Improved method to measure urinary alkoxyacetic acids.

Abstract

To simplify the current preparation of samples, and to improve the specificity and reliability of the conventional analytical methods to measure urinary alkoxyacetic acids.
Samples containing alkoxyacetic acids including methoxy, ethoxy, and butoxyacetic acids (MAA, EAA, and BAA) were acidified with HCl and extracted with a mixed solvent of methylene chloride and isopropyl alcohol, then analysed by gas chromatography/mass spectrometry (GC/MS).
Optimal results were obtained when pH was 1.05-1.45, the ratio of methylene chloride and isopropyl alcohol was 2:1, and when extraction time was 10 minutes. Over the concentration range 0.3-200 micrograms/ml, MAA, EAA, and BAA could be determined with a pooled coefficient of variation (nine concentrations, six replicate samples) of 5.55%, 6.37%, and 6.41%, respectively. Urine samples were stable for at least 5 months and 3 freeze-thaw cycles at -20 degrees C. The limits of detection of MAA, EAA, and BAA were 0.055, 0.183, and 0.009 microgram/ml, respectively. The matrix effect of urine samples was negligible for MAA and EAA, but were marginally significant for BAA. The average recoveries of alkoxyacetic acids were 99%-101%. In urine samples MAA from 15 exposed workers showed a strong linear correlation (r = 0.999, slope = 1.01) between the new GC/MS method and Sakai's GC method.
The simplified non-derivatisation pretreatment of samples coupled with GC/MS can provide a specific, sensitive, simple, safe, and reliable method for the biological monitoring of occupational exposure of ethylene glycol ethers.

Full text

Page 1

Improved method to measure urinary alkoxyacetic
acids
Tung-Sheng Shih, Jui-Shu Chou, Cheng-Yao Chen, Thomas J Smith
Abstract
Objectives—To
preparation of samples, and to improve
the specificity and reliability of the con-
ventional analytical methods to measure
urinary alkoxyacetic acids.
Methods—Samples containing alkoxyace-
tic acids including methoxy, ethoxy, and
butoxyacetic acids (MAA,EAA,and BAA)
were acidified with HCl and extracted
with a mixed solvent of methylene chlo-
ride and isopropyl alcohol, then analysed
by gas chromatography/mass spectrom-
etry (GC/MS).
Results—Optimal results were obtained
when pH was 1.05–1.45,the ratio of methy-
lene chloride and isopropyl alcohol was
2:1, and when extraction time was 10 min-
utes.Over the concentration range 0.3–200
µg/ml, MAA, EAA, and BAA could be
determined with a pooled coeYcient of
variation (nine concentrations, six repli-
cate samples) of 5.55%, 6.37%, and 6.41%,
respectively. Urine samples were stable
for at least 5 months and 3 freeze-thaw
cycles at −20°C. The limits of detection of
MAA, EAA, and BAA were 0.055, 0.183,
and 0.009 µg/ml, respectively. The matrix
eVect of urine samples was negligible for
MAA and EAA, but were marginally
significant for BAA. The average recover-
ies of alkoxyacetic acids were 99%–101%.
In urine samples MAA from 15 exposed
workers showed a strong linear correla-
tion (r=0.999,slope=1.01) between the new
GC/MS method and Sakai’s GC method.
Conclusions—The simplified
vatisation pretreatment of samples cou-
pled with GC/MS can provide a specific,
sensitive,simple,
method for the biological monitoring of
occupational exposure of ethylene glycol
ethers.
(Occup Environ Med 1999;56:460–467)
simplify thecurrent
non-deri-
safe,and reliable
Keywords: alkoxyacetic acids, ethylene glycol ethers,
biological monitoring
The
2-methoxyethanol,
2-butoxyethanol and their acetates are widely
used as industrial solvents because of their
excellent chemical and physical properties.1
Animal studies and case reports have found
that ethylene glycol ethers can cause strong
haematological, developmental, and reproduc-
tive eVects through inhalation, dermal absorp-
tion,andingestion.2–6
showed that ethylene glycol ethers are mainly
ethyleneglycol ethers,such as
2-ethoxyethanol,and
Metabolic studies
metabolised to their corresponding alkoxyace-
tic acids, methoxyacetic acid (MAA), ethoxy-
acetic acid (EAA), and butoxyacetic acid
(BAA) through alcohol dehydrogenase and
aldehyde dehydrogenase.7
showed that ethylene glycol ethers themselves
evoke no toxic responses in either Sertoli cell
culture8or in rat whole embryo culture,9
whereas the corresponding alkoxyacetic acids
do. In vivo studies showed that equivalent
doses of ethylene glycol ethers and their corre-
sponding metabolite alkoxyacetic acids are
equally eVective in causing adverse eVects.10 11
Alkoxyacetic acids have been found to be the
main metabolites in the urine of both tested
animals and humans occupationally or experi-
mentally exposed to ethylene glycol ethers.12–16
Thus, it is suggested that alkoxyacetic acids are
the toxicants of exposure to ethylene glycol
ethers that produce these health eVects.Several
studies also found that skin absorption could
be the main route of entry of ethylene glycol
ethers, and that traditional environmental
monitoring for air contamination might not be
able to evaluate the total dose from all routes of
exposure.17–19
Thesefindings
measurement of urinary alkoxyacetic acids can
be viewed as an indicator of potential health
eVects as well as an assessment of total uptake
throughbothinhalation
absorption.1
Several methods of biological monitoring to
measure the urinary alkoxyacetic acids have
been published.20–30Lyophilisation, counter ion
pair, liquid-solid phase extraction, derivatisa-
tion with pentafluorobenzyl bromide, diaozo-
methane, and trimethylsilyldiazomethane were
most commonly used in the pretreatment of
samples. However, pentafluorobenzyl bromide
derivatisation can generate a very irritating gas,
hydrogen bromide. Diazomethane derivatisa-
tion is hazardous for laboratory chemists.24
Trimethylsilyldiazomethane is very expensive
and the impurity in products from diVerent
manufacturers might cause interference in
analysis of alkoxyacetic acids. Lyophilisation
and liquid-solid phase extraction are time con-
suming, expensive, and require considerably
more sample preparation steps.
Flame ionisation detectors (FIDs), and elec-
tron capture detectors (ECDs) are commonly
used in the current methods, but they are not
specific. The ECD is very sensitive for com-
pounds containing halogen,but further isolation
steps have to be applied to remove reagents con-
taining halogen and contaminants to avoid
interference,and the total method error is much
higher. Large variation and matrix eVects were
also noticed in measurement of urinary MAA
In vitro studies
suggestthat
and dermal
Occup Environ Med 1999;56:460–467460
Institute of
Occupational Safety
and Health, Council of
Labor AVairs, Taipei,
Taiwan, Republic of
China
T-S Shih
J-S Chou
C-Y Chen
Department of
Environmental Health,
Harvard School of
Public Health, Boston,
Massachusetts, USA
T J Smith
Correspondence to:
Dr Tung-Sheng Shih,
Division of Method
Development and Analysis,
Institute of Occupational
Safety and Health, Council
of Labor AVairs, 4th Floor,
132 Min-Sheng East Road,
Section 3, Taipei, Taiwan,
Republic of China.
Telephone 00886 2 8770
1559; fax 00886 2 545 1048
1049; email
stone@mail.cla.gov.tw
Accepted 13 January 1999
group.bmj.com on July 15, 2011 - Published by oem.bmj.comDownloaded from

Page 2

with the GC/ECD method.24Wittfoht et al
used N-methyl-N-(tert-butyl-dimethyl-silyl)-
trifluoro-acetamide and
tography/mass spectrometry (GC/MS) method
to measure urinary MAA.26That method is
sensitive and specific compared with other
existing methods. However, the time consum-
ing pretreatment of the samples make it less
eYcient. Negative chemical ionisation mode
GC/MS was used to measure the extremely low
concentrations of 2-butoxyethanol and BAA in
human blood and urine samples.17 29 30The
negative chemical ionisation needs more com-
plex analytical skill and expensive equipment,
and is not available in many research or service
laboratories.The deuterated internal standards
used are very expensive and not commercially
available in Taiwan. Besides these problems,
most of the existing methods were not
validated either in diVerent laboratories or by
diVerent analytical methods or field samples.
The main purposes of this study were to
simplify the preparation of samples, to improve
the specificity and reliability of the existing
analytical methods, and to provide adequate
laboratory and field validation data on a new
GC/MS method. In this study, urinary MAA,
EAA, and BAA, the main metabolites of the
three most popularly used ethylene glycol
ethers(2-methoxyethanol, 2-ethoxyethanol,
the gaschroma-
and 2-butoxyethanol), were chosen as the
target research chemicals.
Materials and methods
REAGENTS
Chemicals used in this study were MAA (98%,
E Merck,Darmstadt,Germany),EAA (98%,E
Merck), isopropyl alcohol (99.7%, E Merck),
methylene chloride (99.7%, E Merck), HCl
(37%, E Merck), and internal standard dichlo-
robenzene (99%, E Merck) and BAA (> 99%,
Acros, NJ, USA).
SAMPLE PREPARATION
Pooled urine samples from more than six
healthy non-exposed people were used as the
matrix blank. Concentrated HCl (60 µl) was
added to 1 ml urine spiked with 120 µg/ml
MAA, EAA, and BAA. The MAA, EAA, and
BAA were extracted for 10 minutes by an
orbital shaker (TKS, Model OS-701, Taipei,
Taiwan) with 1 ml of mixed solvents of methy-
lene chloride and isopropyl alcohol (2:1 v/v),
containing 0.005µl/ml
dichlorobenzene. After mixing and centrifug-
ing at 3000 rpm for 5 minutes, the lower
organic layer was transferred to a 1.5 ml vial,
and 2 µl of this organic layer was injected into
the GC/MS apparatus for analysis.
internal standard
SAMPLE ANALYSIS
The MAA, EAA, and BAA samples were ana-
lysed by GC/MS (Hewlett Packard, G1,800A,
GCD, CA, USA) equipped with an HP 7673A
automatic sampler and a capillary column (J
and W DB Wax, 60 m×0.25 mm ID, 0.25 m
film thickness, CA, USA). The oven tempera-
ture was initially maintained at 60°C for 5
minutes, then was programmed to 195°C at
10°C/minutes, held at 195°C for 1 minute,
programmed again to 205°C at 10°C/min,held
at 205°C for 5 min, and was finally pro-
grammed to 220°C at 5°C/minute, and held at
220°C for 2 minutes. The injector temperature
was set at 210°C, and the detector temperature
was at 260°C. Aliquots of 2 µl extraction
solvents were automatically injected at 33
minute intervals. Splitless mode was used, and
the flow rate of helium carrier gas was 1
ml/minute.
Electron energy used for mass spectrometry
was 70 eV. Mass spectra (fig 1) were first used
for qualitative confirmation of MAA, EAA,
BAA, and the internal standard. Selected ion
monitoring mode was then used for quantita-
tive measurement based on the abundance
ratio of the major fragments of MAA (m/z 45),
EAA (m/z 31), BAA (m/z 57) with respect to
the major fragment of internal standard
dichlorobenzene (m/z 146). Qualitative confir-
mation was checked by the abundance ratio of
the secondary to the primary fragment peaks of
MAA (m/z 60:45), EAA (m/z 59:31), BAA
(m/z 41:57), and for the internal standard (m/z
148:146) in selected ion monitoring.
Figure 1
and (C) butoxyacetic acid (BAA) in urine samples.All those spectra have been confirmed
by the reference mass spectra.
Mass spectra of (A) methoxyacetic acid (MAA),(B) ethoxyacetic acid (EAA),
350 000
0
100 9070 80 60 4050
m/z
C
B
A
Abundance
30
2743
29
41
60
73
87
89
107 114 125133
74
57
18
2010 110120130 140
300 000
250 000
200 000
150 000
100 000
50 000
0
100 9080 6040 30
26
39
31
45
46
59
6975 79
86
94 105
57
18
14
2010 11050 70
2000
4000
6000
8000
7000
6000
5000
4000
3000
2000
1000
0
100 9070 806040 5030
29
45
41
60
6272
89
15
2010 110 120 130 140 150 160 170
8000
9000
Improved method to measure urinary alkoxyacetic acids
461
group.bmj.com on July 15, 2011 - Published by oem.bmj.com Downloaded from

Page 3

EXAMINATION OF THE OPTIMAL PRETREATMENT
OF SAMPLES
The alkoxyacetic acids are highly water and
lipid soluble, and most when dissolved in urine
are in ionised form. The addition of HCl can
convert the ionised forms of alkoxyacetic acids
to non-ionised form for better solvent extrac-
tion. Extraction procedures which give the
highest response on GC/MS are chosen as the
optimal conditions. The main variables exam-
ined in this study were pH value,ratio of mixed
extraction solvents, and extraction time. Cali-
bration curves between and within days, recov-
ery, precision, and matrix eVects were tested.
Sample storage stability included testing 3
freeze-thaw cycles. Limits of detection (mean
+3 SD of the blank baseline signals divided by
the slope of the standard curve) and limits of
measurement (mean +10 SD of the blank
baseline signal divided by the slope of the
standard curve)31with and without derivatisa-
tion procedures were also examined in this
study. The lowest concentration of the stand-
ard curve was chosen as the lowest reliable
measurement when the correlation coeYcient
was greater than 0.995, the relative error and
coeYcient of variation were <15%, the signal
over noise ratio was >4, and the ratio of the
secondary to the primary fragments of alkoxy-
acetic acids was within 20% of that in the ref-
erence mass spectra.
METHOD COMPARISON
Urine samples were collected from 19 workers
exposed to 2-methoxyethanol in a semiconduc-
tor copper laminate circuit board manufactur-
ing plant and were separated into six aliquots
after 5 minutes of mixing. They were stored at
−20°C before analysis. A previously validated
GC/FID method,25was used as the reference
method to compare with our GC/MS method.
The procedures are summarised as follows: 0.1
ml BAA (internal standard, 10.17 g/ml), was
added to 1 ml urine, which was acidified with
0.1 ml concentrated HCl. The MAA and EAA
were then extracted for 10 minutes by an
orbital shaker with 1 ml of the mixture of 2:1
(v/v) methylene chloride and isopropyl alcohol.
After centrifuging at 3000 rpm for 5 minutes,
0.5 ml of the lower organic layer was
transferred to a 1.5 ml vial, and was derivatised
with 25 µl of a solution of trimethylsilyldia-
zomethane (2.0 M solution in hexane, Aldrich,
WI, USA). The esterified acid solution was
analysed by a GC (Hewlett-Packard 5890
Series II, CA, USA) equipped with an HP
7673A autosampler and an FID. A magabore
column (J and W DB-WAX, 30 m×0.53 mm,
1.0 m film thickness, CA, USA) was used. The
oven temperature was initially maintained at
50°C for 2 minutes, then was programmed to
65°C at 5°C/min, programmed again to 130°C
at 10°C/min, held at 130°C for 8 minutes, then
finally programmed to 225°C at 25°C/min,and
held at 225°C for 10 minutes. The injection
temperature was 220°C, and the detector tem-
perature was 230°C. Of the esterified acid
solution 2 l were injected into the GC appara-
tus after 30 minutes stirring by an orbital
shaker. The flow rate of nitrogen carrier gas
was 13 ml/min, air flow was 300 ml/min, and
hydrogen flow was 30 ml/min. The concentra-
tion range of the standard curves for MAA and
EAA was 3.5 to 200 µg/ml.
DATA ANALYSIS
The GC/MS response, either the abundance of
the main fragment or the ratio of the
abundance of the main fragment to that of the
internal standard, was expressed as mean (SD)
of triplicate analyses. As replicate analyses per-
formed on an industrial hygiene sample are
usually normally distributed,32one way analysis
of variance (ANOVA) and Tukey’s pair wise
tests were used to compare the diVerences
among diVerent conditions of analytical vari-
ables such as pH values, extraction time, and
extraction solvent combination. A Wilcoxon
non-parametric test was also used to evaluate
the results. Eleven concentrations of standard
solutions and simple linear regression were
used to establish the standard curves. Analysis
of covariance (ANCOVA) was used to test for
the diVerences between slopes of standard
curves with diVerent matrices (deionisation
water and urine matrices). Linear regression,
the plot of concentration diVerence versus
mean concentration of two methods, and the
limits of agreement33 34were used for the com-
parison of the new GC/MS method and the
reference method. Pooled coeYcients of varia-
tion (CV) across diVerent concentrations were
used to show the precision of the new GC/MS
method.31
Results
STANDARD CURVES AND GC/MS SPECTRA
Three standard curves between days and three
standard curves within days were obtained for
MAA, EAA, and BAA in the concentration
range of 0.3 µg/ml to 200 µg/ml (the concentra-
tion range of the calibration curves for MAA
was 0.15 µg/ml to 200 µg/ml). For these a con-
centration ranges, the correlation coeYcients
of standard curves between the GC/MS
response and the concentrations of MAA,
EAA, and BAA were all >0.997. The pooled
coeYcients of variation (CVs) of the calibra-
tion curves within days (11 concentrations)
were 5.48% (1.26%–12.87%), 6.59% (1.41%–
17.44%) and 6.17% (2.59%–12.09%) for
MAA, EAA, and BAA, respectively (n=3). The
pooled CVs of calibration curves between days
(11 concentrations) were 5.50% (0.42%–
12.44%), 5.58% (0.17%–12.53%) and 4.62%
(0.64%–14.26%) for MAA, EAA, and BAA
respectively (n=3).
Figure 2 shows the gas chromatograms of
blank and spiked urine samples. MAA, EAA,
and BAA, and the internal standard were com-
pletely separated by the column under the ana-
lytical conditions. No interfering chemicals
were found in GC or MS of urine samples.The
mean (SD) retention times of MAA, EAA,
BAA, and internal standard were 21.452
(0.005), 21.780 (0.003), 24.745 (0.005), and
16.548 (0.001) minutes, respectively (n=12).
The MAA, EAA, BAA, and internal standard
were reconfirmed by matching the MS of urine
462
Shih,Chou,Chen,et al
group.bmj.com on July 15, 2011 - Published by oem.bmj.com Downloaded from

Page 4

samples (fig 1) against the data bank reference
spectra.
PRETREATMENT OF THE SAMPLES
Ratio of extraction solvents
Urine samples spiked with 120 g/ml MAA,
EAA, and BAA, were acidified to a pH of 1.05
with 60 µl concentrated HCl and were
extracted separately with 1 ml of pure methyl-
ene chloride, pure isopropyl alcohol, and a
mixed solvent with diVerent ratio of methylene
chloride and isopropyl alcohol that ranged
from 1:1 to 5:1 (v/v) for 10 minutes. Figure 3
showed that the main fragment abundance of
alkoxyacetic acids was influenced (p<0.05) by
the diVerent ratio of methylene chloride and
isopropyl alcohol. The maximal abundance
within the experiment tested in this study was
found at a ratio of 2:1.
Extraction time
Urine samples spiked with 120 µg/ml MAA,
EAA, and BAA, were acidified to a pH of 1.05
with 60 µl concentrated HCl and were
extracted with 1 ml methylene chloride and
isopropyl alcohol (2:1, v/v) with 2, 5, 10, and
20 minute extraction times. The ANOVA and
Tukey’s pairwise tests showed the ratio of main
fragment abundance of the alkoxyacetic acids
to the internal standard was significantly
diVerent between 2 minutes and other extrac-
tion time intervals for MAA (p=0.024) and
BAA (p=0.016), but not significantly diVerent
for EAA (p=0.055). There were no significant
diVerences among 5, 10, and 20 minute
extraction times (p=0.10 for all three alkoxy-
acetic acids) by ANOVA and Wilcoxon non-
parametric tests. Figure 4 showed that the
extraction efficiency became stable after 5
minutes of extraction. As 10 minutes of extrac-
tion will slightly improve eYciency and maybe
also reduce the fraction of the method error
Figure 2
(B) Gas chromatogram of 2-methoxyethanol (ME),2-ethoxyethanol (EE),
2-butoxyethanol (BE),methoxyacetic acid (MAA),ethoxyacetic acid (EAA),butoxyacetic
acid (BAA),and internal standard (IS) dichlorobenzene in a urine sample.The
concentration of alkoxyacetic acids spiked was 110 µg/ml.
(A) Gas chromatogram of urine blank samples spiked with internal standard.
2 200 000
2 400 000
0
10.00
Time (h)
MAA
IS
ME
EE
EAA
BAA
BE
B
Abundance
11.33
16.56
15.15
15.67
13.26
12.15
29.32
24.38
24.77
23.11
21.46
21.79
18.06
18.23
2
12.0014.00 16.0018.00 20.0022.00 24.00 26.0028.00
2 000 000
1 800 000
1 600 000
1 400 000
1 200 000
1 000 000
800 000
600 000
400 000
200 000
0
10.00
IS
A
11.36
4
28.98
23.12
25.10
3
24.39
2
21.84
18.23
16.57
12.0014.0016.00 18.00 20.00 22.0024.00 26.0028.00
400 000
350 000
300 000
250 000
200 000
150 000
100 000
50 000
Figure 3
concentration is 120 µg/ml,pH was 1.05,and extraction time was 10 minutes.Data are means (SDs) of triplicates.Selective ion monitoring was used to
measure alkoxyacetic acids.
Influence of diVerent ratio of extraction solvents:(A) methoxyacetic acid;(B) ethoxyacetic acid;and (C) butoxyacetic acid.Analyte
0.0
MCIPA1:1 2:13:1 4:15:1
A
Abundance
2.0 × 107
1.5 × 107
2.5 × 107
1.0 × 107
5.0 × 106
MCIPA 1:12:1 3:1 4:15:1
Ratio of extraction solvents
B
MCIPA1:1 2:13:14:1 5:1
C
Figure 4
120 µg/ml,ratio of methylene chloride (MC) and isopropyl alcohol (IPA) was 2:1 (v:v),and pH was 1.05.Data are means (SDs) of triplicates.Selective
ion monitoring was used to measure alkoxyacetic acids.
Influence of diVerent extraction times of (A) methoxyacetic acid,(B) ethoxyacetic acid,and (C) butoxyacetic acid.Analyte concentration was
00
510 1520 25
A
Ratio of abundance
4
3
5
2
1
Extraction time (min)
0510152025
B
0510 152025
C
Improved method to measure urinary alkoxyacetic acids
463
group.bmj.com on July 15, 2011 - Published by oem.bmj.comDownloaded from

Page 5

caused by incomplete extraction, 10 minutes
was chosen as the optimal extraction time.
pH Value
Urine samples spiked with 120 µg/ml MAA,
EAA, and BAA, were acidified with diVerent
quantities of concentrated HCl and were
extracted with 1 ml of methylene chloride and
isopropyl alcohol (2:1,v/v) for 10 minutes.The
ANOVA and Tukey’s pairwise tests showed
that no significant diVerences existed over a pH
range of 1.45–1.05 for MAA,pH 1.45–0.85 for
EAA, and pH 2.62–0.85 for BAA, as shown in
figure 5. There were no significant diVerences
at pH 1.45–1.05 (p=0.60 for all three alkoxy-
acetic acids) by ANOVA and Wilcoxon non-
parametric tests. A pH of 1.05 was therefore
chosen for the simultaneous analyses of three
alkoxyacetic acids.
PRECISION
Nine concentrations of MAA, EAA, and BAA
(1.15,2.30,5.76,11.52,23.04,40.32,84,144,
and 200 µg/ml) representing the range to be
expected for biological samples were exam-
ined. Six replicate samples at each concentra-
tion showed that the pooled CVs for MAA,
EAA, and BAA were 5.55% (3.17%–8.24%),
6.37% (4.13%–9.15%), and 6.41% (4.08%–
8.77%), respectively, and the CVs showed no
evidence of dependence on concentration.
MATRIX EFFECTS
Deionised water and non-exposed human
pooled urine were used for matrix eVect
comparison of MAA, EAA, and BAA. The
ANCOVA analysis showed no significant dif-
ferences between the slopes of deionised water
and urine matrix (0.05<p<0.10) for MAA and
EAA, and a significant matrix eVect for BAA
(0.02<p<0.05), in the concentration range of
0.3 µg/ml to 200 µg/ml, as shown in figure 6.
The ratio of the slopes of standard curves
between urine and deionisation water were
0.95,0.93,and 0.91 for MAA,EAA,and BAA,
respectively.
SENSITIVITY
The limits of detection of this new GC/MS
method were 0.055 µg/ml for MAA; 0.183
µg/ml for EAA; and 0.009 µg/ml, for BAA
respectively, and the limits of measurement
were 0.136 µg/ml for MAA, 0.442 µg/ml for
EAA, and 0.018 µg/ml for BAA, respectively.
The concentration ranges of the standard
curves of the new GC/MS method with and
without the trimethylsilyldiazomethane deriva-
tisation procedures were the same, 0.3–200
µg/ml.
RECOVERY
The mean (SD) recoveries of MAA,EAA,BAA
in spiked urine quality control samples of three
concentrations (23, 84, and 144 µg/ml) were
100.8% (5.6%) (triplicate analyses for each
concentration), 100.0% (4.3%), and 98.6%
(2.4%), respectively.
SAMPLE STABILITY TEST
Storage stability
Twenty one spiked quality control samples
with three concentrations (24, 59, and 118
µg/ml) were prepared and stored at -20°C for
Figure 5
acid,(B) ethoxyacetic acid,and (C) butoxyacetic acid.
Analyte concentration was 120 µg/ml,ratio of methylene
chloride (MC) and isopropyl alcohol (IPA) was 2:1 (v:v),
and extraction time was 10 minutes.Data are means
(SDs) of triplicates.Selective ion monitoring was used to
measure alkoxyacetic acids.
Influence of urinary pH on (A) methoxyacetic
0.0
0.85 1.05 1.17 1.31
pH
1.45 6.53 2.62 1.72
C
Abundance
2.0 × 107
1.5 × 107
1.0 × 107
0.5 × 106
0.0
B
8.0 × 106
6.0 × 106
4.0 × 106
2.0 × 106
0.0
A
4.0 × 106
3.0 × 106
2.0 × 106
1.0 × 106
Figure 6
acid,(B) ethoxyacetic acid,and (C) butoxyacetic acid.
Data are means (SDs) of triplicates.Ratio of abundance
represents the abundance ratio of main fragments of
alkoxyacetic acids to that of internal standard.Selective ion
monitoring was used to measure alkoxyacetic acids.Solid
lines and square symbols represent the data in urine.Dotted
lines and circle symbols represent the data in deionised
water.
Influence of matrix eVects on (A) methoxyacetic
0
0 50
Concentration (µg/ml)
200150100
Ratio of abundance
14
12
10
8
6
4
2
0
7
6
5
4
3
2
1
0
A
B
C
6
5
4
3
2
1
464
Shih,Chou,Chen,et al
group.bmj.com on July 15, 2011 - Published by oem.bmj.comDownloaded from