indicator of exposure to low levels of
toluene. Pol J Occup Med Environ Health 1991;4:249–59.
2. Janasik B, Jakubowski M, Jałowiecki P. Excretion of unchanged
volatile organic compounds (toluene, ethylbenzene, xylene and
mesitylene) in urine as a result of experimental human volunteer
study. Int Arch Occup Environ Health 2008;81:443–9.
End of preview.
