Forensic Science International 121 (2001) 7688

Analysis of fatty acid ethyl esters in hair as possible markers

of chronically elevated alcohol consumption by headspace
solid-phase microextraction (HS-SPME) and gas
chromatography-mass spectrometry (GC-MS)
F. Pragst*, V. Auwaerter, F. Sporkert, K. Spiegel
Institute of Legal Medicine, Humboldt University, Hannoversche Strasse 6, D-10115 Berlin, Germany

Abstract
Fatty acid ethyl esters (FAEE) are products of the nonoxidative ethanol metabolism, which are known to be detectable in
blood only about 24 h after the last alcohol intake. After deposition in hair they should be suitable long-term markers of
chronically elevated alcohol consumption. Therefore, a method for the analysis of ethyl myristate, ethyl palmitate, ethyl oleate
and ethyl stearate from hair was developed based on the extraction of the hair sample by a dimethylsulphoxide (DMSO)/nhexane mixture, separation and evaporation of the n-hexane phase and application of headspace solid-phase microextraction
(HS-SPME) in combination with gas chromatography-mass spectrometry (GC-MS) to the extract. For use as internal
standards, the corresponding D5-ethyl esters were prepared. The HS-SPME/GC-MS measurements were automatically
performed using a multi-purpose sampler. The detection limits of the FAEE were between 0.01 and 0.04 ng/mg and the
reproducibility was between 3.5 and 16%. By application of the method to hair samples of 21 fatalities with known heavy
alcohol abuse 0.0452.4 ng/mg ethyl myristate, 0.3513.5 ng/mg ethyl palmitate, 0.257.7 ng/mg ethyl oleate and 0.05
3.85 ng/mg ethyl stearate were measured. For social drinkers (3060 g ethanol per week), the concentrations were about one
order of magnitude smaller. For 10 teetotalers negative results or traces of ethyl palmitate were found. It was shown by
supplementary investigations in single cases that FAEE are also present in sebum, that there is no strong difference in their
concentrations between pubic, chest and scalp hair, and that they are detectable in hair segments after a 2 months period of
abstinence. From the results follows that the measurement of FAEE concentrations in hair is a useful way for a retrospective
detection of alcohol abuse. # 2001 Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Alcohol markers in hair; Fatty acid ethyl esters; Gas chromatography-mass spectrometry; Hair analysis; Headspace solid-phase
microextraction

1. Introduction
Hair analysis has proved to be a suitable tool for the
retrospective detection of the consumption of illicit or
therapeutic drugs. The physiological basis, analytical methods and practical applications were described in books [1,2],
reviews [3,4] and special issues of journals on the occasion
of hair conferences [5]. Different from these advantages up
to now there is no practicable method for the detection of
chronically elevated alcohol consumption by hair analysis,
*
Corresponding author. Tel.: 49-30-2093-7320;
fax: 49-30-2093-7268.
E-mail address: fritz.pragst@charite.de (F. Pragst).

although ethanol is abused most frequently and has highest

blood concentrations of all substances. Possible alcohol
markers in hair and rst investigations for their analytical
measurement were reviewed in a previous paper [6].
Between them some promising results were obtained for
ethyl glucuronide in rst investigations by Sachs [7] and by
Aderjan, Skopp et al. [8,9], but the results of a recent closer
examination of the presence of this ethanol metabolite in
hair and skin horny layer of alcoholics by the same authors
were rather discouraging [10]. First experiments to detect
acetaldehyde modied proteins in hair of alcohol fed
animals were described by Jelinkova et al. [11] and Watson
et al. [12], but this method is also still far from practical
application.

0379-0738/01/$ see front matter # 2001 Elsevier Science Ireland Ltd. All rights reserved.
PII: S 0 3 7 9 - 0 7 3 8 ( 0 1 ) 0 0 4 5 6 - X

F. Pragst et al. / Forensic Science International 121 (2001) 7688

77

Fig. 1. Formation routes and hydrolysis of fatty acid ethyl esters (FA: fatty acids; FAEE: fatty acid ethyl esters).

In orientating investigations the fatty acid ethyl esters

(FAEE) ethyl palmitate, ethyl oleate and ethyl stearate were
detected in methanol extracts of hair of alcoholics [6]. The
enzymatic formation of FAEE after alcohol consumption
was rst discovered by Lange et al. in 1981 [13] and later
thoroughly investigated mainly by Laposata and Laposata
and coworkers [1425]. According to these results, FAEE
are formed in presence of ethanol from free fatty acids,
triglycerides, lipoproteins or phospholipids by action of the
specic cytosolic and microsomal FAEE synthases as well as
of unspecic enzymes such as carboxylesterase, lipoprotein
lipase, carboxylester lipase and cholesterol esterase [18]
(Fig. 1). FAEE synthase activity was detected in liver,
pancreas, myocard, adipose tissue, different regions of brain
or white blood cells. FAEE are regarded as a reason for the
alcohol induced organ damage by different pathogenic
mechanism [18], such as incorporation and disordering of
organic bilayers, uncoupling of oxidative phosphorylation,
increase of lysosomale fragility or decrease of protein
synthesis and cell proliferation. The primary and terminal
half-life in blood is about 3 and 10 h, respectively [23].
Therefore, in blood they can be used as markers of an actual
or recent alcohol intake at least 24 h after completion of
ethanol intake [22]. In adipose tissues of rats a half life of
16  1:6 h was determined [15].
For volunteers with a blood ethanol concentration of
1.0% the FAEE concentrations in serum were 0.4
0.75 mg/ml [24]. In other studies between 1.8 and 3.8 mg/
ml in serum [20] and between 3 and 50 mg/ml in blood
[16,17,20] were described. In alcoholics postmortem, the
FAEE concentrations were 40500 mg/g in heart tissue [26],
4170 mg/g in brain tissue [26] and 90  14 mg/g in fat tissue
[15]. In meconium of an alcohol exposed infant 13.1 mg/ml
were measured [27].
The measurement of the FAEE concentrations was performed in most cases according to a method of Bernhardt
et al. [25] by liquidliquid extraction with acetone/hexane
(2/5, 1/3 or 2/8, v/v), purication of the extract on aminopropyl columns and identication by gas chromatography
with ame ionization detection (GC-FID) [27,28] or with
mass spectrometric detection (GC-MS) [17,20,24,25] using
ethyl heptadecanoate as internal standard. For separation of
individual FAEE, gas chromatography was applied after
purication by high-performance liquid chromatography
[25]. Tissue samples were extracted with ethyl acetate

and the concentrated extract suspension was directly

injected into the GC-FID instrument [26]. Alternatively,
thin layer chromatography on silica gel 60 plates was used
for extract purication [16,23]. In most cases, ethyl laurate,
ethyl myristate, ethyl palmitate, ethyl palmitoleate, ethyl
stearate, ethyl oleate, ethyl linoleate and ethyl arachidonate
were analyzed. In serum, they are bound mainly to albumin
or lipoproteins [17].
Like other substances, FAEE should be deposited in hair
from blood. Until present experiments for detection of FAEE
synthase activity in hair roots were not described in literature. However, it can be expected from the generally high
enzyme activity in these cells that FAEE should be synthesized also in the basal cells of the hair roots or in the
surrounding tissues. A third way of deposition could be
the incorporation from sebum, although FAEE were also not
yet detected in this secretion. In order to examine whether
fatty acid ethyl esters are suitable hair markers of alcohol
abuse an analytical method for their quantitative analysis
from hair was developed and applied to hair samples of
alcoholics, social drinkers and teetotalers. Headspace solidphase microextraction (HS-SPME) in combination with
capillary gas chromatography and mass spectrometry
(GC-MS) was chosen as a suitable technique, which was
successfully applied for the detection of volatile and semivolatile lipophilic drugs from hair in previous investigations
[2932].
2. Material and methods
2.1. Hair samples
The scalp hair samples of alcoholics analyzed in this
investigation were collected from fatalities who were postmortem examined at the Institute of Legal Medicine of the
Humboldt University, Berlin. The history of heavy alcohol
consumption was known from the police reports and at the
autopsy, the corresponding pathologic symptoms of the
internal organs were found. The hair samples of social
drinkers were collected from colleagues and friends together
with data about their drinking behavior and habits of hair
care on a questionnaire. The hair samples of teetotalers were
obtained from children and adults who denitely did not
drink any alcoholic beverage.

78

F. Pragst et al. / Forensic Science International 121 (2001) 7688

For the development of the method, a pool of hair of 25

alcoholics was prepared. The hair was washed as described
below, cut to pieces of 13 mm length and thoroughly mixed
for 3 days in a shaker. In the same way a hair pool of
teetotalers was prepared from children hair.
In order to remove external contamination and sebum, all
hair samples were thoroughly washed before analysis with
an aqueous solution of 0.1% dodecyl sulfate and deionized
water and dried at room temperature. This investigation was
allowed by the ethics commission of the university hospital
Charite of the Humboldt University.
2.2. Reagents and deuterated standards
Ethyl myristate, ethyl pentadecanoate, ethyl palmitate,
ethyl stearate and ethyl oleate as well as the corresponding
free carboxylic acids were obtained from Sigma-Aldrich
(Deisenhofen, Germany). All other reagents were obtained
in analytical grade purity from Merck (Darmstadt, Germany).
The deuterated standards D5-ethyl myristate, D5-ethyl
palmitate, D5-ethyl stearate and D5-ethyl oleate were prepared according to Blau and Dabre [33] in the following way.
In a 5 ml test tube, 50 ml D6-ethanol were added to 10 mg of
the free acid. The mixture was cooled to 788C with dry ice.
Then 10 ml thionyl chloride were added and the closed vessel
was allowed to warm up and held at 408C for 2 h. After that
excessive SOCl2 and D6-ethanol were removed by a nitrogen
stream at room temperature. For complete removal of
remaining traces of the reagents the residue was three times
dissolved in 1 ml n-hexane and the solvent again removed
with the nitrogen stream. The oily residues (1112 mg) were
dissolved in chloroform to obtain a concentration of 2 mg/
ml. These stock solutions were stored at 048C. For analysis
a mixture of the four deuterated esters (concentration 2 mg/
ml of each) was prepared by dilution with chloroform. The
identity of the four D5-ethyl esters was proven by the mass
spectra. By control with GC-MS in the scan mode was
shown that no residual free fatty acids or measurable
amounts of other impurities were present.
2.3. Instruments
A gas chromatograph 6890 with a mass selective detector
5973 (Hewlett-Packard GmbH, Waldbronn, Germany) was
used for the GC-MS measurements. This was combined with
a multi-purpose sampler MPS 2 (Gerstel, Muhlheim/Ruhr,
Germany), with which all steps of the HS-SPME experiments (preheating of the sample and headspace adsorption
in the heating station with or without sample agitation,
desorption in the GC injection port) could be programmed
and automatically carried out. The SPME experiments were
performed with a 65 mm polydimethylsiloxane/divinylbenzene ber (PDMS/DVB) from Supelco (Deisenhofen, Germany) tting to the MPS 2. A mixer 5433 with a rack for 24
samples (Eppendorf, Hamburg, Germany) was used for the

liquid extraction of the hair samples. The solvents were

removed by an evaporator of the rm Liebisch (Bielefeld,
Germany).
2.4. Sample preparation and solid-phase microextraction
After optimization (cf. Section 3.1), the following procedure was used. To 50 mg of the washed and dried hair pieces
in a 4 ml glass vessel with screw cap and Teon septum 25 ml
of the solution of the four deuterated standards in chloroform
(concentration 2 mg/ml of each), 0.5 ml dimethylsulfoxide
and 2 ml n-hexane were added. The vessels were tightly
closed and vigorously shaken with the Eppendorf mixer for
14 h. After centrifugation the hexane phase was separated and
transferred into a 10 ml headspace vial. The solvent was
completely evaporated at 408C by a nitrogen stream.
To the residue 1 ml phosphate buffer (0.1 M, pH 7:6)
and 0.5 g NaCl were added. The vials were tightly closed
with silicon/PTFE septa and steel caps and placed into the
vial rack of the multi-purpose sampler. For HS-SPME the
following conditions were used. Preheating 5 min at 908C
and 250 rpm agitation, headspace adsorption 30 min at 908C
and 150 rpm agitation, desorption 15 min at 2608C. The
agitation mode was 60 s right, 30 s interval, 60 s left, 30 s
interval, etc.
2.5. Gas chromatography-mass spectrometry
An HP5-MS capillary column (28 m  0:25 mm  0:25
mm, Hewlett-Packard) with helium 5.0 as carrier gas
(1.0 ml/min) was used for the gas chromatographic separation. The injection mode was splitless for 3 min. The following temperature program was applied: 2 min at 1008C,
then 208C/min up to 3008C, then 5 min at 3008C. The
temperatures of the injector, the interface, the ion source
and the quadrupol were 260, 280, 230 and 1068C, respectively. The retention times under these conditions and the m/z
values chosen for the detection in the selected ion mode
(SIM, three time windows) are given in Table 1.
Table 1
Retention times and m/z values chosen for the GC-MSSIM
analysis of FAEE and their D5-analogs (three time windows)
Compound

Retention time
(min)

Values of m/z in
SIM mode

D5-Ethyl myristate
Ethyl myristate

9.71
9.92

93, 106, 162, 218, 261

88, 101, 157, 213, 256

D5-Ethyl palmitate
Ethyl palmitate

10.73
10.75

93, 106, 162, 246, 289

88, 101, 157, 241, 284

D5-Ethyl oleate
Ethyl oleate

11.57
11.58

93, 106, 315

88, 101, 310

D5-Ethyl stearate
Ethyl stearate

11.67
11.68

93, 106, 162, 274, 317

88, 101, 157, 269, 312

F. Pragst et al. / Forensic Science International 121 (2001) 7688

3. Results and discussion

3.1. Optimization of the analytical method
Fatty acid ethyl esters are unpolar substances, which are
easily hydrolyzed in basic or alkaline medium. Therefore,
they must be extracted with lipophilic solvents under mild
conditions from the hair matrix and the sample preparation
by alkaline or enzymatic digestion of the hair used for other
compounds cannot be applied. Different from the one-step
method previously used for a series of basic drugs [2932] in
this case, the method consists of two steps: (i) the liquid
extraction of the hair sample and (ii) the selective separation
of the esters from this extract on the basis of their relatively
high volatility by HS-SPME and the measurement by
GC-MS.
At rst the HS-SPME conditions and the GC-MS analysis
were optimized in absence of hair. After that the liquid
extraction of the hair sample was studied and the calibration
and validation of the method were carried out with spiked
hair samples. All experiments were performed twice
(n 2). The difference between both results was always
smaller than 15% of the mean values, which are given in the
gures and tables.
3.1.1. GC-MS and HS-SPME conditions
In HS-SPME, the analyzed compounds are adsorbed at
moderate temperature on a coated silica ber, which is
placed in the vapor phase above the sample [34,35]. After
that, the ber is transferred into the injection port of the gas
chromatograph and the compounds are desorbed at higher
temperature for GC-MS analysis. For these investigations, a
65 mm PDMS/DVB was chosen since this coating is particularly suitable for lipophilic semivolatile compounds.
3.1.1.1. Internal standards and mass spectra. For quantitative measurements with HS-SPME always internal
standards must be used, because of the variability of the
GC-MS sensitivity and of the strong dependence of the
absolute HS-SPME extraction yields on the experimental
conditions. The most reliable results are generally obtained
with deuterated standards. Therefore, the D5-ethyl myristate,
D5-ethyl palmitate, D5-ethyl oleate and D5-ethyl stearate
were prepared from the corresponding carboxylic acids and
D6-ethanol.
The mass spectra are shown in Fig. 2 in comparison to the
corresponding non-deuterated esters. It can be seen that the
D5-ethyl esters have suitable mass spectroscopic properties
for the use as internal standards in these investigations. In the
case of the three saturated esters the molecular peaks and the
peaks m=z 93 (88), 101 (106), 162 (157) and M-43 show
the difference of ve mass units without an essential mutual
contribution by other fragment peaks. Only in the case of
ethyl oleate, which has much more fragment peaks, the SIM
analysis must be restricted to the molecular peak 315 (310)
and to the fragment peaks 93 (88) and 106 (101), for which a

79

certain mutual contribution must be taken into account. In

Table 1, the m/z values used for the GC-MSSIM detection
are given. In the analysis of real samples the retention time
and the characteristic peak area ratios in the single ion
chromatograms of these m/z values were used for identication of the esters. For quantication only the peaks of the
molecular ions were used, which always displayed a sufcient intensity and were only to a minor extent disturbed by
other matrix constituents.
3.1.1.2. Composition of the liquid phase for HS-SPME. For
the optimization of the HS-SPME conditions, the absolute
signal intensity is important. Therefore, these measurements
were carried out in absence of the internal standards. The
liquid phase from which the hair extract is analyzed should
avoid the ester hydrolysis, promote their evaporation and
retain impurities. For this purpose, 1 ml of a neutral aqueous
phosphate buffer (0.1 M, pH 7.6) was chosen. At this pH, the
esters should be stable even at elevated temperatures and the
disturbing free carboxylic acids (pK  5), which are present
in hair lipids to a dominant extent [36], should be in the nonvolatile ionic form. Since, generally, the evaporation from
aqueous solution is improved by a salt-out effect, also
measurements in presence of 0.5 g Na2SO4, NaCl and
Na2HPO4 were performed. Surprisingly, in the case of
these lipophilic compounds, the highest peak areas were
measured in absence of salt. However, in the investigation of
hair extracts instead of the pure reference substances equal
or better results were obtained in presence of the salts.
Therefore, the routine measurements were carried out
with 0.5 g NaCl.
3.1.1.3. Adsorption temperature and adsorption time. The
effects of the adsorption temperature and of the adsorption
time on the peak area were measured in the range from 40 to
1208C and 10 to 60 min, respectively. As expected, the
volatility decreased with increasing length of the aliphatic
chain. Whereas, ethyl myristate was extracted already at
408C to a considerable extent, the extraction of ethyl stearate
and ethyl oleate started not below 608C. For the routine
measurements, 908C were chosen since at higher
temperatures the reproducibility was worse.
From the dependence of the peak area on the adsorption
time followed that the extraction equilibrium between the
liquid phase and the ber for all four esters was attained after
30 min. Therefore, this time was chosen for routine investigations. At longer extraction time a slight decrease of the
peak areas occurred, which can be explained by a slow ester
hydrolysis.
3.1.2. Liquid extraction of the hair sample
The FAEE should be located together with other hair
lipids in the cell membrane complex of the hair matrix.
Furthermore, they should also be present in the sebum
layer on the hair surface. For a retrospective clarication
of alcohol consumption, particularly, the esters deposited

80

F. Pragst et al. / Forensic Science International 121 (2001) 7688

Fig. 2. Mass spectra of ethyl myristate, ethyl palmitate, ethyl oleate and ethyl stearate in comparison to those of their D5-ethyl analogues.

in the structural lipids should be important. Therefore, the

hair was always thoroughly washed with a dodecyl sulfate
solution in order to remove the external lipids.
The solvent used for the extraction must have a sufcient
swelling effect on the hair matrix and must sufciently

dissolve the esters. It was shown in a previous study that

non-polar solvents like hexane or acetone are unsuitable for
hair extraction, since they do not swell the hair matrix [37].
Methanol could be a reasonable compromise and was used in
the preliminary experiments [6]. However, it was found that

F. Pragst et al. / Forensic Science International 121 (2001) 7688

the ethyl esters are partly transformed into methyl esters

under these conditions. For this reason methanol and other
alcohols were excluded.
Alternatively two-phase mixtures of n-hexane as a good
solvent for the FAEE with ethylene glycol, DMSO or an
aqueous buffer (pH 7.6), which have swelling properties for
hair, were examined. Furthermore, n-hexane was applied in
combination with an aqueous solution of 8 M urea and 0.2 M
mercaptoacetate, which causes a cleavage of the disulde
bonds in the hair. The best results were obtained with DMSO
(relative yield B 100%) and the urea/mercaptoacetate solution (98%), whereas, the aqueous buffer and ethylene glycol
led to much smaller yields between 47 and 71%. For these
reasons and in order to minimize ester hydrolysis, further
experiments were performed with the aprotic n-hexane/
DMSO mixture.
For control of the completeness of the extraction in a
series of experiments 50 mg of the alcoholic hair pool were
extracted with this mixture at eight different extraction times
between 3 and 24 h. The results are shown in Fig. 3. For ethyl
palmitate, the extraction was completed after 6 h, but for the
other three esters, the yield increased till 12 h and after that
was nearly constant. For practical application, the extraction
time of 14 h should always be sufcient. This was conrmed
by a second extraction of the same hair samples by which
only less than 4% of the rst 14 h extraction were obtained.
In a third extraction the results were negative.
From the long extraction time follows that the esters are
deposited not only close to the surface, but also in deeper
compartments of the hair. It cannot be decided from these
experiments whether there remains a residue of FAEE in
hair, which is not accessible by this solvent combination
as it was described for other compounds before [38]. It
was tried to improve the extraction by use of an ultrasonic
bath and by grinding the hair pieces, but this lead to a higher

81

chromatographic underground and to a decrease of the

sensitivity.
3.1.3. Effect of the sample amount and absolute HS-SPME
extraction yields
It was found in previous investigations that the efciency
of the HS-SPME decreases with increasing sample amount
[30,32]. In the case of the FAEE, where only the lipid extract
of the hair sample is analyzed by HS-SPME instead of the
dissolved total hair, the same effect was found. This results
from a series of experiments in which the extracts from 0, 10,
20, 50 and 100 mg of the teetotaler hair pool were spiked
with 40 ng of the four FAEE (Fig. 4). The absolute HSSPME extraction yields were calculated from the peak areas
obtained in the HS-SPME experiment in comparison to the
peak areas measured after direct injection of the same
substance amount. It is seen from Fig. 4 that the highest
HS-SPME yields (3239%) were found with a sample
amount of 5 mg. Above that the yields decreased and were
only between 5 and 10% at a sample amount of 100 mg. As
an explanation, it is assumed that constituents of the hair
extract with surface activity like anions of free fatty acids
increase the solubility of the esters in the aqueous phase and
decrease the distribution coefcient between SPME ber
and sample. Since the amount of lipids extracted from
different hair samples strongly varies [36], the effect of
the matrix on the HS-SPME yield should also vary from
sample to sample. This was conrmed in the application of
the method to hair samples of alcoholics, social drinkers and
teetotalers (Section 3.2), where for the same concentration of
the deuterated internal standards from different samples
quite different peak areas were measured.
In routine analysis, 50 mg were chosen as an optimal
sample amount. However, because of the higher extraction
yield also with 10 or 20 mg the sensitivity is sufcient.

Fig. 3. FAEE concentrations extracted from 50 mg of the alcoholic hair pool with n-hexane/DMSO at different extraction times (n 2).

82

F. Pragst et al. / Forensic Science International 121 (2001) 7688

Fig. 4. Effect of the sample amount on the HS-SPME extraction yields of the FAEE. A 1 ml phosphate buffer pH 7.6, 0.5 g NaCl and 40 ng of
each of the four FAEE were added to the extract from 0, 10, 20, 50 to 100 mg of the teetotaler hair pool.

3.1.4. Calibration, limits of detection (LOD), limits of

quantication (LOQ) and reproducibility
For calibration 50 mg of the teetotaler hair pool were
spiked with 14 concentrations between 0.01 and 15 ng/mg of
the four FAEE and analyzed in presence of 40 ng of each of
the D5-FAEE as internal standards according to the procedure described in the experimental part. For quantication
only the molecular ions were used. The calibration curves
were linear up to 2.5 ng/mg. The characteristic data and the
correlation coefcients are listed in Table 2. At higher
concentrations, the curves were slightly bent to lower peak
area ratios. The reason for this deviation from linearity could
not be cleared.
According to the German Industrial Norm (DIN 32645)
the LOD and the LOQ were estimated as the 3- and 11-fold
standard deviation of the base line noise, respectively. The
data are also given in Table 2. The LOD values were between
0.01 ng/mg for ethyl stearate and 0.04 ng/mg for ethyl
oleate. The corresponding LOQ values were between 0.04
and 0.12 ng/mg. These values were conrmed by measurement of spiked hair samples in the concentration range
between 0.01 and 0.1 ng/mg.

The reproducibility of the method was examined by

measuring the alcoholic hair pool 10 times within 2 days.
The mean concentrations and the standard deviations are
also given in Table 2. The relative standard deviation for
ethyl myristate at the concentration of 0.067 ng/mg was
15.5%. For the other esters with higher concentrations
between 3.5 and 6.6% were found.
3.2. Application to hair samples of alcoholics, social
drinkers and teetotalers
The method was applied to the hair samples of 21 fatalities, which were known to be heavy alcoholics from the
police reports and from the typical pathologic changes of the
internal organs. In all cases the hair was investigated in full
length. As a typical example in Fig. 5, the SIM chromatograms of a hair sample with medium concentrations are
shown. It is seen from the ion chromatograms that (within
the usual limits of error) the peak area ratio for the m/z values
of esters from the sample (e.g. ethyl palmitate, m=z 88,
101 and 284) is the same as for the deuterated standard (D5ethyl palmitate, m=z 93, 106 and 289). As generally found

Table 2
Calibration curves, LOD, LOQ and reproducibility in the analysis of FAEE in haira
FAEE

Ethyl
Ethyl
Ethyl
Ethyl
a
b

c a(Asample/Astandard) b

myristate
palmitate
oleate
stearate

LOD
(ng/mg)

R2

1.094
1.135
1.083
1.169

0.024
0.037
0.023
0.014

0.9993
0.9990
0.9987
0.9937

0.015
0.02
0.04
0.01

A: peak area in the SIM chromatogram of the molecular ions.

The 10 measurements of the alcoholic hair pool in one series.

LOQ
(ng/mg)

0.05
0.07
0.12
0.04

Reproducibility (n 10)b
Medium concentration
(ng/mg)
0.067
0.796
0.387
0.133

S.D.
ng/mg

0.010
0.052
0.023
0.005

15.7
6.6
5.9
3.5

F. Pragst et al. / Forensic Science International 121 (2001) 7688

83

Fig. 5. GC-MSSIM chromatogram of the hair extract of case 169/99 obtained after HS-SPME. Measured concentrations: ethyl myristate
0.13 ng/mg, ethyl palmitate 1.31 ng/mg, ethyl oleate 2.76 ng/mg and ethyl stearate 0.37 ng/mg. Deuterated standards 0.8 ng/mg. For control of
the method 1.0 ng/mg ethyl pentadecanoate was added.

for deuterated standards, the retention times are slightly

smaller for the D5-FAEE than for the non-deuterated analytes. As in this case, the identity of the FAEE is unambiguously proven for all alcoholic samples by the peak area ratio
of the measured m/z values and by the exact agreement of the
retention time.
The hair concentrations of the four FAEE measured for
all alcoholics are given in Table 3 together with the postmortem alcohol concentrations in blood and urine and
the cause of death. The esters were detected in all cases.
The concentrations differ strongly from case to case between
0.045 and 2.4 ng/mg for ethyl myristate, 0.35 and 13.5 ng/
mg for ethyl palmitate, 0.25 and 7.1 ng/mg for ethyl oleate
and 0.05 and 3.85 ng/mg for ethyl stearate. In all cases,
ethyl palmitate and ethyl oleate have the highest concentrations, as it was also found in the analysis of blood
samples [22].
The cause of death was not in all cases in a direct context
to the alcohol abuse. Only two subjects died from acute
alcohol overdose. In two cases, hypothermia and in one case
blood loss during the acute alcohol intoxication were stated.

Three subjects died from acute withdrawal syndrome. In ve

cases, the cause of death can directly be related to the organ
damage by chronic alcohol abuse. The residual eight cases
were suicides or other diseases, which are not in a direct
context to the alcohol consumption, but might indirectly
have been favored by it. In 15 of the 21 cases, alcohol was
not detected in blood. Therefore, the analysis for FAEE in
hair is an additional useful diagnostic tool in order to prove
heavy drinking behavior in fatalities.
For comparison, the hair samples of 10 children or adult
teetotalers and of 10 social drinkers with an alcohol consumption between three and six alcohol units (3060 g
ethanol) per week were analyzed. The results are given in
Table 4. The chromatogram of an 11-year-old child is shown
in Fig. 6. Beside the peaks of the internal standards no peaks
of the four FAEE are seen. For three of the teetotalers traces
of ethyl palmitate were found below the LOQ, the origin of
which is not clear.
For the social drinkers only ethyl palmitate was detected
in all cases with a concentration range between LOD and
0.40 ng/mg (mean 0.12 ng/mg), which is clearly below that

84

F. Pragst et al. / Forensic Science International 121 (2001) 7688

Table 3
Alcohol concentrations in blood and urine, cause of death and concentrations of FAEE in scalp hair of fatalities with known heavy alcohol
abusea
Case
No.

Age
(sex)

169/98
252/98
325/98
4/99
88/99
156/99
169/99
223/99
451/99
453/99
102/00
132/00
140/00
173/00
185/00
188/00
232/00
241/00
260/00
276/00
287/00
a

45
57
35
40
59
43
57
65
51
40
34
65
50
40
64
38
45
46
49
56
53

(m)
(m)
(f)
(m)
(m)
(m)
(m)
(m)
(m)
(m)
(m)
(m)
(m)
(m)
(f)
(m)
(m)
(m)
(m)
(m)
(f)

Postmortem alcohol
concentration (mg/g)
Blood

Urine

0.1
0.0
1.4
0.0
0.0
2.3
1.5
3.2
4.8
0.0
0.1
0.1
0.0
5.1
0.1
0.0
0.0
0.0
0.0
0.0
0.0

0.2
0.0

0.0
0.0
2.2

3.1
6.1

0.1
0.0
0.0
5.6
0.1
0.0
0.1

0.0
0.1

Cause of death

FAEE in hair (ng/mg)

Hepatic cirrhosis
Pneumonia and lung cancer
Hypothermia during alcoholic intoxication
Withdrawal syndrome, ketoacidosis
Hypothermia during alcoholic intoxication
Alcoholic intoxication and loss of blood
Suicide by jumping from 11th floor
Suicide by electricity in bathtub
Acute alcohol intoxication
Withdrawal syndrome
Suicide by electricity in shower cubicle
Hepatic cirrhosis
Hyperglycemic coma at known diabetes
Acute alcohol intoxication
Heavy coronary arteriosclerosis
Withdrawal syndrome and liver failure
Pontine myelolysis
Bronchopneumonia
Hepatic cirrhosis, pancreatitis
Gastric hemorrhage, hyperglycemia
Streptococcus infection

Ethyl
myristate

Ethyl
palmitate

Ethyl
oleate

Ethyl
stearate

0.47
0.22
0.08
0.03
0.09
0.27
0.13
2.44
0.88
0.72
0.045
0.42
0.19
0.20
0.11
0.17
0.38
0.11
0.11
0.06
0.14

1.73
1.37
0.39
0.57
0.40
1.25
1.31
13.5
3.76
2.28
0.46
1.17
0.49
1.15
0.35
0.86
3.33
0.60
0.45
0.73
0.99

1.09
1.23
0.65
0.68
0.40
2.14
2.76
7.07
6.88
2.75
0.25
1.47
1.25
2.89
0.62
1.08
8.94
0.79
1.02
1.23
1.01

0.22
0.35
0.10
0.25
0.08
0.19
0.37
3.85
0.68
0.36
0.05
0.28
0.11
0.27
0.10
0.26
1.39
0.26
0.15
0.47
0.47

m: male; f: female.

Table 4
Concentrations of FAEE in hair of teetotalers (T) and social drinkers (SD)
No.

Age (sex)

T01
T02
T03
T04
T05
T06
T07
T08
T09
T10
SD01
SD02
SD03
SD04
SD05
SD06
SD07
SD08
SD09
SD10
a
b

9
11
13
15
30
32
54
56
65
68
19
22
25
25
26
26
29
30
35
35

(m)
(m)
(f)
(m)
(f)
(f)
(f)
(f)
(f)
(f)
(m)
(f)
(f)
(m)
(f)
(m)
(m)
(f)
(f)
(f)

Drinking behaviora

Concentrations in hair (ng/mg)b

Beverage units per week

Ethyl myristate

Ethyl palmitate

Ethyl oleate

Ethyl stearate

0
0
0
0
0
0
0
0
0
0
6
6
5
6
6
5
3
3
6
6

n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
pos.
n.d.
n.d.
n.d.
pos.
n.d.
pos.
0.05
pos.
n.d.

n.d.
n.d.
n.d.
n.d.
n.d.
pos.
pos.
pos.
n.d.
n.d.
0.40
pos.
0.07
0.12
0.09
0.08
0.07
0.22
pos.
pos.

n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
0.32
n.d.
pos.
pos.
pos.
n.d.
0.12
0.23
0.13
pos.

n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
0.12
n.d.
n.d.
n.d.
n.d.
n.d.
pos.
0.04
n.d.
n.d.

Self-reported data, one beverage unit is 250 ml beer, 100 ml wine or 20 ml high percent spirits.
Here, n.d.: not detected; pos.: between LOD and LOQ (cf. Table 2); m: male; f: female.

F. Pragst et al. / Forensic Science International 121 (2001) 7688

85

Fig. 6. GC-MSSIM chromatogram of the hair extract of a child (11 years, male, No. T04 in Table 4) obtained after HS-SPME. Deuterated
standards 0.8 ng/mg. For control of the method 1.0 ng/mg ethyl pentadecanoate was added. FAEE from the sample were not detected and
should be seen at the retention times marked by arrows.

of the alcoholics (mean 1.69 ng/mg). Ethyl oleate was found

only in eight cases with concentrations up to 0.32 ng/mg,
and ethyl myristate and ethyl stearate were only in one or
two samples above the LOQ.
Since the FAEE should be present also in sebum at the
surface of unwashed hair of alcoholics each 50 mg of the
samples 102/00 and 173/00 were rinsed with cold water,
dried and two times washed with 2 ml n-hexane. To the nhexane washing solutions the internal standards were added,
the solvent was evaporated and the residues (1.31 mg in case
102/00 and 1.42 mg in case 173/00) were analyzed in the
same way as the hair extracts. After that the hair was
extracted with DMSO/n-hexane and analyzed for the FAEE.
In Fig. 7a, the concentrations of the FAEE from the washing
solution and the hair extract are compared. In both cases, the
FAEE amount obtained from sebum is lower than that
obtained from hair and ethyl oleate is present to a higher
degree in sebum than in the hair matrix.
In case 241/00 chest hair and pubic hair were investigated
in addition to scalp hair. The concentrations are shown in
Fig. 7b. The highest concentrations were measured in pubic
hair, but the differences between the three sampling positions are not very distinctive and beside a smaller concen-

tration of ethyl myristate in chest hair the concentration ratio

of the different FAEE is almost the same.
The hair sample of a living female alcoholic (age 63) with
a previous periodic drinking behavior was analyzed in three
segments. She was in a withdrawal treatment and reported a
medium consumption of about 80 g ethanol per day during
the last months before she stopped drinking after an excessive episode exactly 2 months before sampling. The results
of the hair analysis are shown in Fig. 7c. There is a
signicant decrease of the FAEE concentrations to about
60% from the distal to the proximal segment, but the
expected nearly negative result for the proximal segment
(01 cm) is not found. A low concentration (up to 20% of the
distal segment) could be explained from resting or slowly
growing hair [6]. Since the claim of abstinence during the
last 2 months before sampling was very convincing the much
higher proximal concentration (about 60% of the distal
segment) may be caused by a retarded deposition into hair
from surrounding tissues.
Altogether, it can be concluded from the results that
FAEE are suitable markers for the detection of heavy alcohol
consumption by hair analysis. For alcoholics positive results
are also obtained after a longer period of abstinence.The

86

F. Pragst et al. / Forensic Science International 121 (2001) 7688

Fig. 7. Concentrations of FAEE in hair samples of alcoholics: (a) comparison of sebum and hair matrix for the fatalities 102/00 and 173/00;
(b) comparison of scalp hair, pubic hair and chest hair for the fatality 241/00; (c) segmental analysis of the hair sample of an alcoholic (63-oldfemale), which stopped drinking 2 months before sampling.

concentrations are clearly distinguishable from those of

social drinkers, and for teetotalers negative results are found.
The regions for the sum of all four esters found in these rst
investigations are given in Table 5. The analysis by HSSPME and GC-MS with deuterated internal standards
proved to be very reliable and sufciently sensitive for

the routine analysis of the FAEE in hair. With the multipurpose sampler MPS 2 the manual work was minimized
and the reproducibility was improved. Further investigations
are in progress to examine the applicability of the method in
forensic and clinical practice of the detection and treatment
of alcohol abuse.

Table 5
Drinking behavior and total FAAE concentrations in hair
Drinking behavior

Total concentration of FAEE (ng/mg)a

Teetotalers
Social drinkers (three to six drinking units per week)
Fatalities with known heavy alcohol consumption

0.0 (Traces of ethyl palmitate)

0.00.87 (Mean 0.27)
0.8126.9 (Mean 4.6)

Sum of the concentrations of ethyl myristate, ethyl palmitate, ethyl oleate and ethyl stearate.

F. Pragst et al. / Forensic Science International 121 (2001) 7688

Acknowledgements
The authors thank the Deutsche Forschungsgemeinschaft
(DFG) for supporting these investigations.

References
[1] P. Kintz (Ed.), Drug Testing in Hair, CRC Press, Boca Raton,
1996.
[2] A. Chatt, S.A. Katz, Hair Analysis Applications in the
Biomedical and Environmental Sciences, VCH Publishers,
New York, 1988.
[3] H. Sachs, P. Kintz, Testing for drugs in hair. Critical review of
chromatographic procedures since 1992, J. Chromatogr. 713
(1998) 147161.
[4] F. Pragst, M. Rothe, K. Spiegel, F. Sporkert, Illegal and
therapeutic drug concentrations in hair segments a
timetable of drug exposure? Forensic Sci. Rev. 10 (1998)
81111.
[5] P. Kintz (Ed.), Special issue, Proceedings of the Selected
Papers from the 2nd International Meeting of the Society of
Hair Testing, Forensic Sci. Int. 13 (2000) 107.
[6] F. Pragst, K. Spiegel, F. Sporkert, M. Bohnenkamp, Are there
possibilities for the detection of chronically elevated
alcohol consumption by hair analysis? A report about the
state of investigation, Forensic Sci. Int. 107 (2000)
201233.
[7] H. Sachs, Drogennachweis in Haaren, in: H. Kijewski (Ed.),
Proceedings of the Symposium on Das Haar als SpurSpur
der Haare, 24 November 1993, Goettingen, Schmidt-Romhild, Lubeck, 1997, pp. 119133.
[8] R. Aderjan, K. Besserer, H. Sachs, G. Schmitt, S. Skopp,
Ethyl glucuronide a non-volatile ethanol metabolite in
human hair, in: V. Spiehler (Ed.), Proceedings of the 1994
Joint TIAFT/SOFT International Meeting, 31 October4
November 1994, Tampa/Florida, DABFT Newport Beach,
CA, 1995, pp. 3945.
[9] G. Skopp, G. Schmitt, P. Dronner, R. Aderjan, Trinkverhalten
und Haaranalyse, in: T. Daldrup, F. Muhoff (Eds.),
Proceedings of the 1995 GTFCh Symposium, 2022 Aprol
1995, Mosbach/Baden, Verlag Dieter Helm, Eppenheim,
1995, pp. 175179.
[10] G. Skopp, G. Schmitt, L. Poetsch, P. Droenner, R. Aderjan, R.
Mattern, Ethyl glucuronide in human hair, Alcohol Alcoholism 35 (2000) 283285.
[11] D. Jelinkova, Z. Deyl, I. Miksik, F. Tagliaro, Capillary
electrophoresis of hair protein modied by alcohol intake in
laboratory rats, J. Chromatogr. A 709 (1995) 111119.
[12] R.R. Watson, D. Solkoff, J.Y. Wang, K. Seeto, Detection of
ethanol consumption by ELISA assay measurement of
acetaldehyde adducts in murine hair, Alcohol 16 (1998)
279284.
[13] L.G. Lange, S.R. Bergmann, B.E. Sobel, Identication
of fatty acid ethylesters as a product of rabbit myocardial
ethanol metabolism, J. Biol. Chem. 256 (1981) 12968
12973.
[14] E.A. Laposata, L.G. Lange, Presence of nonoxidative ethanol
metabolism in human organs commonly damaged by ethanol
abuse, Science 231 (1986) 497499.

87

[15] E.A. Laposata, D.E. Scherrer, L.G. Lange, Fatty acid ethyl
esters in adipose tissue. A laboratory marker for alcoholrelated death, Arch. Pathol. Labor Med. 113 (1989) 762766.
[16] M. Laposata, Z.M. Szczepiorkowski, J.E. Brown, Fatty acid
ethyl esters: non-oxidative metabolites of ethanol, Prostaglandins Leukotriens Essential Fatty Acids 52 (1995) 8791.
[17] D.A. Bird, A. Kabakibi, M. Lapasota, The distribution of fatty
acid ethyl esters among lipoproteins and albumin in human
serum, Alcoholism: Clin. Exp. Res. 21 (1997) 602605.
[18] M. Laposata, Fatty acid ethyl esters: ethanol metabolites
which mediate ethanol-induced organ damage and serve as
markers of ethanol intake, Prog. Lipid Res. 37 (1998) 307
316.
[19] M. Laposata, Fatty acid ethyl esters: nonoxidative ethanol
metabolites with emerging biological and clinical signicance, Lipids 34 (1999) 281285.
[20] B.L. Soderberg, E.T. Sicinska, E. Blodget, J.E. CluetteBrown, P.M. Suter, T. Schupisser, W. Vetter, M. Laposata,
Preanalytical variables affecting the quantication of fatty
acid ethyl esters in plasma and serum samples, Clin. Chem.
45 (1999) 21832190.
[21] M. Laposata, Fatty acid ethyl esters: current facts and
speculations, Prostaglandins Leukotriens Essential Fatty
Acids 60 (1999) 313315.
[22] L. Dan, M. Laposata, Ethyl palmitate and ethyl oleate are the
predominant fatty acid ethyl esters in the blood after ethanol
ingestion and their synthesis is differentially inuenced by the
extracellular concentrations of their corresponding fatty acids,
Alcohol Clin. Exp. Res. 21 (1997) 286292.
[23] K.M. Doyle, D.A. Bird, S. Al-Salihi, Y. Hallaq, J.E. CluetteBrown, K.A. Goss, M. Laposata, Fatty acid ethylesters are
present in human serum after ethanol ingestion, J. Lipid Res.
35 (1994) 428437.
[24] K.M. Doyle, J.E. Cluette-Brown, D.M. Dube, T.G. Bernhardt,
C.R. Morse, M. Lapasota, Fatty acid ethylesters in the blood
as markers of ethanol intake, J. Am. Med. Assoc. 276 (1996)
11521156.
[25] T.G. Bernhardt, P.A. Cannistraro, D.A. Bird, K.M. Doyle, M.
Laposata, Purication of fatty acid ethyl esters by solid-phase
extraction and high-performance liquid chromatography, J.
Chromatogr. B Biomed. Appl. 675 (1996) 189196.
[26] K. Yamazaki, T. Gilg, G. Kauert, L.v. Meyer, W. Eisenmenger, Nonoxidative ethanol and methanol changes in the heart
and brain tissue of alcohol abusers, Jpn. J. Legal Med. 51
(1997) 380387.
[27] J. Klein, T. Karaskov, G. Koren, Fatty acid ethyl esters: a
novel biological marker for heavy in utero ethanol exposure: a
case report, Ther. Drug Monit. 21 (1999) 644646.
[28] C.F. Bearer, S. Lee, A.E. Salvator, S. Minnes, A. Swick, T.
Yamashita, L.T. Singer, Ethyl linolate in meconium: a
biomarker for prenatal ethanol exposure, Alcohol Clin. Exp.
Res. 23 (1999) 487493.
[29] I. Koide, O. Noguchi, K. Okada, A. Yokoyama, H. Oda, S.
Yamamoto, H. Kataoka, Determination of amphetamine and
methamphetamine in human hair by headspace solid-phase
microextraction and gas chromatography with nitrogen
phosphorus detection, J. Chromatogr. B Biomed. Sci. Appl.
707 (1998) 99104.
[30] F. Sporkert, F. Pragst, Use of headspace solid-phase
microextraction in hair analysis for organic compounds,
Forensic Sci. Int. 107 (2000) 129148.

88

F. Pragst et al. / Forensic Science International 121 (2001) 7688

[31] F. Sporkert, F. Pragst, Determination of lidocaine in hair of

drug fatalities by headspace solid-phase microextraction, J.
Anal. Toxicol. 24 (2000) 316322.
[32] F. Sporkert, F. Pragst, Determination of methadone and its
metabolites EDDP and EMDP in human hair by headspace
solid-phase microextraction and gas chromatography-mass
spectrometry, J. Chromatogr. B Biomed. Appl. 746 (2000)
255264.
[33] K. Blau, A. Dabre, Esterication, in: K. Blau, J. Halket
(Eds.), Handbook of Derivatives for Chromatography, 2nd
Edition, Wiley, Chichster, 1993, p. 16.

[34] J. Pawliszyn, Solid Phase Microextraction Theory and

Practice, Wiley, New York, 1997.
[35] J. Pawliszyn (Ed.), Applications of Solid Phase Microextraction, The Royal Society of Chemistry, Cambridge, 1999.
[36] C.R. Robbins, Chemical and Physical Behaviour of Human
Hair, Springer, New York, 1994, pp. 71, 7984.
[37] M. Rothe, F. Pragst, Solvent optimization for the direct
extraction of opiates from hair samples, J. Anal. Toxicol. 19
(1995) 236240.
[38] W.A. Baumgartner, V.A. Hill, Sample preparation techniques,
Forensic Sci. Int. 63 (1993) 121135.