Journal of Analytical Toxicology, Vol. 34, April 2010
Headspace In-Tube Extraction Gas Chromatography–
Mass Spectrometry for the Analysis of Hydroxylic
Methyl-Derivatized and Volatile Organic Compounds
in Blood and Urine
Ilpo Rasanen, Jenni Viinamäki, Erkki Vuori, and Ilkka Ojanperä*
Department of Forensic Medicine, P.O. Box 40, Kytösuontie 11, FI-00014 University of Helsinki, Finland
Abstract
A novel headspace in-tube extraction gas chromatography–mass
spectrometry (ITEX-GC–MS) approach was developed for broadscale analysis of low molecular weight organic compounds in
blood and/or urine. One sample was analyzed following in-vial
derivatization with dimethyl sulfate for ethylene glycol (EG),
glycolic acid (GA), formic acid (FA), other hydroxylic compounds,
and another sample for underivatized volatile organic compounds.
Tenax adsorbent resin was used in the microtrap, and a porous
layer, open tubular GC capillary column was used for separation.
MS was operated in the full-scan mode, identification was based
on the Automated Mass Spectral Deconvolution and Identification
System, and quantification was based on extracted ions. The limits
of quantification for EG, GA, and FA in blood were 10, 50, and
30 mg/L, respectively, and the expanded uncertainties of
measurement were 20%, 16%, and 14%, respectively. The
procedure allowed for the first time the inclusion of EG and GA as
their methyl derivatives within a quantitative HS analysis. The ITEX
method described here was more sensitive for analysis of volatile
organic compounds than the corresponding static headspace
analysis as demonstrated for 11 representative compounds.
Introduction
Ethylene glycol (EG), 1,2-ethanediol, is an industrial chemical and coolant available to the public in the form of automotive antifreeze products. The inherent toxicity of unmetabolized EG is fairly low, but life-threatening toxicity follows from
its metabolites. In particular, glycolic acid (GA) is responsible
for the anion gap metabolic acidosis characteristic of EG
poisonings, and nephrotoxicity is thought to be due to calcium oxalate (1–4). As the estimated lethal dose of EG is only
1.4 mL/kg, both accidental and suicidal poisonings have been
regularly reported (5). The importance of analyzing GA to sup* Author to whom correspondence should be addressed.
port the clinical diagnosis of EG poisoning has often been emphasized (6).
Methanol is another industrial, low molecular weight alcohol used in paint remover and automotive antifreeze liquids, and it causes toxic exposures via its metabolites. Methanol
is also a typical component of embalming fluids. The major
toxic metabolite is formic acid (FA), which causes characteristic
visual defects and even blindness (7). FA may also induce deep
anion gap metabolic acidosis, which frequently results in a
fatal outcome (8). The lethal dose of methanol is estimated to
be 1–2 mL/kg, but as little as 0.3–0.9 mL/kg may be lethal
without medical treatment (9). As with EG, methanol poisoning can be accidental or imply suicidal intention. In a clinical context, the presence of a determinable FA concentration
in blood has been recommended as an indication for hemodialysis when methanol poisoning is suspected (10). The wide variation in lethal dose of methanol in humans to a large extent depends on the fact that many people drink this toxic alcohol
mixed with the antidote ethanol, whereas those who drink
neat methanol die after ingesting much smaller amounts. The
other important variable is the time elapsed between drinking
and initiation of emergency treatment.
Volatile organic compounds (VOC) are a diverse group of low
molecular weight substances, many of which possess abuse
potential. The following product groups were found to be frequently associated with volatile substance abuse (VSA) fatalities
in Britain (11): gas fuels, aerosols, glues, fire extinguishers,
cleaning products, anesthetic agents, and alkyl nitrites. Butane
from all sources accounted for the majority (70%) of VSA
deaths in 2006 (11). VOC analysis is hampered by a low prevalence of positive cases combined with the wide variety of compounds to be screened, posing a challenge for organizing an appropriate analysis service. Technical challenges in sampling
and storage include possible contamination from non-glass
containers and stoppers and the loss of analytes due to leaking
or only partially filled containers.
Static headspace (HS) gas chromatographic (GC) methods
Reproduction (photocopying) of editorial content of this journal is prohibited without publisher’s permission.
113
Journal of Analytical Toxicology, Vol. 34, April 2010
are the backbone of blood ethanol analysis in the medicolegal context (12). As blood alcohol methods aim for high
quantitative performance and high throughput, static HS
methods with isothermal column conditions are regularly applied to this task, which results in analysis cycle times of
around 5 min. However, such methods cannot be accommodated to the requirements of analyzing polar alcohols other
than methanol, ethanol, propanols, and butanols, much less a
wider range of VOCs.
A number of GC methods are available for EG analysis. The
standard method in analytical toxicology involves derivatization of the diol to form the cyclic phenylboronate ester (13–17),
but this method is not suited for the simultaneous analysis of
GA. Methods that are able to simultaneously determine both
EG and GA include silyl derivatization prior to GC with flame
ionization (FID) (18) or mass spectrometric (MS) detection
(19). Derivatization to methyl glycolate has also been used
prior to GC–FID analysis (20,21). FA is almost exclusively determined as a methyl formate ester by HS-GC–FID following
methylation with methanol in sulfuric acid (22–25). Clearly,
this procedure does not allow the simultaneous analysis of FA
and methanol.
VOCs are usually extracted from biological material by static
HS, dynamic HS, or high concentration capacity HS techniques such as HS solid-phase microextraction (SPME) prior to
GC or GC–MS analysis. A comprehensive review discusses the
sample handling and analytical techniques in detail and focuses
on the postmortem diagnosis of VSA (26). Interestingly, derivatization has been infrequently used for VOCs, and consequently, the hydroxylic analytes EG, GA, and FA have not been
included in these assays.
As shown earlier, many separate analysis methods are generally needed for low molecular weight organic compounds
of toxicological relevance. The purpose of the present study
is to establish a rational, cost-effective approach to cover
the analysis of both hydroxylic involatile and volatile substances in blood and urine using a single analytical technique based on in-tube extraction (ITEX). ITEX is a novel,
dynamic HS technique in which a microtrap filled with adsorbent material is placed between a heated HS syringe and
a syringe needle while a part of the HS volume is pumped by
the syringe repeatedly through the microtrap. The microtrap
is then rapidly flash-heated, and the enriched analytes are released by thermal desorption into the GC injector. Unlike
other, similar techniques such as solid-phase dynamic extraction, ITEX does not use a needle as a support for the adsorbent. Thermodesorption takes place using an external
heater around the needle body independently of GC injector
temperature.
In this study, we apply ITEX-GC–MS to a broad-scale scheme
of low molecular weight organic compounds in two parallel
samples: one analyzed following in-vial derivatization with
dimethyl sulfate (DMS) for EG, GA, FA, and other hydroxylic
compounds, and another analyzed for underivatized VOCs.
Tenax adsorbent resin for the microtrap and a porous layer,
open tubular GC capillary column (CP-PoraPLOT Q-HT) were
chosen because they allow polar compounds to be analyzed together with light hydrocarbons with low affinity for water (27).
114
The procedure allows for the first time the inclusion of EG and
GA as their methyl derivatives within a quantitative HS analysis. Further, the higher sensitivity of ITEX over the corresponding static HS in VOC analysis is demonstrated with 11
representative compounds.
Materials and Methods
Materials
EG, methyl formate, and methyl methoxy acetate were purchased from Fluka (Buchs, Switzerland); ethylene-d4 glycol
(EG-d4), GA, dimethyl sulfate (DMS), and 1,2-dimethoxyethane
were from Aldrich (Steinheim, Germany); glycolic-d2 acid
(GA-d2) was obtained from Cambridge Isotope Laboratories
(Andover, MA); FA was from Merck (Darmstadt, Germany);
formic-d acid (FA-d) was from Isotec (Miamisburg, OH); and
tetrabutylammonium hydrogen sulfate (TBA-HSO4) was from
Riedel-de Haën (Seelze, Germany).
Inhalation agents included desflurane (Suprane, Pharmacia,
Stockholm, Sweden), enflurane (Efrane, Abbott, Campoverde,
Italy), halothane (Trothane, ISC Chemicals, Bristol, England),
isoflurane (Forene, Abbott, Queenborough, Kent, England),
and sevoflurane (Sevorane, Abbott, Solna, Sweden).
A qualitative mixture consisting of propane, butane, 2methylpropane, and 2-methylbutane in water was prepared
from Newport butane lighter gas (Keen World Marketing, Newbury, Berkshire, England) by introducing the gas underneath
the water surface. A qualitative solution of 1,1,1,2-tetrafluoroethane was prepared from Plus plasting spray (Würth, Riihimäki, Finland). All other chemicals were generally pro-analysis grade and from various suppliers.
Standard solutions were prepared for each analyte separately by adding a volume corresponding to 10 mg of standard
compound underneath the liquid surface into a 25-mL volumetric flask containing water or methanol and filling the bottle
to the mark. All standard solutions were stored and refrigerated
(4°C).
Sample preparation for hydroxylic organic compounds by
in-vial methylation (ITEX 1)
The internal standard solutions (IS-1) contained 20 mg/mL
each of EG-d4, GA-d2, and FA-d in water.
Urine samples
Urine (0.5 mL) was measured into a 20-mL HS vial; 25 µL of
IS-1 solution was added, and the sample was mixed. For standard
samples, the total volume of the standard solution and urine was
0.5 mL. Anhydrous sodium sulfate (250 mg) was mixed into the
solution to obtain a salting-out effect, followed by adding 50 µL
of 5 M sodium hydroxide solution, 50 µL of 0.1 M TBA-HSO4, and
150 µL of DMS. The vial was closed and mixed.
Blood samples
Blood (0.5 g) was measured into a 10-mL test tube, 25 µL of
IS-1 solution was added, and the sample was mixed. For
preparing standard samples, 0.5 g of bovine blood was used and
Journal of Analytical Toxicology, Vol. 34, April 2010
the corresponding standard solutions were added into the vial
before the IS-1 solution. Acetonitrile (1.5 mL) was gradually
added while simultaneously mixing the sample in a vortex
mixer. The sample was mixed for an additional 1 min and centrifuged. The top layer was transferred into another 10-mL
test tube, and the acetonitrile was evaporated under a gentle
Sample preparation for volatile organic compounds without
derivatization (ITEX 2)
Table I. Static Headspace and In-Tube Extraction
Method Parameters
Parameter
Incubation temperature
Incubation time
Agitator speed
Syringe temperature
Extraction volume
Extraction strokes
Extraction speed
Desorption temperature
Desorption speed
Injection volume
Needle flush time
Trap cleaning temperature
Fill speed
Pull-up delay
Injection speed
Pre inject delay
Flush time
HS
ITEX1*
ITEX2†
60°C
300 s
500 rpm
55ºC
90°C
300 s
500 rpm
55ºC
1500 µL
10
150 µL/s
230°C
20 µL/s
2000 µL
600 s
230°C
60°C
300 s
500 rpm
55ºC
1500 µL
30
150 µL/s
230°C
20 µL/s
2000 µL
600 s
230°C
2000 µL
stream of nitrogen at 40°C. The residue of approximately
250 µL was transferred into a 20-mL HS vial, and 125 mg of anhydrous sodium sulfate was mixed into the solution for saltingout. Thereafter, 50 µL of 5 M sodium hydroxide solution, 50 µL
of 0.1 M TBA-HSO4, and 50 µL of DMS were added, and the vial
was closed and mixed.
The internal standard solution (IS-2) contained 5.0 µg/mL of
diethylketone in water. Blood (1.0 g) was measured into a 20mL HS vial, and 1.0 mL of IS-2 solution was added. The vial
was closed and mixed. For preparing standard samples, 1.0 g of
bovine blood was used, and the corresponding standard solutions were added into the vial before the IS-2 solution. Static
HS sample preparation for VOCs without derivatization was the
same as described previously for ITEX2.
ITEX-GC–MS
The ITEX1 and ITEX2 methods used the same hardware
configuration. In the static HS method, the ITEX injection
unit was replaced with a standard HS unit. ITEX2 parameters
were optimized to obtain high sensitivity for VOCs. The same
parameters were used in ITEX1 if applicable, but the derivatization procedure for EG, GA, and FA required a higher incubation temperature, and the required sensitivity was obtained
with fewer extraction strokes. Final ITEX1, ITEX2, and static
HS method parameters are shown in Table I.
GC–MS was performed with a 5975B VL mass selective
detector coupled with a 6890N GC from Agilent Technologies
(Santa Clara, CA). The GC was equipped with a CTC Analytics
100 µL/s
1000 ms
20 µL/s
500 ms
10 s
* Ethylene glycol, glycolic acid, formic acid, and polar organic compounds.
† Volatile organic compounds.
Table II. Validation Data for Ethylene Glycol, Glycolic Acid, and Formic Acid in Blood and Urine
Blood
Target
Ion*
(m/z)
Q1*
(m/z)
Q2*
(m/z)
Ethylene-d4
glycol-2ME
94
64
47
Ethylene
glycol-2ME
90
60
45
Glycolic-d2
acid-2ME
76
47
Glycolic
acid-2ME
74
45
Formic-d
acid-ME
61
32
Formic
acid-ME#
60
31
Derivative
Linear
range
(mg/L)
R2
LOD†
(mg/L)
S/N‡
Urine
LOQ
(mg/L) S/N
U95§
(%)
Linear
range
(mg/L)
R2
LOD
(mg/L)
LOQ
S/N (mg/L)
S/N
U95
(%)
10–20,000 0.9996 3
6
10
23
20
30–20,000 0.996
10
4
30
29
13
50–20,000 0.997 10
9
50
28
16
50–20,000 0.996
30
10
50
14
12
30
699
14
50–5000 0.9989
50
595
14
30–4000 0.9991
* Extracted target ions and qualifier ions (Q) of methyl derivatives used for quantification in full-scan gas chromatography–mass spectrometry.
† LOD = lower limit of detection; LOQ = lower limit of quantification.
‡ Signal-to-noise ratio at LOD or LOQ.
§ Expanded uncertainty of measurement corresponding to 95% confidence.
# LOD was not determined because of endogenous FA.
115
Journal of Analytical Toxicology, Vol. 34, April 2010
CombiPal autosampler with ITEX and static HS options
(Zwingen, Switzerland). The column was a CP-PoraPLOT
Q-HT capillary column from Varian (25 m × 0.32-mm i.d.,
10-µm film thickness, and 2.5-m particle trap, Lake Forest,
CA). A 2.5-mL HS syringe was used (Hamilton, Bonaduz,
Switzerland). In the ITEX methods, a microtrap containing
44 mg of Tenax TA 80/100 mesh from Supelco (Bellefonte,
PA) was attached to the syringe needle. The GC–MS was operated with ChemStation software (Agilent) with integrated
CTC Control software. Qualitative data analysis was performed using AMDIS software.
The GC was used in the splitless mode with a constant
column flow of 2 mL/min. The injector port temperature was
200°C, and the transfer line temperature was 280°C. The oven
temperature was initially kept at 40°C for 2 min and then increased by 15°C/min to 250°C, which was kept for 4 min
(ITEX1) or 10 min (ITEX2). The mass detector was operated in
electron-ionization and full-scan mode in the range m/z 15–
550. For the analysis of EG, GA, and FA methyl derivatives,
target and qualifier ions were selected from the full-scan data
(Table II), and quantification was performed using these extracted ions.
Validation of quantification
In the validation of the ITEX1 method, the following parameters were evaluated: selectivity (analysis of blank and
spiked matrix samples), linearity, accuracy (bias), precision
(spiked and real samples), repeatability (duplicate analysis on
two different days; sample preparation
performed by different persons), lower
A
B limit of detection (LOD), and lower limit
of quantification (LOQ). Five parallel
spiked samples in 15 levels were prepared in the concentration range of 5
mg/L to 20 g/L. The criteria for the LOQ
and linear range were ± 20% relative
standard deviation (RSD) in precision,
bias within ± 20% of the reference value,
and signal-to-noise ratio (S/N) of > 10.
The criterion for LOD was S/N > 3. The
estimation of uncertainty of measurement (U) was based on the EURACHEM
approach (28), using data described earlier. U was reported as the expanded uncertainty of measurement corresponding to 95% confidence according
to the following equation: U95 = 2(U12 +
U22)½, where U1 is the systematic error
(mean accuracy from calibration and
proficiency test samples) and U2 is the
random error (mean precision and reFigure 1. Analysis scheme for low molecular weight hydroxylic organic compounds in blood and urine
peatability from calibration and real
(ITEX1) (A) and for volatile organic compounds in blood (ITEX2) (B). NIST MS or other library search was
samples).
performed only when required.
AMDIS
Figure 2. In-vial methylation of ethylene glycol (EG), glycolic acid (GA),
and formic acid (FA) with dimethyl sulfate producing EG-2ME, GA-2ME,
and FA-ME, respectively.
116
AMDIS is a computer program that extracts spectra for individual components in a GC–MS data file and identifies target
compounds by matching the spectra against a reference library. The software was developed at the U.S. National Institute of Standards and Technology (NIST) and is freely available
(http://chemdata. nist.gov/mass-spc/amdis/). AMDIS is included in the NIST/EPA/NIH mass spectral library and also includes Lib2NIST converter software, which can be used to
convert ChemStation libraries to AMDIS libraries. In the present study, the AMDIS search was performed with an in-house
target library containing spectra of underivatized compounds
and the corresponding methyl esters or ethers. For the calculation of AMDIS match factors, the correlation between extracted spectrum and library spectrum was utilized. An NIST
MS search or a manual search against other libraries was
performed for unidentified peaks when necessary.
Journal of Analytical Toxicology, Vol. 34, April 2010
Results and Discussion
EG, GA, FA, and hydroxylic organic compounds
Figure 1 shows the analysis scheme for low molecular
weight hydroxylic organic compounds in blood and urine
(ITEX1) and for VOCs in blood (ITEX2). Different sample
preparation procedures were applied in these two methods,
while the GC–MS analysis was the same, except for the
isothermal time kept at the final oven temperature. Hydroxylic
and carboxylic compounds in biological material could be
readily derivatized by in-vial methylation with DMS to form
ethers and esters, respectively, making EG, GA, and FA
amenable to HS analysis (Figure 2). The GC retention times of
hydroxylic organic compounds, as their methyl derivatives
(ITEX1) and underivatized VOCs (ITEX2), are shown in Table
III. Many glycols, alcohols, phenol as well as γ-hydroxybutyric acid could be detected by the method. The boiling points
of the compounds and derivatives included were generally
below 160°C. Methanol could not be included as an analyte in
ITEX1 as the decomposition of the DMS reagent produces
Table III. Gas Chromatographic Retention Times (RT) on a Porous Layer, Open Tubular Column* for
Low Molecular Weight Hydroxylic Organic Compounds as Their Methyl Derivatives (ITEX1) and Underivatized Volatile
Organic Compounds (ITEX2)
Compound
1,1,1,2-Tetrafluoroethane
Propane
Methanol
Acetaldehyde
Formic-d acid
Formic acid (FA)
Methyl formate
Isobutane
Desflurane
Ethanol
Butane
Acetonitrile
Acetone
Isopropyl alcohol
Methylene chloride
Sevoflurane
Carbon disulfide
2-Methylbutane
Diethyl ether
Dimethoxymethane
Isoflurane
1-Propanol
Pentane
Enflurane
tert-Butanol
2-Methylpropanal
Halothane
Ethylmethyl ketone
Butanal
Chloroform
Ethyl acetate
Tetrahydrofuran
Methyl-tert-butyl ether (MTBE)
2-Butanol
Methyl propionate
2-Methyl-1-propanol
1,2-Dichloroethane
Hexane
1-Butanol
RT (min)
ITEX1†
RT (min)
ITEX2‡
8.4
9.5
9.5
10.9
11.7
11.8
16.0
16.4
16.2
16.8
11.8
11.9
12.3
12.3
12.4
13.0
13.7
13.9
14.0
14.3
14.5
14.5
14.5
14.5
14.6
14.6
14.7
14.9
15.2
15.4
15.7
15.8
15.8
16.0
16.0
16.0
16.1
16.1
16.2
16.3
16.6
16.7
16.8
Compound
Ethylene-d4 glycol
Ethylene glycol (EG)
1,1,1-Trichloroethane
Di-isopropyl ether
Benzene
Carbon tetrachloride
Ethyl-tert-butyl ether (ETBE)
Cyclohexane
Trichloroethylene
3-Methylbutanal
Diethyl ketone
Pentanal
Ethyl propionate
2-Pentanol
tert-Amyl methyl ether (TAME)
1,2-Propanediol
Triethylamine
Glycolic-d2 acid
Glycolic acid (GA)
2,3-Butanediol
Heptane
Pyridine
1,3-Propanediol
Isopentyl alcohol
1-Pentanol
Methyl isobutyl ketone
Toluene
Tetrachloroethylene
Hexanal
Butyl acetate
1,2-Butanediol
1,3-Butanediol
1,4-Butanediol
m-Xylene
p-Xylene
o-Xylene
Phenol
1,1,1,2-Tetrachloroethane
γ-Hydroxybutyric acid (GHB)
RT (min)
ITEX1†
RT (min)
ITEX2‡
16.9
16.9
18.3
17.0
17.1
17.2
17.2
17.2
17.4
17.4
17.4
17.7
17.9
17.9
18.0
18.1
18.2
18.5
18.5
18.5
18.6
18.7
18.7
18.9
18.8
18.9
19.0
19.0
19.4
19.6
20.0
20.4
20.5
20.5
20.8
21.9
22.9
22.9
23.0
23.7
23.8
24.7
* CP-PoraPLOT Q-HT capillary column (25 m × 0.32-mm i.d., 10-µm film thickness, 2.5-m particle trap).
† Retention time for methyl derivative.
‡ Retention times sorted according to ITEX2.
117
Journal of Analytical Toxicology, Vol. 34, April 2010
methanol and sulfuric acid. However, methanol was readily desilylation (34). However, the methylation reaction with DMS
tected with ITEX2. In ITEX1, some dimethoxy methane was
appears to be the method of choice where direct derivatization
detected in all samples and reagent blanks, probably forming
in an aqueous sample is an option. DMS has been recently
from methanol. Putrefied samples, especially urine samples,
used in several in situ derivatization applications for organic
showed a higher background than usual, but this was effiacids in water (35–38) or urine samples (39) prior to GC–MS
ciently compensated for by the use of deuterated internal
analysis with HS (35), HS-SPME (36,38,39), or large-volume
standards. FA was found in many postmortem samples at
on-column injection (37). The methylation procedure used in
approximately 50 mg/L, possibly due to postmortem changes.
this study is a modification of these previously presented
The validation data for EG, GA, and FA are shown in Table
methods. The reaction mixture consisted of DMS as the methyII. All three analytes possessed good linearity over the toxicolation agent, the catalyzing ion-pairing agent TBA-HSO4,
sodium sulfate to improve the salting-out effect by increasing
logically relevant concentration range. The expanded uncerthe ionic strength, and sodium hydroxide for adjusting the
tainty of measurement (level of confidence 95%) was no
pH. The toxicity of DMS warrants appropriate precautions in
higher than 20%. Comparison to previous studies indicates
sample work-up and waste disposal.
that, for EG and GA, the present values of LOD and LOQ in
blood, which range from 3 to 50 mg/L, are similar to the
VOC
values obtained with GC–FID or GC–MS (4 to 25 mg/L) folThe ITEX2 method allowed the qualitative analysis of a comlowing derivatization with phenyl boronic acid or silylation
prehensive range of VOCs in blood without sample work-up
reagents (29). As the plasma EG concentration of 200 mg/L
other than the addition of an internal standard. Table III lists
has been considered the threshold of toxicity for systemic exposure (19,30) and plasma GA concentrations of 980 mg/L or higher have been
associated with renal injury (30), the present method clearly meets the sensitivity
and quantification range requirements.
In the treatment of EG poisoning,
hemodialysis has traditionally been indicated for patients with a serum EG
concentration greater than 500 mg/L (4),
but there are no established action limits
for GA (31). Very low LOD and LOQ have
been reported for EG and GA using direct
injection GC–MS following derivatization with bis-N,O-trimethylsilyl trifluoroacetamide (32), but according to those
Figure 3. ITEX1-GC–MS total ion chromatogram from a postmortem blood sample related to a fatal
authors, the values obtained are far
poisoning case due to ingestion of EG. Peak identification: 1, FA-d-ME; 2, acetonitrile; 3, EG-d4-2ME
below the limits seen in patient intoxiand EG-2ME; and 4, GA-d2-2ME and GA-2ME. The concentrations of EG and GA in blood were 1.4 and
cation.
2.1 g/L, respectively.
For FA, the LOQ obtained in blood was
30 mg/L (Table II). This value is lower
than the LOQ of 100 mg/L reported with GC–FID following
Table IV. Comparison of Identification Limits* for VOCs
derivatization with methanol in sulfuric acid (24,25). FA conin Blood by ITEX2 and Static HS Methods
centrations in postmortem blood between 600 and 1400 mg/L
have been found in 97% of 74 fatal methanol poisoning cases.
ITEX2
HS
Compound
(µg/L)
(µg/L)
Unlike methanol concentrations alone, FA concentrations have
been highly correlated with fatal outcome (24). In a clinical
Acetone†
0.9
100
context, serum methanol concentration over 500 mg/L is an inDesflurane
0.5
70
dication for hemodialysis (33), but there are no established acEnflurane
3
20
tion limits for FA (31).
Ethanol
15,000
120,000
Figure 3 shows an ITEX1-GC–MS total ion chromatogram
Ethyl acetate
5
30
from a postmortem blood sample related to a fatal poisoning
Diethyl ether
1
100
case due to ingestion of EG. The concentrations of EG and GA
Hexane
0.5
7
in blood were 1.4 and 2.1 g/L, respectively, and in urine were
Methanol
100
80,000
3.4 and 4.2 g/L, respectively.
MTBE
5
40
In-vial derivatization applied directly to the biological maSevoflurane
5
40
Toluene
2
20
terial is a key step in terms of ease of use and speed of the
ITEX1 method for hydroxylic organic compounds. In GC anal* Minimum required AMDIS spectral match factor set to 80 (maximum 100).
ysis, numerous derivatization methods are available for hy† Determined in a water sample because of endogenous acetone present in blood.
droxyl groups, such as acylation, alkylation, esterification, and
118
Journal of Analytical Toxicology, Vol. 34, April 2010
the retention times for 62 VOCs of toxicological relevance, intimes, independence from GC injector temperature and, imcluding the gas fuel components propane, butane, 2-methylportantly, lower LODs (40). Compared with a purge and trap
propane, and 2-methylbutane. A spectral library was created by
GC–MS method (41), higher LODs were obtained, but a benefit
ChemStation software and, after adding retention times in
of ITEX was a much shorter analysis cycle time.
seconds and CAS numbers, the library was converted to AMDIS
The identification limits obtained here for VOCs compared
library by the Lib2NIST converter. ITEX2 method was
favorably with those reported in the literature for blood using
amenable to semiquantitative analysis using diethyl ketone as
various extraction and analysis techniques (26,42). The ITEX2
an internal standard. However, for precise quantification, inmethod appears to be sensitive enough to reveal occupational
dividually selected internal standards should be used.
exposure well below acutely toxic levels. This is exemplified
AMDIS software proved to be an indispensable tool for recwith the workers exposed to toluene at low levels during proognizing target analytes in the full-scan GC–MS acquisition
duction of adhesive tapes, who exhibited a mean blood toluene
data. The program analyzes individual single ion chroconcentration of 18 µg/L (43). Another example relates to the
matograms to automatically identify possible component peaks
exposure to the common gasoline oxygenate methyl tert-butyl
within a data set. All masses that have the same chromatoether (MTBE) during tank truck loading. The fuel loaders were
graphic characteristics are then put forward as the unknown
found to have MTBE levels in blood averaging 13 and 19 µg/L
spectrum for a spectral search within a selected target library.
at colder and warmer ambient temperature, respectively (44).
The overall process involves four sequential steps: noise analysis, component perception, spectral deconvolution, and compound identification.
Conclusions
Identification limits of 11 VOCs based on AMDIS spectral
match were compared between ITEX2 and the corresponding
The ITEX1-GC–MS procedure is a viable method for the
static HS method. Default parameters were used with AMDIS
screening or quantitative confirmation of suspected EG poiwith the minimum required spectral match factor set to 80
sonings and for the confirmation of methanol findings from a
(maximum 100). Table IV shows that ITEX2 was superior to the
standard HS blood alcohol analysis. Delayed sampling folcorresponding static HS with all compounds tested. The GC
lowing ingestion of a toxic alcohol may result in undetectable
peak shape was generally good, but further improvement in the
EG or methanol but still measurable GA or FA in biological
analysis of some very volatile and early eluting compounds
could probably be obtained by cryofocusing. The fairly high identification
limit of ethanol compared to methanol
was mainly due to a non-optimal chromatographic peak shape of ethanol.
Figure 4 shows an ITEX2-GC–MS
analysis of a postmortem blood sample
related to a case of fatal poisoning due to
inhalation of butane. The main findings,
as verified from mass spectra and retention times, were butane, isobutane, and
propane.
As ITEX is a new microextraction technique, not much literature on its applications is currently available. A recent,
thorough study reported the optimization and evaluation of a fully automated
ITEX method for the enrichment of
volatile organic hydrocarbons from
aqueous samples prior to GC–MS analysis (40). The technique was found to be
comparable to other in-needle extraction
techniques in terms of the governing
parameters for method optimization.
These include extraction temperature,
the number of extraction cycles, extraction flow rate and volume, and desorpFigure 4. ITEX2-GC–MS total ion chromatogram and the corresponding mass spectra from a postmortem
tion flow rate and volume. The advanblood sample related to a fatal poisoning case due to inhalation of butane. Findings: propane (peak 1,
tages provided by ITEX over the
spectrum A), isobutane (peak 2, spectrum B), butane (peak 3, spectrum C), and internal standard diethyl
commonly used HS-SPME were lower
ketone (peak 4).
fragility, longer extraction phase life-
119
Journal of Analytical Toxicology, Vol. 34, April 2010
fluids. The present approach is the first practical HS method
for the quantitative analysis of EG, GA, and FA in biological material. The ITEX2-GC–MS allows screening for a wide range of
VOCs to reveal exposure to fuel gases, gasoline, alcohols, solvents, glues, anesthetic agents, and organic nitrites. The procedure has a sufficiently high performance to disclose the majority of harmful agents in blood at subtoxic concentrations. As
findings of EG, methanol, and VOCs are rather infrequent, the
shared analytical set-up used for ITEX1 and ITEX2 is advantageous in utilizing the sophisticated instrumentation in a costeffective manner. Because of the full-scan MS data acquisition
and AMDIS capabilities, the current procedure can be extended
to analyze any other low molecular weight compound that is
sufficiently volatile or can be volatilized by methylation.
References
1. K.L. Clay and R.C. Murphy. On the metabolic acidosis of ethylene
glycol intoxication. Toxicol. Appl. Pharmacol. 39: 39–49 (1977).
2. D. Jacobsen, N. Ostby, and J.E. Bredesen. Studies on ethylene
glycol poisoning. Acta Med. Scand. 212: 11–15 (1982).
3. D. Jacobsen, S. Ovrebø, J. Ostborg, and O.M. Sejersted. Glycolate
causes the acidosis in ethylene glycol poisoning and is effectively removed by hemodialysis. Acta Med. Scand. 216: 409–416
(1984).
4. J. Brent. Current management of ethylene glycol poisoning. Drugs
61: 979–988 (2001).
5. P.M. Leth and M. Gregersen. Ethylene glycol poisoning. Forensic
Sci. Int. 155: 179–184 (2005).
6. A.D. Fraser. Importance of glycolic acid analysis in ethylene
glycol poisoning. Clin. Chem. 44: 1769–1770 (1998).
7. G. Martin-Amat, K.E. McMartin, S.S. Hayreh, M.S. Hayreh, and
T.R. Tephly. Methanol poisoning: ocular toxicity produced by
formate. Toxicol. Appl. Pharmacol. 45: 201–208 (1978).
8. D. Jacobsen and K.E. McMartin. Antidotes for methanol and ethylene glycol poisoning. J. Toxicol. Clin. Toxicol. 35: 127–143 (1997).
9. The International Programme on Chemical Safety (IPCS). Environmental Health Criteria 196. http://www.inchem.org/documents/ehc/ehc/ehc196.htm. Accessed July 2009.
10. J.D. Osterloh, S.M. Pond, S. Grady, and C.E. Becker. Serum formate concentrations in methanol intoxication as a criterion for
hemodialysis. Ann. Intern. Med. 104: 200–203 (1986).
11. M.E. Field-Smith, B.K. Butland, J.D. Ramsey, and H.R. Anderson.
Report 21: Trends in death associated with abuse of volatile substances 1971–2006, July 2008. St. George’s University of London,
London, U.K. http://www.vsareport.org. Accessed July 2009.
12. A.W. Jones and J. Schuberth. Computer-aided headspace gas
chromatography applied to blood-alcohol analysis: importance of
online process control. J. Forensic Sci. 34: 1116–1127 (1989).
13. W.H. Porter and A. Auansakul. Gas-chromatographic determination of ethylene glycol in serum. Clin. Chem. 28: 75–78 (1982).
14. R.J. Flanagan, S. Dawling, and B.M. Buckley. Measurement of
ethylene glycol (ethane-1,2-diol) in biological specimens using
derivatisation and gas-liquid chromatography with flame ionisation detection. Ann. Clin. Biochem. 24: 80–84 (1987).
15. N.B. Smith. Identification and elimination of an ethylene glycol
determination artifact. Clin. Chim. Acta 162: 105–108 (1987).
16. P. Houzé, J. Chaussard, P. Harry, and M. Pays. Simultaneous determination of ethylene glycol, propylene glycol, 1,3-butylene
glycol and 2,3-butylene glycol in human serum and urine by
wide-bore column gas chromatography. J. Chromatogr. 619: 251–
257 (1993).
17. W.H. Porter, M.C. Jarrells, and D.H. Sun. Improved specificity for
ethylene glycol determined as the phenylboronate by capillary
column gas chromatography. Clin. Chem. 40: 850–851 (1994).
120
18. H.H. Yao and W.H. Porter. Simultaneous determination of ethylene glycol and its major toxic metabolite, glycolic acid, in serum
by gas chromatography. Clin. Chem. 42: 292–297 (1996).
19. W.H. Porter, P.W. Rutter, and H.H. Yao. Simultaneous determination of ethylene glycol and glycolic acid in serum by gas chromatography–mass spectrometry. J. Anal Toxicol. 23: 591–597
(1999).
20. A.D. Fraser and W. MacNeil. Colorimetric and gas chromatographic procedures for glycolic acid in serum: the major toxic
metabolite of ethylene glycol. J. Toxicol. Clin. Toxicol. 31: 397–405
(1993).
21. C.L. Moreau, W. Kerns, II, C.A. Tomaszewski, K.E. McMartin,
S.R. Rose, M.D. Ford, and J. Brent. Glycolate kinetics and
hemodialysis clearance in ethylene glycol poisoning. META Study
Group. J. Toxicol. Clin. Toxicol. 36: 659–666 (1998).
22. C. Abolin, J.D. McRae, T.N. Tozer, and S. Takki. Gas chromatographic head-space assay of formic acid as methyl formate in biologic fluids: potential application to methanol poisoning.
Biochem. Med. 23: 209–218 (1980).
23. A.D. Fraser and W. MacNeil. Gas chromatographic analysis of
methyl formate and application in methanol poisoning cases.
J. Anal. Toxicol. 13: 73–76 (1989).
24. G.R. Jones, P.P. Singer, and K. Rittenbach. The relationship of
methanol and formate concentrations in fatalities where methanol
is detected. J. Forensic Sci. 52: 1376–1382 (2007).
25. H.R. Wallage and J.H. Watterson. Formic acid and methanol
concentrations in death investigations. J. Anal. Toxicol. 32: 241–
247 (2008).
26. S.M. Wille and W.E. Lambert. Volatile substance abuse—postmortem diagnosis. Forensic Sci. Int. 142: 135–156 (2004).
27. I. Ojanperä, K. Pihlainen, and E. Vuori. Identification limits for
volatile organic compounds in the blood by purge-and-trap
GC–FTIR. J. Anal. Toxicol. 22: 290–295 (1998).
28. EURACHEM/CITAC Guide GC4: Quantifying Uncertainty in Analytical Measurement, 2nd ed, S.L.R. Ellison, M. Rosslein, and
A. Williams, Eds. EURACHEM/CITAC, 2000, http://www.
measurementuncertainty.org/index.html (October 2009).
29. R. Hess, M.J. Bartels, and L.H. Pottenger. Ethylene glycol: an
estimate of tolerable levels of exposure based on a review of
animal and human data. Arch. Toxicol. 78: 671–680 (2004).
30. J. Brent, K. McMartin, S. Phillips, K.K. Burkhart, J.W. Donovan,
M. Wells, and K. Kulig. Fomepizole for the treatment of ethylene
glycol poisoning. N. Engl. J. Med. 340: 832–838 (1999).
31. A.D. Fraser, L. Coffin, and D. Worth. Drug and chemical metabolites in clinical toxicology investigations: the importance of ethylene glycol, methanol and cannabinoid metabolite analyses. Clin.
Biochem. 35: 501–511 (2002).
32. P. van Hee, H. Neels, M. de Doncker, N. Vrydags, K. Schatteman,
W. Uyttenbroeck, N. Hamers, D. Himpe, and W. Lambert. Analysis of gamma-hydroxybutyric acid, DL-lactic acid, glycolic acid,
ethylene glycol and other glycols in body fluids by a direct injection gas chromatography–mass spectrometry assay for wide
use. Clin. Chem. Lab. Med. 42: 1341–1345 (2004).
33. D.G. Barceloux, G.R. Bond, E.P. Krenzelok, H. Cooper, and
J.A. Vale. American Academy of Clinical Toxicology Ad Hoc
Committee on the Treatment Guidelines for Methanol Poisoning.
American Academy of Clinical Toxicology practice guidelines
on the treatment of methanol poisoning. J. Toxicol. Clin. Toxicol.
40: 415–446 (2002).
34. M. Rompa, E. Kremer, and B. Zygmunt. Derivatisation in gas
chromatographic determination of acidic herbicides in aqueous
environmental samples. Anal. Bioanal. Chem. 377: 590–599
(2003).
35. P.L. Neitzel, W. Walther, and W. Nestler. In situ methylation of
strongly polar organic acids in natural waters supported by ionpairing agents for headspace GC–MSD analysis. Fresenius J. Anal.
Chem. 361: 318–323 (1998).
36. M.N. Sarrion, F.J. Santos, and M.T. Galceran. In situ derivatization/solid-phase microextraction for the determination of
haloacetic acids in water. Anal. Chem. 72: 4865–4873 (2000).
Journal of Analytical Toxicology, Vol. 34, April 2010
37. M.I. Catalina, J. Dallüge, R.J.J. Vreuls, and U.A.T. Brinkman.
Determination of chlorophenoxy acid herbicides in water by in
situ esterification followed by in-vial liquid-liquid extraction combined with large-volume on-column injection and gas chromatography–mass spectrometry. J. Chromatogr. A 877: 153–166
(2000).
38. L. Araujo, J. Wild, N. Villa, N. Camargo, D. Cubillan, and A. Prieto.
Determination of anti-inflammatory drugs in water samples, by in
situ derivatization, solid phase microextraction and gas chromatography–mass spectrometry. Talanta 75: 111–115 (2008).
39. N. Li, C. Deng, and X. Zhang. Determination of methylmalonic
acid and glutaric acid in urine by aqueous-phase derivatization
followed by headspace solid-phase microextraction and gas chromatography–mass spectrometry. J. Sep. Sci. 30: 266–271 (2007).
40. M.A. Jochmann, X. Yuan, B. Schilling, and T.C. Schmidt. In-tube
extraction for enrichment of volatile organic hydrocarbons from
aqueous samples. J. Chromatogr. A 1179: 96–105 (2008).
41. E. Martínez, S. Lacorte, I. Llobet, P. Viana, and D. Barceló. Multicomponent analysis of volatile organic compounds in water by
automated purge and trap coupled to gas chromatography–mass
spectrometry. J. Chromatogr. A 959: 181–190 (2002).
42. F. Pragst. Application of solid-phase microextraction in analytical
toxicology. Anal. Bioanal. Chem. 388: 1393–1414 (2007).
43. T. Kawai, H. Ukai, O. Inoue, Y. Maejima, Y. Fukui, F. Ohashi,
S. Okamoto, S. Takada, H. Sakurai, and M. Ikeda. Evaluation of
biomarkers of occupational exposure to toluene at low levels. Int.
Arch. Occup. Environ. Health 81: 253–262 (2008).
44. S. Vainiotalo, K. Pekari, and A. Aitio. Exposure to methyl tert-butyl
ether and tert-amyl methyl ether from gasoline during tank lorry
loading and its measurement using biological monitoring. Int.
Arch. Occup. Environ. Health 71: 391–396 (1998).
Manuscript received August 21, 2009;
revision received November 4, 2009.
121