CLIN. CHEM. 29/11, 1959-1960 (1983)
Effect of Short-Term Storage Conditions on Alcohol Concentrations in Blood
from Living Human Subjects
Charles L Winek and Louette J. Paul
We examined the effects of time, temperature, and a preservative (sodium fluoride) on ethanol concentrations in stored
samples of whole blood from living human subjects. We
measured the ethanol in the first, second, seventh, and 14th
day of storage, by gas chromatography. Samples were
stored at 0-3 #{176}C
and at 22-29 #{176}C,
with and without preserva-
tive. None of these showed significant gains or losses in
concentration. The average differences between ethanol as
measured on the day of collection and after storage were all
within the range of experimental error of the method (±5%).
AdditIonalKeyphrases: ethanol
medicine
sample handling
forensic
Recently, the increasing number of traffic accidents and
fatalities caused by individuals who drive while under the
influence of alcohol has resulted in an increase
in the
number
of medicolegal
proceedings held to establish the
sobriety of the accused. Results of determinations
of blood
ethanol are often the deciding factor between innocence and
guilt, so these values must be accurate.
Current laws in
many states and countries do not provide for standardized
collection and storage of specimens taken for such purposes,
nor is the interval before analysis specified. Thus conditions
under which specimens are stored during this interval
can
vary, including storage in a police officer’s desk or in a police
vehicle. What effect this has on the ethanol content has not
been fully established. The possibility of in vitro synthesis of
ethanol in such specimens has been raised, as well as loss
due to evaporation or adsorption of the ethanol onto the
rubber stopper.
Using one-month-old blood-bank blood to which ethanol
and sodium fluoride (10 g/L of blood) were added, Brown et
al. (1) demonstrated
that the factors most affecting stored
blood samples to be used for ethanol determination
were the
duration and temperature
of storage and the concentration
of preservative.
Studies of freshly collected blood samples
have shown that concentrations
do not change in preserved
samples stored at room temperature
for up to two months or
in refrigerated
samples stored for up to 10 months (2).
Reportedly,
ethanol concentrations
will remain stable for
six months when preserved samples are stored at -20 #{176}C
(3),
but effects of temperature
and time for unpreserved
samples
were not evaluated.
Thus the question of the stability of ethanol remains
incompletely
resolved,
because
studies
have not simulated
the actual conditions under which samples are submitted for
legal purposes.
Therefore, we undertook to determine the
effects of time, temperature, perservative, and delayed analToxicology Department,
Allegheny County Department of Laboratories, 542 Fourth Ave., Pittsburgh, PA i52i9; and Department
of Pharmacology-Toxicology, Duquesne University, Pittsburgh, PA
15219.
Received April i9, i983; accepted July 18, i983.
on the concentration
of ethanol in blood samples from
living persons, as submitted by law-enforcement
agencies
and hospitals without altering their normal collection procedure, and to evaluate the effects of these same variables in
blood samples collected from volunteers.
ysis
Materials and Methods
Apparatus
A Sigma 4B gas chromatograph
equipped with a flame
ionization detector (all from Perkin-Elmer Corp., Norwalk,
CT 06856) was used for ethanol quantification. The chromatograph
was connected to a Sigma 10 console (PerkinElmer), which calculated from the detector
response the
amount of ethanol in a sample.
Standards
De-ionized
water was used throughout. The 8.04 g/L stock
internal
standard
consisted
of 5.0 mL of n-propanol (ACS
grade) diluted to 500 mL with water. A 3.86 g/L working
internal standard
solution was made by diluting 24.0 mL of
the stock internal standard solution to 50 mL with water.
Procedures
Quantitative
analysis
for ethanol in blood samples.
Add
100 L of blood to a 5.0-mL test tube, with a micropipette.
With the same tip, add 100 L of water, using water to rinse
any remaining sample from the tip. With a fresh tip, add
100 L of the working internal standard to the same tube,
also rinsing with 100 L of water, and also adding to the
same tube. Stopper the tube and vortex-mix for 3 to 5 s.
Using a microliter syringe (Hamilton
Co., Reno, NV 89504),
inject 1 pL into the gas chromatograph.
Prepare all samples
in duplicate
and inject each in triplicate.
The gas chromatograph was equipped with a 6 ft. (182.88
cm) x 1/8 in. (3.17 mm) stainless steel column packed with a
0.2% Carbowax 1500 on 80/100 mesh Carbopak
(Supelco,
Inc., Bellefonte,
PA 16823). The instrument
conditions
were
as follows: column temperature,
85 #{176}C;
injection
port and
detector temperature,
225 #{176}C;
carrier gas (helium) flow rate,
35 mL/min, and pressure, 36 psi; hydrogen pressure,
22 psi;
air pressure, 32 psi; attenuation,
10 x 16. The detector
response to n-propanol and ethanol was converted to peak
areas by the Sigma 10 console, which then calculated
the
ethanol
concentration of the sample from the ethanol/npropanol
peak-area
ratio.
Study
of samples
collected
under
noncontrolled
conditions.
These whole-blood samples were from clinical cases submitted to this laboratory by hospital and various law-enforcement agencies for ethanol determination. We had no control
over any of the above-mentioned collection conditions. Upon
receipt of a case, the date, time of collection, and the
presence and identity of additives,
if any, were noted. Each
case was assigned
a consecutive number; if a case was
submitted
with more than one blood sample, the same
CLINICAL
CHEMISTRY,
Vol. 29, No. 11, 1983
1959
number was assigned to all samples of that case. The day of
collection of a sample was designated
as day 1 and considered the first day of storage. Samples from cases that were to
be stored refrigerated
(0-3 #{176}C)
were promptly placed in the
refrigerator.
On the day a refrigerated
sample was to be
analyzed, 100 L was taken for analysis without allowing
the sample to reach room temperature, and the sample was
resealed
and
replaced
in the refrigerator.
The 100-zL alis described above.
quot was diluted with internal standard
For these specimens stored at room temperature,
the
ambient temperature was continually monitored. They were
sampled as described above, and the specimen was resealed
and placed back on the laboratory bench.
Samples
from 30 cases were stored under refi-igeration
and analyzed
on the day of collection (day 1), then reanalyzed on the second, fifth, seventh,
and 14th day of
storage.
Samples from 42 cases stored at room temperature
were
also analyzed
on days 1, 2, 5, 7, and 14. The ethanol
concentration
was recorded and the percent differences
between day 1 values and values after storage were calculated.
Study
of blood specimens
collected
under
controlled
condi-
tions. Six volunteers
drank
between six and eight 12-oz.
cans of beer (4% ethanol) during a 1.5- to 2.0-h period.
Thirty
minutes after the last drink, venous blood was
sampled into 4.0-mL sterile Vacutainer
Tubes (Becton Dickinson, Rutherford,
NJ 07070). We took 20 samples from each
volunteer, 10 of which were collected with 100 USP units of
sodium
heparmn as an anticoagulant.
The remaining
10
samples were collected with 8 mg of potassium oxalate as an
anticoagulant
and 10 mg of sodium fluoride as a preservative. Five of the 10 samples that did not contain preservative
were stored under refrigeration,
five at room temperature.
Similarly, five of the 10 samples with preservative were
stored refrigerated and five at room temperature.
The day of collection was designated
as day 1. One of the
five samples in each storage condition per volunteer
was
analyzed
immediately
(day 1); the remaining four samples
in each storage condition were stored until analysis, i.e.,
either day 2, 5, 7, or 14.
Results
Samples
collected
under
noncontrolled
conditions.
We
could detect no significant changes in ethanol content,
either in the 30 nonpreserved
samples
collected
under
noncontrolled conditions and stored under refrigeration
(03#{176}C)
or in the 42 nonpreserved
samples
collected
under
controlled conditions and stored at room temperature
(2229 #{176}C).
The greatest change was 0.62% for the refrigerated
samples,
0.23% for the samples kept at room temperature
(Table 1).
Samples collected under controlled conditions. Again, we
could detect no significant changes in ethanol content of the
Table 1. Effect of Storage on Concentrations of
Ethanol In Blood: Clinical Samplesa
RetrIerMed
Room temp
Aver. %
Aver, ethanol
Aver. %
Day
Aver, ethanol
dIfference5
dIfference5
concn, g/L
analyzed concn, g/L
1
2.00
1.96
0.23 ± 2.18
2
2.00
0.62 ± 1.87
1.95
1.95
0.11 ± 2.30
5
1.99
0.03 ± 3.21
1.95
-0.01 ± 2.88
7
1.99
-0.35 ± 2.87
0.34 ± 3.01
1.94
-0.23 ± 2.52
14
2.00
n = 30 for refrigerated samples and 42 for samples kept at room
temperature.
5The percentage change in the ethanol concentration since day 1.
1960 CLINICALCHEMISTRY,Vol.29, No. 11, 1983
Table 2. Effect of Storage on Concentrations of
Ethanol In Blood of Six Beer-Drinking Volunteers
Room temp
RefrIgerated
Day
analyzed
1
2
5
7
NaF
added
+
Aver.
ethanol
concn,
/L
Aver. %
dlflerence
0.88
0.87
concn,
g/L
Aver. %
dltterence
0.87
-
-
+ 1.27(± 2.60)
-0.44 (± 2.81)
0.85
-0.48 (± 2.1)
+0.44 (± 1.51)
+
0.87
0.87
0.88
0.86
0.86
+2.01
+0.64(±
+
0.88
0.88
0.87
0.85
0.87
0.86
-1.05
-1.31
-0.62
+0.08
0.84
0.86
-2.12 (± 1.61)
-0.65 (± 1.89)
+
0.89
14
Aver.
ethanol
+
0.90
0.87
a Changesinceday 1.
-
+2.53
+ 1.57
+0.98
+ 1.59
(± 2.5)
2.64)
(± 3.12)
(± 2.38)
(± 3.33)
(± 3.60)
-
(±
(±
(±
(±
2.07)
2.21)
3.3)
2.22)
blood specimens obtained from the six beer-drinking
volunteers regardless
of the storage conditions
(Table 2). The
greatest change was 0.97% for the fluoridated-refrigerated
samples, 4.21% for the fluoridated-room
temperature samples. The greatest change was 1.97% for the nonfluoridatedrefrigerated
samples,
-1.14% for the nonfluoridated-room
temperature
samples.
Discussion
Extensive
studies have been conducted on the stability of
ethanol in blood sampled postmortem.
Postmortem
synthesis of ethanol was previously demonstrated
in our laboratory
by Collom (4); significant ethanol production was observed
in rabbit blood after 14 days of carcass storage at room
temperature.
Avbel (5) studied nonpreserved
postmortem
human-blood
samples
stored under refrigeration
(3 #{176}C)
for
18 months to two years, observing increases and decreases
in ethanol content. The decreases were attributed to oxidation and (or) evaporation, the increases to postmortem
synthesis of ethanol by microbial
fermentation
of glucose.
If there were such variations
in alcohol content in stored
blood sampled
from inebriated
living human
beings, it
would present a serious forensic implication:
blood samples
taken from living human
subjects for the legal testing of
sobriety would need to be evaluated immediately; otherwise,
a legal defense could challenge the validity of the results.
From our experiments,
we conclude that alcohol analyses
of blood obtained
sterilely
from living
humans
can be
delayed for as long as 14 days without a significant
change
in alcohol content. This holds true whether the blood sample
is refrigerated
or not, or whether a preservative is added to
the sample.
References
1. Brown GA, Neylan D, Reynolds WJ, Smaildon KW. The stability
of ethanol in stored blood. Part 1: Important variables and interpretations of results. Anal Chim Ada 66, 271-283 (1973).
2. Glendening BL, Waugh TC. The stability of ordinary blood
alcohol samples held various periods of time under different conditions. J Forensic Sci 10, 192-200 (1965).
3. Meyer T, Monge PK, Sakshaug J. Storage of blood samples
containing alcohol. Ada Pharinacol Toxicol 45, 282-286 (1979).
4. Collom WD. Postmortem synthesis of alcohol. In The Toxicology
Annual, 1974. CL Winek, Ed., Marcel Dekker Inc., New York, NY,
1975, pp 269-274.
5. Avbel AJ. Some factors affecting the analytical determination of
ethanol in human blood and tissues. Master’s thesis, Duquesne
University Graduate School, Pittsburgh, PA, 1972.
6. Blume P, Lakatua DJ. The effect of microbial contamination of
the blood sample on the determination of ethanol levels in serum.
Am J Clin Pathol 60, 700-702 (1973).