Journal of Analytical Toxicology, Vol. 20, January/February1996
Incorporation of Isotopically Labeled Cocaine and
Metabolites into Human Hair: 1. Dose-Response
Relationships
Gary t. Henderson*, Martha R. Harkey, and ChihongZhou
Department of Medical Pharmacologyand Toxicology, School of Medicine, University of California, Davis, CA 95616
Langley Porter Institute, University of California, San Francisco, CA 94 t 43
Abstract
Deuterium-labeled cocaine (cocaine-d s) was administered
intravenously and/or intranasally in doses of 0.6-4.2 mg/kg to 25
human volunteers under laboratory clinical conditions. Sequential
blood samples were collected for up to 3 days, and hair samples
were collected for up to 10 months. Samples were analyzed by gas
chromatography-mass spectrometry (GC-MS) for cocaine-d s and
its major metabolite, benzoylecgonine-d5 (BZE-ds). The parent
drug, cocaine-ds, was the predominant analyte in hair, whereas
BZE-ds was the major analyte in blood, especially at later time
periods. The amount of cocaine-d s incorporated into hair ranged
from 0.1 to 5 ng/mg hair, whereas the amount of BZE-ds was
approximately one-sixth of that concentration. The threshold dose
for detection was estimated to be 25-35 mg of drug administered
intravenously. A single dose could be detected for 2-6 months.
Subjects receiving the same dose differed (from two to 12 times as
much depending upon how it was measured) in the amount of
cocaine-d5 incorporated into their hair. Non-Caucasians, in
particular, incorporated more cocaine-d5 in hair than did
Caucasians. Also, segmental analysis of the samples revealed
considerable intersubject variability in the time drug first appeared
in hair and the rate at which the drug moved along the hair shaft
with time. These interindividual differences could not be explained
by differences in plasma pharmacokinetics. Considered together,
these results suggest that cocaine incorporation into hair may
occur by way of multiple mechanisms~by way of sweat and
sebum, for example--and at various times during the hair growth
cycle. Thus, hair analysis using GC-MS appears to be a very
sensitive method for detecting cocaine ingestion. However, within
the range of doses used in the present study, hair does not provide
a particularly accurate record of either the amount, time, or
duration of drug use.
Introduction
For decades,hair has been used as a biomarker for exposure
to environmental contaminants such as arsenic, lead, and mercury. High concentrations of these metals in hair correlate
*Author to whomcorrespondenceshould be addressed,
well with the signs and symptoms of toxicity. However, because of the great variability in testing methods and the wide
range of values reported for these metals in "normal" populations, there is less agreement on the value of hair analysis in
identifying cases of subchronic poisoning or evaluating the
health risk of populations exposed to environmental contaminants (1).
More recently, hair analysis has been used to detect the use
of both licit and illicit drugs. In 1979, Baumgartner et al. (2)
reported that radioimmunoassay could be used to detect
nanogram concentrations of morphine in the hair of heroin
abusers and suggested that the position of drug along the hair
shaft correlated with the time of drug use. Since then, hair has
been used by an increasing number of forensic laboratories
throughout the world as a specimen to detect drugs of abuse,
such as cocaine (3-8), methamphetamine (9-13), phencyclidine (14-16), and nicotine (17-21).
Many of these early studies used immunoassay screening
tests that were not confirmed by more specific methods, such
as gas chromatography-mass spectrometry (GC-MS), and
relied upon self-reported drug use to determine drug dosage
rather than controlled dose administration. As a result, the correlation between drug dose and concentration in hair has not
been well established. Recent studies (22) using controlled
doses of drugs and more specific methods have shown a better
relationship, but the numbers of subjects were small.
The objectives of the research reported herein were to
determine the following:the relationship between the dose of
cocaine and the concentration of parent drug and metabolites
in plasma and hair; the relationship between the time of
cocaine use and the position of cocaine or metabolites along
the hair shaft; and the time interval between cocaine use and
appearance of cocaine or metabolites in hair. Precise doses of
deuterium-labeled cocaine (cocaine-ds)were administered to
human volunteers under controlled laboratory conditions. A
specific and sensitive method, GC-MS, was used to assay the
plasma and hair samples. Because cocaine-d5is pharmacologically and metabolically identical to cocaine and produces a
unique response in a mass spectrometer, we were able to distinguish the cocaine administered in the laboratory from any
Reproduction(photocopying)of editorialcontentof thisjournalis prohibitedwithoutpublisher'spermission,
I
Downloaded from https://academic.oup.com/jat/article/20/1/1/750825 by guest on 20 April 2021
Reese T. Jones and Peyton Jacob, III
Journal of Analytical Toxicology, Vol. 20, January/February 1996
residual cocaine in the subjects' tissues or any cocaine taken
surreptitiously during the course of the study.
Experimental
Synthesis of deuterium-labeled cocaine
Cocaine hydrochloride labeled with five deuterium atoms on
the benzoyl group (deuterated cocaine hydrochloride, cocaineds, (benzoyl)-cocaine-d5 HCI) and suitable for intravenous and
intranasal use was synthesized according to the general
method of Bell and Archer (23). Briefly, unlabeled cocaine HCI
was hydrolyzed to ecgonine HCI by refluxing in dilute
hydrochloric acid; it was then esterified with methanol and
anhydrous hydrogen chloride to form ecgonine methyl ester
HCI. Ecgonine methyl ester HCI was converted to the free
base with aqueous potassium carbonate and benzoylated with
benzoylchloride-ds to yield crude (benzoyl)-cocaine-ds free
base. This was converted to (benzoyl)-cocaine-d5 HC] with one
equivalent of concentrated aqueous hydrogen chloride in
2-propanol-diethylether. Chemical purity was certified by
GC-MS, thin-layer chromatography (TLC), melting point, and
elemental analysis. There were no impurities detectable by
GC-MS or TLC.
The melting point was identical to the melting point of
Isotope effect of cocaine-d5
The positions of the deuterium label on the benzoyl group of
cocaine are not involved in the major routes of cocaine
metabolism, and thus, a significant isotope effect is unlikely.
Nevertheless, this was confirmed by the following experiment.
An additional seven subjects were recruited for the evaluation of an isotope effect of cocaine-ds. These subjects are not
included in Table I; however, their demographics and medical
histories are similar. A 50:50 mixture of labeled (5 mg/mL)
and unlabeled (5 mg/mL) cocaine was administered as a
0.6-mg/kg total dose (i.e., 0.3 mg/kg labeled and 0.3 mg/kg
unlabeled cocaine) given intravenously over a l-rain injection
time. Plasma and urine samples were analyzed by GC-MS. At
all sampling times, the concentrati~)ns of nondeuterated
cocaine and cocaine-ds were not significantly different. This
NCH.
0
benzoylecgonine
MW = 289
MW = 303
OOH3
NOI"I3
cocaine-d5
~OH
benzoylecgonine-d5
MW - 294
MW - 308
Figure 1. Chemicalformulasand molecularweightsof cocaine,benzoylecgonine,cocaine-ds,and benzoylecgonine-ds.
Table I. Demographics of the Human Volunteers Participating in the Studies
Mean age*
(years)
Mean body
weight*
(kg)
Gender
30
(21-39)
63.6
(54.5-68.2)
21 male,
4 female
"The range is in parenthesis.
Race
21 Caucasian,
4 non-Caucasian
Hair color
12 brown, 7 black,
4 graying,2 blonde
Hair
pattern
15 straight,
5 curly, 5 wavy
Cosmetic
treatments
Cocaineuse
3 dyed, 2 bleach,
1 straightened
13 heavy,
9 moderate,3 light
Downloaded from https://academic.oup.com/jat/article/20/1/1/750825 by guest on 20 April 2021
Research subjects
Twenty-five volunteers were recruited into the study. Moderate users (defined as use once or twice every 6 weeks to
weekly use) of cocaine were sought. None were cocaine dependent as judged by DSM III-R criteria or gave a history of
cocaine dependence in the past. Subjects were healthy, as
judged by medical, laboratory, and psychiatric evaluation. All
were able to give adequate informed consent. None of the subjects were HIV positive. Written consent was obtained from all
subjects. All protocols were approved by an institutional review
committee. All studies were performed in a general medical
hospital environment.
The demographics of the subjects are in Table I. Most subjects had straight brown hair, but there was a range of hair
colors, types, and textures. The hair of four subjects was described as "graying". Six of the subjects had treated their hair
with cosmetic products such as bleach, dyes, and straighteners. Subjects were questioned about how often they washed
their hair and their use of hair cosmetic products, but no attempt was made to control these variables.
unlabeled U.S. Pharmacopeial cocaine obtained from Mallinckrodt Corp.
Sterile solutions of labeled cocaine for intravenous injection
were prepared by dissolving the labeled cocaine in sterile saline
and then filtering the solution through a Millipore filtration
unit.
The benzoyl site was chosen for deuteration because the
benzoyl moiety remains intact during metabolic conversion to
the primary metabolite, benzoylecgonine. The increased mass
of pentadeuterated cocaine (molecular weight [MW], 308) is
sufficient to ensure its distinction from unlabeled cocaine
(MW, 303) by MS. The chemical formulas and molecular
weights of cocaine, cocaine-ds, BZE, and BZE-d5 are shown in
Figure 1.
Journal of Analytical Toxicology, Vol. 20, January/February 1996
pattern was observed for BZE and BZE-ds as well. Relevant
pharmacokinetic parameters, such as half-life, clearance,
volume of distribution, and amount of metabolites (benzoylecgonine and ecgonine methyl ester), were determined
from the measured concentrations of drug and metabolite in
the biofluids.
Table II shows half-lifeand clearance values for the test subjects. The mean values for plasma half-life and clearance of
nondeuterated and deuterated cocaine are not significantly
different and correlate with an r 2 of approximately 0.8, which
indicates that the pharmacokinetics of nondeuterated and
deuterated cocaine are similar.
Table II. Plasma Half-Life and Clearance Values for NonDeuterated and Deuterated Cocaine in Seven Subjects
Plasma half-life
Plasma clearance
(min)
(mL/minper kilogram)
Subject
cocaine
cocaine-d s
cocaine
cocaine-d s
1
83.6
81.1
21.5
19.3
2
61.7
61.4
32.8
28.9
3
81.1
102.4
30.8
22.4
4
37.8
41.6
23.0
21.5
5
83.9
76.9
15.5
15.1
6
44.5
46.7
24.5
26.5
7
78.3
68.9
17.6
15.2
Mean
SD*
67.3
19.5
68.4
20.9
23.7
6.4
21.3
5.5
* SD = Standarddeviation.
Procedures for collecting biological samples
Blood collection. For the pharmacokinetic studies, blood
was collected from a forearm vein through an indwelling
venous catheter that was removed at the end of the first study
day. On days 2-4, blood was obtained by venipuncture. Blood
samples, usually 8--10 mL, were obtained before and at 15, 30,
60, 180, and 360 min and at 24, 48, and 72 h after drug administration. Blood samples were placed in test tubes conraining 0.5 mL of a saturated solution of sodium fluoride to
Table Iii. Dosing Regimen for Single-Dose and MultipleDose Studies
Amount
cocaine-ds
(mg/dose)
Dose
cocaine-ds
(mg/kg)
No. of
subjects*
Single IVt
35-46
63-108
0.6
1.2
10
7
Single nasal
120-170
0.6*
6
Multiple doses
41 IV
270 nasal*
(4 months later)
0.6 IV
1.2 nasal*
Multiple doses
108 IV
170 nasal*
(7 months later)
1.2 IV
0.6 nasal*
Multiple doses
750 nasal*
(multiple doses
over 1 month)
3.0 nasal*
Multiple doses
108 IV
825 nasal*
(multipFe doses
over 1 month)
1.2 IV
3.0 nasal*
Dosing
regimen
1
* Somesubjectsreceivedmore than one dose regimen.
.t IV = Intravenous.
* Nasal doseslistedwere correctedfor bioavailability, estimatedto be 30%.
Downloaded from https://academic.oup.com/jat/article/20/1/1/750825 by guest on 20 April 2021
Drug administration and formulation
Intravenous administration. Cocaine for intravenous administration was prepared from synthesized deuterium-labeled
cocaine hydrochloride. A stock solution was prepared under
sterile conditions as a Millipore filtered 10-20% stock solution
in sterile 0.9% sodium chloride. The administered dose was
diluted to a volume of 3-7 mL for intravenous injection by
syringe pump. Stock solutions were refrigerated at 10~ and
were used within 1 month of preparation. As a quality-control
procedure, samples from each dose administered were assayed
to verify concentration and stability.
lntranasal administration. For intranasal doses, cocaine
was administered as a 20% solution of the hydrochloride salt
in saline deliveredas a fine mist to the nasopharynx. Although
snuffing of the crystalline material is the preferred route by
cocaine users and some researchers, in our laboratory we find
that the spray delivers a more reproducible dose than does
crystalline cocaine, and we find little difference in kinetics or
effectwhen the two dose forms are compared. The advantages
of using cocaine solution introduced by insufflation are better
control of the dose administered and reduced variability of
dose deliverydue to subject variability in the method of dosing.
A Macintosh Oxfordsprayer was modifiedso that the reservoir
held approximately I mL of solution and delivered a 150-mg
dose in about four squeezes. Assayof the residual material in
the spray system allowed precise determination of the actual
dose delivered to the nasopharynx.
Doses used. A number of dose ranges and a number of dose
regimens were used in the study (as shown in Table III). We
attempted to span as wide a range of doses as possible and
to simulate both one-time use and chronic cocaine use. For
the single-dose studies, an intravenous dose of 0.3, 0.6, or
1.2 mg/kg or an intranasal dose of 0.6 or 1.2 mg/kg was
administered.
To administer larger doses of cocaine safely,we gave either
multiple intranasal doses (spaced 3-5 days apart) or a single
intravenous dose followed by multiple intranasal doses. With
this dosing scheme, more than three-quarters of a gram of
cocaine (825 rag) was administered to one subject. This dosing
regimen allowed us to simulate, to a degree, chronic cocaine
use. Obviously, many chronic cocaine users use considerably
higher doses and for longer periods of time, but we were
limited by safety considerations for the research subjects.
Journal of Analytical Toxicology, Vol. 20, January/February1996
Analytical procedures
Quanfitation of cocaine-ds and metabolites in human hair.
A sensitive and specific GC-MS method for the simultaneous
detection and quantitation of cocaine and its two principle
metabolites was used. Details of the method have been published previously (24), but the essentials of the method are
described below.
Sample preparation and derivatization. Root ends of the
hair sample were aligned carefully, cut into 1-cm segments,
washed with 25 mL 1% SDS, and then rinsed with deionized
water (50 mL, 10 times) and then methanol (30 mL, three
times). Samples were allowed to drain completely between
each rinse and dried under a hood overnight before weighing.
Dried hair samples were cut into approximately 2-mm sections, and 10-rag aliquots were digested with proteinase K and
dithiothreitol in sodium acetate buffer at 40~ overnight. Cocaine-d5, cocaine, BZE-ds, BZE, and ecgonine methyl ester
(EME) were extracted from the digested hair samples using
Bond Elut CertifyTM columns. After adding the hair digests, the
columns were rinsed with deionized water, HC1, and MeOH.
Drugs were then eluted with methylene chloride-isopropyl alcohol (80:20) with 2% ammonium hydroxide. Extracts were
evaporated under N2 at 40~ reconstituted in methylene chloride, and then derivatized with N-methyl-N-(tertbutyldimethylsilyl)-trifluoroacetamide.
CC-MS. Derivatized hair extracts were analyzed by chemical
ionization MS using a Finnigan ITS-40 ion-trap mass spectrometer coupled to a Varian 3400 gas chromatograph fitted
with a DB-5 capillary column (15 m x 0.25-mm i.d.; 0.1-mm
film thickness). Helium was used as the carrier gas, and isobutane was the reagent gas. Our gas chromatographic conditions yielded retention times of 4.13 min for the tert-butyldimethylsilyl derivative of EME (EME-TBDMS), 5.62 rain for
the internal standard, 5.85 rain for cocaine-ds, and 6.90 min for
the tert-butyldimethylsilyl derivative of BZE-ds (BZE-dsTBDMS).
Quantitation. A five-point calibration plot was prepared daily
by analyzing 10 mg of drug-free hair samples fortified with
cocaine-ds, cocaine, BZE-ds, BZE, and EME at concentrations
of 0.1, 0.5, 1, 5, and 10 ng/mg hair. In addition, both positive
quality-control hair samples (prepared by fortifying drug-free
hair samples with cocaine-d5, BZE-d5,and EME at either 0.1 or
I ng/mg hair) and negative quality-control hair samples were
analyzed daily.
A single ion (MH§ was used for quantitation based on the
peak-area ratios of cocaine-d5, BZE-ds-TBDMS, and EMETBDMS to the internal standard, diflurococaine. Ions at m/z
308, 408, and 340 were used for cocaine-d5, BZE-ds-TBDMS,
and the internal standard, respectively. Ions at m/z 304, 404,
314, and 340 were used for cocaine, BZE-TBDMS,
EME-TBDMS, and the internal standard, respectively.Because
cocaine-d5 was one of the analytes, diflurococaine was used as
the internal standard rather than cocaine-d3, which is more
typically used. The detection limit was set at 0.1 ng/mg hair for
cocaine-ds and BZE-d5 and 0.5 ng/mg for EME, based on a
signal-to-noise ratio greater than or equal to 3. The cutoff of
EME was set at 0.5 ng/mg hair because of a small coeluting
peak observed in some negative control samples. The detection
limit of the cocaine analytes varied somewhat on the type of
hair used to produce the calibration curves. To improve the
ruggedness of the assay, coarse black Asian hair was used to
validate the method and prepare all standards and controls.
This hair type is more difficult to digest, and it produced a
higher chemical background than Caucasian hair.
Precision and recovery. For hair samples spiked with each
drug at 0.1 ng/mg hair, the observed mean plus or minus standard deviation (SD) and percent coefficient of variation (%CV)
AUC = Area 1 + A r e a 2 + Area 3 + Area 4
i~
r
v
Hair Sample
(100 mg)
.43
amples
U3
10
$
Tip
Root
8O
r
Seg 1 S e g 2
Seg3
Seg4
Seg5
Seg6
Seg7
O
E
,,r
Hair Segments (10 rag)
Figure 2. Diagram showing a typical hair sample sectioned into sequential
1-cm segments,which represent approximately 1 month's growth, prior to
GC-MS analysis.
0
1
2
3
Time post dose (months)
Figure 3. Illustration showing the calculation of the amount of cocaine-ds
incorporated into hair expressed as area under the curve (AUC).
Downloaded from https://academic.oup.com/jat/article/20/1/1/750825 by guest on 20 April 2021
prevent degradation of cocaine by blood esterases. The tubes
were mixed gently to ensure thorough mixing and iced immediately after filling. Plasma was separated and frozen at
-20~ within 1 h.
Hair collection. Hair samples were collected from the posterior vertex region of the scalp before each drug administration and at monthly intervals thereafter. At each sampling
time, approximately 100 hair strands (at least 100 mg per
subject) were collected by cutting the hair as close to the scalp
as possible (within i ram). Root ends of the hair sample were
aligned carefully, and the bundle of hair was cut into 1-cm segments. Each 1-cm segment corresponds to approximately 1
month's growth. The sampling and sectioning procedure is
illustrated in the diagram in Figure 2. Care was taken to orient
hair so that the proximal ends matched and could be identified
for precise segmental analysis. Hair sampling continued at
monthly intervals until 1 month past the time when drug was
no longer measurable in hair.
Journal of Analytical Toxicology, Vol. 20, January/February 1996
were as follows:0.11 • 0.02 ng/mg (18.2%) for cocaine, 0.09 •
0.02 ng/mg (22.2%) for BZE-TBDMS, and 0.14 • 0.04 ng/mg
(28.6%) for EME-TBDMS.At I ng/mg hair, the observed mean
plus or minus SD and %CVwere 1.07 + 0.11 ng/mg (10.3%),
1.10 + 0.18 ng/mg (16.3%), and 0.94 + 0.26 ng/mg (27.7%) for
cocaine, BZE-TBDMS, and EME-TBDMS, respectively. The
percentage recovery of cocaine and BZE was typically in the
range of 90.4-115.4%.
Presentation of hair analysis data
Currently, there is no universally accepted way of expressing
hair analysis data, although it is most often expressed in units
of concentration (e.g., nanograms of drug per milligram of
hair). However, in our studies, in which a bolus of drug was
given, this was not a useful unit because hair length varied
greatly between individuals. Subjects with the same amount of
drug incorporated into their hair would
have different concentrations because of difTable IV. Segmental Analysis Resultsfor Subject Number 88173
fering hair lengths.
Demonstrating the Calculation of the Area Under the Curve (AUC) by the
Some investigators correct for differing
Trapezoidal Rule*
hair lengths by using a standardized hair
Hair
Amount
sample such as the first 3 cm of hair from
collected
cocaine-d 5
Amount cocaine-d s in segment (ng)
the
root. Theoretically, only drug ingested
(months post
in sample
segment segment segment segment segment
5
during
the past few months would be predose)
(ng)
1
2
3
4
sent in this section of hair. However, our
0.26
4.01
1.20
2.82
sampling times were too long (up to 10
1.17
2.40
1.26
0.54
0.38
0.22
months in some individuals) and the hair
2.20
1.89
0.83
0.38
0.25
0.19
0.24
lengths too variable to use this method.
3.13
0.68
0.11
0.23
0.34
Further, we found that after a single dose of
cocaine, drug was not always confined to a
* Maximum amount of cocaine-ds was 4.01 ng. The AUC of cocaine-ds was 6.84 ng"months.
discrete area adjacent to the root. In some
subjects, drug was distributed over multiple
segments extending far from the root.
1000
! 000
~"
We chose to express our data in terms of
-~
E
800
A
~ 800
concentration of drug in hair only when
8oo.
~ 8oo
B
hair samples were too small to perform
'~
0.8 .~Q/.o iv
segmental analysis. More typically,we chose
o 400
nn 400
. . . . . .
to express our data in terms of amount of
drug found in hair.
h0
Total amount of drug in hair. The total
0
60 120 180 240 300 360
0
60 120 180 240 300 360
amount of drug incorporated into hair was
~- 1000 ,
lOOO 1
determined by analyzing successive 1-cm
segments
of a "standard hair sample". A
800
C
~ 800
D
standard
hair
sample consisted of a bundle
~9
.2 m~ko ~
,~
|
1,2 m . ~
400
m 400 1
of hair fibers, about the thickness of a pencil
but varying in length, cut into successive
1-cm segments starting at the root or prox~0
0
60 120 180 240 300 360
0
60 120 180 240 300 360
imal end of the hair. From each segment, a
Time post dose (min)
Time post dose (min)
10-mg aliquot was weighed out, washed,
dried, and then analyzed. The total amount
Figure4. Plasma concentrations of cocaine-ds and BZE-dsfollowing intravenous (IV) administration
of cocaine-ds to human volunteers. Plasma concentrations of (A) cocaine-ds and (B) BZE-ds folof drug incorporated was calculated by sumlowing a 0.6-mg/kg dose to 10 subjects. Plasma concentrations of (C) cocaine-ds and (D) BZE-ds folming the amount of drug found in all the
lowing a t.2-mg/kgdoseto sevensubjects.
individual segments.
~
~
i
i
Downloaded from https://academic.oup.com/jat/article/20/1/1/750825 by guest on 20 April 2021
Quantitation of deuterated and nondeuterated drug and
metabolites in blood
Nondeuterated cocaine and BZE were measured according to
the method of Jacob et al. (25) using automated capillary gas
chromatography with nitrogen-phosphorus detection. Structural analogs, m-toluylecgonine and m-toluylecgonine methyl
ester, were used as internal standards, and the limit of detection
was 10 ng/mL using 1-mL samples.
Deuterated cocaine and deuterated BZE were determined by
GC-MS. The extraction procedure and derivatization of benzoylecgonine were identical to the method for unlabeled
cocaine and benzoylecgoninedescribedabove. GC-MSanalyses
were performed using a methyl silicone capillary column
(25 m • 0.2 mm) coupled to a desktop quadrupole mass spectrometer (Hewlett-Packard mass selective detector). Quantiration was achieved by selected-ion monitoring of the molecular ions produced by electron ionization of the analytes and
internal standards: the ion at m/z 303 was used for cocaine, the
ion at m/z 308 was used for cocaine-ds, the ion at m/z 317 was
used for the internal standard, m-toluylecgonine methyl ester,
the ion at m/z 345 was used for the butyl ester derivative of
benzoylecgonine, the ion at m/z 350 was used for the butyl
ester derivative of BZE-ds, and the ion at m/z 359 was used for
the butyl ester derivative of the internal standard, m-toluylecgonine. The limit of detection of this method was 10 ng/mL.
Journal of Analytical Toxicology, Vol. 20, January/February 1996
Intravenous
Nasal
MultipleDoses
Results
O
~.~s ~
[]
[]
O
I
I
I
I
I
I
I
m
~
I
I
I
I
o
I
o
I
,
"5"-~ lS
<.=_ ~ lO
v
o
$
o
[]
811
L"I
Cocaine-ds dose
Cocaine-ds dose
(mg/kg)
(mg/kg)
Cocaine-ds dose
(mg/kg)
Figure 5. Scattergrams showing the relationship between dose and the amount of cocaine-ds incorporated into hair expressedas the maximum amount of drug (lop row) and as the area under the curve
(bottom row). Open squares ([3) represent values for Caucasian subjects, and open circles (O) represent values for non-Caucasian subjects.
4
3
2
90308
90339
90340
eJ
A
5
3
2
90314
0 1 2 3 4 5 6 7 8 9 1 0 0 1 2 3 4 5 5 ? 8 9 1 0 0 1 2 3 4 5 5 7 8 9 1 0 0 1
Time post dose {months)
Time POStdose {months)
Time post dose {months)
110375"
2345678910
Time post dose {months)
Figure 6. Total amount of cocaine-d s found in hair at various times after the intravenous administration of 0.6 mg/kgcocaine-ds. Drug amounts are calculated by summingthe amounts found in all positive segments of the hair samples. Subject numbers with an asterisk indicate non-Caucasian subjects.
Plasma pharmacokinetics of cocaine-ds in
the test subjects
Figure 4 shows plasma concentrations of
cocaine-ds and BZE-ds after intravenous
doses of 0.6 and 1.2 mg/kg cocaine-ds.
These curves show that cocaine-d5 is the
primary analyte in plasma in the early time
periods, whereas BZE-d5is the primary analyte at later times. For the 0.6-mg/kg dose
group, peak cocaine-ds concentrations of
approximately 400-500 ng/mL occurred in
the first few minutes and then declined
rapidly over the next few hours. Conversely,
BZE-d5 concentrations increased over time
to a plateau value of approximately 200400 ng/mL.
For the 1.2-mg/kg dose group, peak
cocaine-ds concentrations of approximately
600-800 ng/mL occurred in the first few
minutes and then declined rapidly over the
next few hours. Conversely, BZE-ds concentrations increased over time to a plateau
value of approximately 400 ng/mL. The
mean half-life of cocaine-ds in these subjects was 66 + 14 min.
There was intersubject variability in the
drug concentration curves for the two analytes; however, the range of valueswas similar to those reported by other investigators
(26-31). Also, there appeared to be no raceor gender-related differences in the plasma
pharmacokinetics for the group of subjects.
Analyte profiles in hair
Cocaine-ds was the major analyte in hair
following all doses and all routes of administration. Even when multiple doses were
given, cocaine was still the primary anatyte.
Downloaded from https://academic.oup.com/jat/article/20/1/1/750825 by guest on 20 April 2021
are summed. Figure 3 illustrates how AUC calculations are
performed in present studies.
Table IVsummarizes how hair analysis data are expressed in
this report. Hair samples obtained from this subject were positive for cocaine-d5 at 0.26, 1.17, 2.20, and 3.13 months after
drug administration. The maximum amount of drug found
was 4.01 ng (the sample obtained at 0.26 months), and the
amount expressed as AUC was 6.84 ng months.
Pharmacokinetic calculations. Plasma half-life values for
deuterated and nondeuterated cocaine and metabolites were
calculated by traditional single-dose noncompartmental model
calculations using terminal rate constants. Peak plasma drug
concentrations were determined by inspection of the plasma
decay curves, and AUC values were determined by the trapezoidal rule.
Maximum amount of drug in the hair sample. The maximum amount of drug in hair is defined as the largest amount
of drug found in hair obtained at any sampling time. Typically, the maximum amount of drug was found in samples obtained 1-2 months after drug administration; however, this
time varied considerably between subjects. In our studies, the
maximum amount of drug in hair was used as a functional
equivalent to peak plasma concentration, a measure of drug
bioavailability typically used in pharmacokinetics.
Amount of drug expressed as area under the curve (AUC).
Area under the time-concentration curve (AUC) is a commonly used pharmacokinetic term used to define the total
amount of drug incorporated into the body over time and is a
useful measure of total drug bioavailability. AUC values for
hair samples were calculated using the trapezoidal rule, in
which areas of trapezoids comprising the amount-time curve
Journal of Analytical Toxicology, Vol. 20, January/February
1996
BZE was detected in only 10 of the 25 subjects and in only a few
of the hair samples obtained from these individuals. Usually,it
was found in those subjects who received the higher doses of
cocaine. Total amounts of cocaine-ds found in positive hair
samples were typically in the high picogram to low nanogram
range (0.1-5.05 ng per sample). BZE-dsamounts in hair were
less than 1 ng per sample. The mean ratio plus or minus SD of
cocaine-d5 to BZE-d~was 5.5 • 3.3. The metabolite EME was
detected infrequently, usually at the limit of detection; therefore, data for this metabolite are not shown.
doses greater than 35.2 mg (approximately 0.3 mg/kg). Thus,
the minimum detectable dose for our analytical methods
appeared to be between 22 and 35 mg, an amount somewhat
less than that found in a single "line" of street cocaine (e.g.,
50-100 rag).
c
~o
3
90364
91033
91031
r
o
: m. - 4. 1 .- - 4. l -.. 4 .k . a. s _ m . ~
: :
, , - T ". T , ,: : :
01
2 3 4 5 6 7 8 9 1 0 0 1 2 3 4 5 6 7 8 9 1 0 0 1 2 3 4 5 6 7 8 9 1 0
T i m e post dose (months)
T i m e post d o s e (months)
T i m e post dose (months)
Figure7. Total amount of cocaine-ds found in hair at various times after the intranasal administration
of 0.6 mg/kg cocaine-d5. Drug amounts are calculated by summing the amounts found in all positive
segments of the hair samples.
5
c
3
90376
0
5
4i i
3
2
9037
90378
90381
Effect of route of administration on
amount of cocaine incorporated into hair
None of the routes of administration produced a predictable relationship between
the dose and the amount of drug incorporated into hair. Theoretically, intravenous
administration should yield the best doseresponse relationship because absorption is
not a variable. However,even this route was
associated with considerable variability.
Effect of subjects' gender on amount of
cocaine incorporated into hair
All female subjects received the same
dose so we are unable to infer anything
about their dose-response relationship.
There was, however, little variability between the subjects--all four female subjects had very similar amounts of cocaine-d5
in their hair.
1
8
0..~:-5
:
:
:
:
:
:
:
:
4
c
3
,//
-
2
1
o
0
91003
91015
~
9 : : : : : : : " I
1 2 3 4 5 6 7 8 9 1 0
T i m e post d o s e (months)
I
0 1 2 3 4 5 6
78910
Time post dose (months)
0 1 2 3 4 5 6 7 8 9 1 0
Time post dose (months)
Figure8. Total amount of cocaine-ds found in hair at various times after the intravenous administration of 1.2 mg/kg cocaine-ds. Drug amounts are calculated by summing the amounts found in all positive segmentsof the hair samples. Subject numbers with an asterisk indicate non-Caucasian subjects.
Effect of subjects' race on amount of
cocaine incorporated into hair
A significant variable affecting the incorporation of cocaine into hair appeared to
be race. All four non-Caucasians had significantly more cocaine incorporated into
their hair than did their Caucasian counterparts (Figure 5). The non-Caucasians
were clearly outliers whether the amount of
cocaine was expressedas maximum amount
or AUC. Non-Caucasians had between two
Downloaded from https://academic.oup.com/jat/article/20/1/1/750825 by guest on 20 April 2021
Effect of dose on amount of cocaine incorporated into hair
The relationship between the dose of cocaine-ds administered and the amount of cocaine incorporated into hair is
illustrated in Figure 5. In this figure the amount of drug is
expressed as either maximum amount of drug or as the AUC.
As can be seen by inspection, there was a poor correlation beThreshold dose
tween the amount of drug incorporated into hair and the dose
No drug could be detected in the hair of subjects who
receivedby the subjects. This was true over the nearly fourfold
received the lower doses of cocaine (11.8-22 rag), even when
range of doses administered (from 0.6 to 4.2 mg/kg). Statistical
two small doses were given I week apart. However,cocaine was
analysis (linear regression by the least-squares method) of the
found in the hair of all the subjects who received intravenous
data confirmed this observation. Evenwhen
the non-Caucasianoutliers were eliminated,
5
there was at best a weak correlation between dose and the amount of drug in hair
91030
91021
c
(correlation coefficientsranged between 0.5
and 0.6), primarily because incremental
increases in dose resulted in only small
,'~'~..
=, _ , , - : : I I I I : : : : : : :
8
0
, ' . . . . . . .
: :
increases
in the amount of cocaine incor0 1 2 3 4 5 6 7 8 9 1 0
5
porated into hair.
Journal of Analytical Toxicology, Vol. 20, January/February
and 12 times (depending upon how the amount of drug was expressed) as much drug in their hair as did Caucasians.
t996
Detection window for cocaine
Data in Figures 6-9 show the total amount of cocaine-d 5
found in hair at various times after drug administration. In
general, there was a decrease in the amount of drug in hair
with time, and the larger the dose, the longer the drug could
be detected. However, the period of time during which cocaine could be detected in hair after a single dose varied considerably between subjects. Figure 6 shows that for the subjects
receiving 0.6 mg/kg cocaine-d5 intravenously, parent drug
could be detected in their hair for 2-3 months. When the drug
5
4
3
t.6
Time until drug is first detected in hair
IV 1.2 N
2
88173*
I
~;
Although determining the m i n i m u m
~ 90350
0
5
3,0 N
=
3
0
0
I
:1
; 3.0 N
....
:':'...
O 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10
Time post dose (months)
Time post dose (months)
Time post dose (months)
Figure 9. Total amount of cocaine-dsfound in hair at various times following the administrationof
0.6 mglkgcocaine-ds.Drug amountsare calculatedby summingthe amountsfound in all positivesegmentsof the hair samples.Subject numberswith an asteriskindicate non-Caucasiansubjects.Arrow
symbols mark time of dosing.
Distance from root
Distance from r o o t
Distancefrom root
(cm)
(om)
(cm)
(cm)
0 1 2345678910
0 1 2345678910
0 1 234 56 7891C 100 1 2345678910
0 ~
I~
1 1 I I I I I/
I I~1 I I I I I I I I
1
~ i
j i 90339 1 ~I l~J I I 90340_]
I l~ll,J
I I I I II
3
I I~IB4
1 1 1 1 II
Jllll
4
I I I',I [J I I I [ I I I I',I [] I i i I J
I I t [] P[ I t I I I ] [ 14 P'I [ [ [ t [ [ [ [J PI I ti i t I ' ',J I't L ' I L
] I~1"11xl I I I
S I I I I I'l I',ll [ I ,lllrll\llll
Distance from root
t
;IIIIIIHJt]
9
tOIIIIIII
.~
I IJ
I
0
1 ~
III
IN
g9 3 4
5
6
7 i i i
n- 6 i i i
9 I I I
i=~o I I I
o
II
8
i
i
I
I
i
i
I
I
i"1 i~1 i i
i i\1 iq i
I I Iq IJ
I I I 1.11
I
I
I
I
i
I
t
'\' q l J
IIIIJ
IIIIII
I
II
I I
I
i
I
I
I
I I~11
U
1'4 i'J I I I I
IJ PIIIII
I F I N I I I
I I I F.I N I
i i i i I\1 14
I I I I I 14 I
I I I I I I IJ
',11IV \ll
IIIIIIIIII.I
I I I I I ',",J t lI
I
I
I
J
I
I I I I I I I
'',-t
I II
II 14 IJ II
El I IJ P]I
I I I
II I [PI
hi I I I
I ] I I I r',l I\1 I I
I I I I I I I\l I,,11
I I I I I I I Iq I]
I1111
III1.t
I1'"IIII
III
III
III
IJI
J
I I I I I II
IJIIIII
Jr'lllll
r'l P,i i i i
i",l I\1 I I
]'4 141
i 14 i,l
I I [Jl
Segmental analysis
2
411141JIIIII
S IIIIl
PI I I I
I I I I PI IXl I I
7 I I I I I P.I IXJ I
~ 8 I I I I I I I\1 14
t I I I I I I 114 I
i.:t
I I I I I I I I IJ
time for drug to appear in hair was not an
original objective of this study and thus was
not studied systematically, we did observe
considerable variability in the time it took
for drug to first appear in hair. We
inadvertently discovered that cocaine could
be detected within 8 h of drug administration when a control (i.e., supposedly
predose) hair sample from one subject
(number 90377) was found to contain
cocaine-ds (Figure 8). After checking the
clinical laboratory records, it was discovered that the sample was indeed collected
on day 0 (the dosing day), but in fact, it was
collected 8 h after drug administration
rather than before drug administration. A
control hair sample obtained from this subject 3 weeks prior was negative. We subsequently obtained hair samples from four
other subjects (91021, 91030, 91031, and
91033) on days 1 and 3 following drug administration. Three of the four subjects
(91021, 91030, and 91031) had positive hair
samples the day after the drug was administered (Figure 7). Although these results
are limited, they do suggest that cocaine
may be detected in hair within hours after
drug administration.
I
I
I
I
]
II
I
I
I
I
I
I
I
I
I
[
IIl
I I
I I
I I
I I
I I
F~III
I'1 1~11 I
I IXl I\/I
I I N n
I I I 14 I
I I I I I,I
I
I
I
I
J
I
Figure 10. Segmentalanalysisdata from hair samplesobtained from subjects receuvung0.6 mglkg
cocaine-dsintravenously.Eachdata point (11)representsa 1-cm segmentpositivefor cocaine-ds. The
diagonal parallel lines show the theoreticalpath a bolus of incorporateddrug would take assuminga
hair growth rate of 1 cm/month. Subject numberswith an asteriskindicate non-Caucasiansubjects.
Figures 10-13 show the results of segmental analysis of all positive hair samples
obtained from the research subjects. In a
few subjects, for example subjects 90339
and 91031 (Figure 11), the incorporated
drug was confined to one or two segments
that moved down the hair shaft at the rate
consistent with a hair growth rate of
I cm/month. Similarly, the administration
Downloaded from https://academic.oup.com/jat/article/20/1/1/750825 by guest on 20 April 2021
was administered at higher doses or intranasally, cocaine-ds
could be detected in hair for up to 8 months. However, the detection window did not always appear to be related to the
amount of drug incorporated into hair after drug administration, and it would be difficult to predict with any accuracy
how long a single dose of cocaine can be detected. For example,
some subjects receiving a single dose had hair samples positive
for cocaine-ds for only 2 months, whereas others receiving a
single dose had positive hair samples for up to 8 months. In
Figure 9, subject 88173 had relatively large amounts of cocaine-ds in his hair (4 ng) 7 days after drug administration, yet
the drug disappeared rapidly and could not be detected after the
third month. In contrast, subject 91021 had much less drug in
hair (0.2 ng on day 1 and 0.14 ng on day 28),
but the drug could still be detected in hair
8 months after administration (Figure 7).
Journal of Analytical Toxicology, Vol. 20, January/February 1996
i: "'i
-
Distance from root
(cm)
~, O
1
g3
r
2
Distance from root
(cm)
O 12345678910
0t
~ 1
~
I I I-J
Distance from root
(cm)
234567891001234
5678910
Distance from root
(cm)
0 12345678910
LJ~J I I I I I I II
Discussion
4
s
| 8
E 9
~-10
9
1
2
I I I I I I I\1 I~l I
I I I In I l l ~ IJ
IIIllllliJI
I I I I ~1
I\1
I I
IllllllfJI
Ix.r',l I I t I t i i i
ll',,I
I I ~ 110 1 5 ~
I hal ],d I n I t J /
~ 4
~ 5
6
8
~ 9
NIO
I I Ill
Ill
II liB
I'1 I\1 I I
IIII II i Ill~
I t\t I~ I
iJ
It il.I J
Figure 12. Segmental analysis data from hair samples obtained from subjects receiving 1.2 mg/kg
cocaine-ds intravenously. Each data point ( l ) represents a 1-cm segment positive for cocaine-d s. The
diagonal parallel lines show the theoretical path a bolus of incorporated drug would take assuming a
hair growth rate of 1 cm/month. Subject numbers with an asterisk indicate non-Caucasian subjects.
The key feature of this study was the use
of isotopically labeled drug administered
under controlled laboratory conditions to
a relatively large number of human research subjects. In addition, the sensitive
and specific analytical method used for
quantitation and chemical ionization iontrap GC-MS easily distinguished between
the drug we administered, cocaine-d5 and
BZE-ds, from any tissue stores of drug or
self-administered drug use by the subjects
during the study. As a result, we were able
to conduct controlled studies over long
Downloaded from https://academic.oup.com/jat/article/20/1/1/750825 by guest on 20 April 2021
received a single intranasal dose of 0.6 mg/kg cocaine-ds. Segmental analysis of hair from subject 91031 showed that the
drug was confined to only one segment that, over the next 8
months, appeared to move along the hair shaft at a rate of approximately I crn/month. On the other hand, segmental analysis of hair collected from subject 91030 the day after drug administration revealed that cocaine-ds was distributed over 10
segments. Subsequent samples showed diminishing amounts
of incorporated drug that did not move down the hair shaft at
any predictable rate.
There was a similar lack of correlation between dosing regimen and segmental analysis results in subjects who received
multiple doses. For example, subjects 91001 and 91014 both
received the highest doses of any subjects during the study.
Also, in an attempt to partially simulate chronic drug use, the
drug was administered in successive doses
Distance from root
Distance from root
Distance from root
over a 1-month period. However, only
(cm)
(cm)
(cm)
modest amounts of cocaine-d5 were found
0 1 234567
89100
123 456
78910
0 12345
678910
in their hair, and the drug was confined to
only two or three segments. Segmental
analysis results similar to these were ob- s , _ o,379,_
- i 90364__
~ I
91021
~
_-(
I I t I
i I I Iserved in subjects who received only a
"I I I I
"I ii I
single bolus dose, for example numbers
[Xl I I
r',l-I I
67
" "~ I
\II\I I
\I Ix] I
90308 (Figure 10), 91030 (Figure 11), and
s
", ,~1
l'J Ix]
i /i Id
91003 and 91013 (Figure 12).
t= 1~
I I I,J
ii~l-"ld~
In addition, there was variability in the
f J I,
Illr]
rate at which the incorporated band of drug
traveled from the root to the tip. Segmental
32
",
_l I I I_
~
-9~,o,33analysis results from subjects 90376 (Figure
mm
-3_
12) and 91001 (Figure 13) indicate the
incorporated drug moved at a rate of
R 7
\i
\
\
\
9
1 cm/month, the most typically reported
8
\m,
,,
,
growth
rate for human hair. However, seg~- 9
x
,,
mental analysis data for subject 91033
10
(Figure 11) suggest this subject's hair grew
Figure 11. Segmental analysis data from hair samples obtained from subjects receiving 0.6 mg/kg
at
a slightly faster rate (approximately 1.3
cocaine-ds intranasaliy. Fach data point (l) represents a 1-cm segment positive for cocaine-d s. The
cm/month),
whereas data for subjects 91003
diagonal parallel lines show the theoretical path a bolus of incorporated drug would take assuming a
(Figure 12) and 91031 (Figure 11) suggest
hair growth rate of 1 cm/month.
these subjects' hair grew at a slightly slower
rate--0.7 and 0.8 cm/month, respectively.
of two doses, given 4 months apart, to subject 90350 (Figure
13) resulted in two discrete areas of incorporated drug that correlated to a surprising degree to the time of drug administration. However, for most subjects, segmental analysis revealed
considerable variability in the area over which incorporated
drug was distributed in the hair shaft and in the rate of axial
distribution of drug along the hair shaft.
Only six of the 23 subjects (26%) receiving a single dose had
cocaine-d5 and BZE-d5 confined to a single 1-cm segment.
Most of the subjects (11 of 23 or 48%) had drug distributed
over two segments, whereas three (13%) had drug distributed
over three segments and another three (13%) had drug distributed over four or more segments. The extreme differences
in segmental analysis results are illustrated by data from two
subjects shown in Figure 11. Subjects 91030 and 91031 both
Journal of Analytical Toxicology, Vol. 20, January/February 1996
higher concentrations of cocaine have been reported by others
is not surprising because most of the subjects in previously
reported studies were self-identified, chronic cocaine users
generally in treatment for their chemical dependency. Thus, it
is to be expected that the cocaine concentrations in their hair
would be considerably higher than the amounts found in our
subjects who, because of medical and ethical concerns, received
much smaller doses for shorter periods of time.
In the present study, increasing the dose of cocaine generally
resulted in a greater amount of drug in hair; however, there
was not a predictable relationship between dose and amount in
hair. Between the subjects there was considerable variability in
the amount of drug incorporated into hair, the time until drug
first appeared in hair, and the distribution of drug along the
hair shaft with time. Some variability is expected because of
experimental error or bias and biological factors. For example,
hair growth rate may be more variable than is typically
assumed. Studies have demonstrated an up to sixfolddifference
(0.3-1.8 cm/month) in growth rates for various populations
(37-39). Accordingly, Uematsu et al. (39) have shown that
intersubject variability in segmental analysis results can be
reduced when only hairs in the growing stage are analyzed and
great care is taken to align the strands precisely. Other investigators have found different concentrations of drugs like morphine (40) and methadone (41) in hair samples collected from
different anatomical sites.
Experimental error may explain some but not all of the observed intersubject variability in the study reported herein.
Hair samples were obtained only from the vertex region of the
scalp, and considerable care was taken to precisely align the
hair strands before analysis. Also, the range of hair growth
rates found in our subjects (0.7-1.3 cm/month) is similar to the
overall range of growth rates reported by others (37-39). Thus,
variability related to collecting and sectioning the hair samples
may explain some of the intersubject differences
in the time cocaine was first deDistance from root
Distance from root
Distance from root
(cm)
(cm)
(cm)
tected in hair, the number of segments in
012345678910012345678910
012345678910
which the drug was confined, and in the
0
rate at which the band of drug traveled from
irxll
iI i + ,,
1
1 1
\
,
3,ON I
the root to the tip. However, experimental
l l q I,
g 3
E== 4
error will not explain the unusual findings
g 5
in the outlier subjects who had considerably
-~ 6
higher
concentrations of cocaine-ds in mul~. 7
mmtiple hair segments obtained from many
~- 9
hair samples collected months apart and
I[,l
IrJ
10
analyzed
at different times.
0
"=mmmmmmm
1
The outlier subjects, all non-Caucasians,
--mmmmm
2
could have genotype-related differences in
,
-mmmmstl=l
~ 4
their drug distribution, metabolism, or
, 9 10,14
5
elimination. However, this does not appear
g 6
to be the cause of their unusual hair anal~ 6
ysis results. There were some intersubject
lid
|
~= 9
differences in cocaine plasma pharmacoki10
netics as would be expected, but these difFigure 13. Segmental analysis data from hair samples obtained from subjects receiving multiple
ferences
were small compared with differdoses of cocaine-d s. Each data point (m) represents a 1-cm segment positive for cocaine-d s. The diences in the amounts of drug incorporated
agonal parallel lines show the theoretical path a bolus of incorporated drug would take assuming a hair
into hair. Non-Caucasian subjects did not
growth rate of 1 cm/month. Subject numbers with an asterisk indicate non-Caucasian subjects. Arrow
symbols mark time of dosing.
differ significantly from Caucasians in any
--l_
10
Downloaded from https://academic.oup.com/jat/article/20/1/1/750825 by guest on 20 April 2021
periods of time without having the subjects confined or under
surveillance. To our knowledge, this is the first time such experimental procedures have been employed in hair analysis
research.
An important finding of our study is that hair analysis can be
an extremely sensitive indicator of cocaine use; however, because of the considerable intersubject variability, it is impossible to infer either the dose or time of dose from hair analysis
results alone. Surprisingly, perhaps, cocaine, not the metabolite BZE, is the major analyte found in hair. This unexpected
finding has been confirmed by others and is true regardless of
whether cocaine is administered to human subjects or experimental animals or whether the drug is administered acutely or
even chronically (15,32-35). Particularly in the latter case,
the BZE concentrations in the body would greatly exceed the
concentrations of parent drug.
The threshold dose for detecting cocaine in hair, using a
sensitive method like chemical ionization GC-MS, appears to
be approximately 25-35 mg cocaine administered intravenously. Once incorporated into hair a single dose of cocaine
can be detected for 2-6 months. This detection window varies
considerably between subjects for reasons that are not yet
apparent. Intersubject differences in hair hygiene or the use of
cosmetic hair treatments may be involved, but there is no
direct evidence for this.
The cocaine-ds and BZE-ds concentrations found in the present study (0.1-5 ng/sample for cocaine-d5 and less than
1 ng/sample for BZE) are in general agreement but at the low
end of the reported ranges of cocaine and BZE concentrations
reported by others, which include 6.4-19.2 ng/mL (33),
1.4-50.6 ng/mL (34), and 0.6-29.1 ng/L (36). The cocaineds/BZE-ds ratio found in our subjects was approximately 6.
This compares favorably with the 1-10 range of cocaine/BZE
ratios reported in other studies (33-36). The finding that
Journal of Analytical Toxicology, Vol. 20, January/February1996
higher doses. Finally, our studies suggest that the mechanism
generally proposed for drug incorporation into hair--that of
passive diffusion of drug into the growing hair follicle--is
probably too simplistic, and multiple mechanisms may be involved. Until these mechanisms are better understood and the
reasons for the intersubject variability clarified, it seems inappropriate to use hair analysis to infer either the dose, time, or
duration of cocaine use.
Acknowledgments
This research was supported, in part, by grants from the
National Institute on Justice (USJ-NIJ-90-IJ-CX-0012) and
from the National Institute on Drug Abuse (R01 DA082228).
References
I. A. Chatt, C.A. Secord, B. Tiefenbach, and R.E.Jervis. Scalp hair as
a monitor of community exposure to environmental pollutants. In
Hair, Trace Elements, and Human Illness. A.C. Brown and R.G.
Crounse, Eds., Praeger Publishers, New York, NY, 1980, pp
46-73.
2. A.M. Baumgartner, P.F.Jones, W.A. Baumgartner, and C.T. Black.
Radioimmunoassay of hair for determining opiate-abuse histories.
J. Nucl. Med. 20:748-52 (1979).
3. W. Arnold and K. Puschel. Experimental studies on hair as an indicator of past or present drug use. J. Forensic Sci. Soc. 21:83
(1981).
4. W.A. Baumgartner, C.T. Black, P.F. Jones, and W.H. Blahd.
Radioimmunoassay of cocaine in hair: concise communication.
J. Nucl. Med. 23(9): 790-92 (1982).
5. M. Michalodimitrakis. Detection of cocaine in rats from analysis
of hair. Meal., Sci. Law. 27(1): 13-15 (1987).
6. L. Parton, D. Warburton, V. Hill, and W. Baumgartner. Quantitation of fetal cocaine exposure by radioimmunoassay of hair.
Pediatr. Res. 21 (4): 372A (1986).
7. F.P.Smith and R.H. Liu. Detection of cocaine metabolite in perspiration stain, menstrual bloodstain, and hair. J. Forensic Sci.
31(4): 1269-73 (1986).
8. D. Valente, M. Cassini, M. Pigliapochi, and G. Vansetti. Hair as
the sample in assessing morphine and cocaine addiction. Clin.
Chem. 27(11): 1952-53 (1981).
9. T. Niwaguchi, S. Suzuki, and T. Inoue. Determination of methamphetamine in hair after single and repeated administration to rat.
Arch. Toxicol. 52:157-64 (1983).
10. S. Suzuki, T. Inoue, H. Hori, and S. Inayama. Analysis of methamphetamine in hair, nail, sweat, and saliva by mass fragmentography. J. Anal. Toxicol. 13:176-78 (1989).
11. O. Suzuki, H. Hattori, and M. Asano. Detection of methamphetamine and amphetamine in a single human hair by gas
chromatography/chemical ionization mass spectrometry. J.
Forensic Sci. 29(2): 611-17 (1984).
12. Y. Nakahara, K. Takahashi, Y. Takeda, K. Konuma, S. Fukui, and
T. Tokui. Hair analysis for drug abuse, Part II. Hair analysis for
monitoring of methamphetamine abuse by isotope dilution gas
chromatography/mass spectrometry. Forensic Sci. Int. 46:243-54
(1990).
13. Y. Nakahara, M. Shimamine, and K. Takahashi. Hair analysis for
drugs of abuse. Ill. Movement and stability of methoxyphenamine
(as a model compound of methamphetamine) along hair shaft
with hair growth. J. Anal. Toxicol. 16:253-57 (1992).
11
Downloaded from https://academic.oup.com/jat/article/20/1/1/750825 by guest on 20 April 2021
pharmacokinetic parameter, but they differed by two to 12
times (dependingupon how it was measured) in the amount of
cocaine-ds incorporated into their hair. The lack of significant
differencesin the plasma pharmacokineticsbetween the dosing
groups also precludes errors in dosing as a cause of variability.
Although the outliers were non-Caucasians,they were not all
African Americans. Two of the four were of mixed Hispanic,
Asian, and East Indian decent. The one common feature this
group did share was their coarse, dark hair. The differences in
drug incorporation by different hair types have been raised by
a number of investigators but remain controversial.
Kidwell (15) found that coarse, black hair takes up drugs
such as cocaine and phencyclidine more slowly from solution
and releases them more slowlythan fine, brown hair. Similarly,
black hair has been found to contain higher concentrations of
haloperidol, chlorpromazine, certain quinolone antimicrobials
(39,42), and nicotine (21) than white (nonpigmented) hair.
This is thought to be related to the melanin content in hair,
and drug-melanin binding was one of the earliest mechanisms
proposed for drug incorporation into hair (43,44). However,
simple differences in melanin content are an unlikely explanation for the differences observed in our study. The very low
drug concentrations in the studies described were found in
subjects with completely nonpigmented hair, and there were
no such individuals in our study. Most of our subjects had
dark brown or black hair.
A more likely explanation for the intersubject variability
observed in our study is that drugs may be incorporated into
hair by way of multiple mechanisms. For example,highly polar
drugs such as chlorpromazine, amphetamines, and quinolones
may depend on binding to melanin, and thus, their uptake
kinetics may vary with a subject's hair color. On the other
hand, unionized and lipid-soluble drugs like cocaine may enter
hair by way of contact with sweat and sebum as well as through
passive transfer from the blood. Thus, intersubject variability
in cocaine uptake into hair could be related to differences in
sweat and sebum secretion. Secretion of cocaine by these
glands, which are in intimate contact with the hair follicle,
could also explain how cocaine can be detected in hair within
a few hours after drug administration and how a single dose of
cocaine can be distributed over multiple hair segments. We
have direct evidence that cocaine and BZE are found in sweat
or sebum or both in higher concentrations and for a longer
time than they are measurable in plasma. Further, the cocaine/BZE ratio in sweat is similar to that found in hair. Details
of these findings will be presented in future communications.
In summary, hair analysis can be a sensitive technique for
detecting cocaine use. However, the significant variability
observed in the amount and time course of cocaine incorporation suggests that hair analysis does not necessarily
provide an accurate calendar of drug use; that is, hair analysis
results cannot be used with any certainty to determine either
the amount, frequency, or time of last cocaine use. Our studies
were limited by the amount of cocaine that could be administered. Compulsive cocaine users consume considerably more
cocaine than was used in our studies and repeat doses over
longer periods of time. Nevertheless, there is nothing in our
data to suggest that the intersubject variability is any less with
Journal of Analytical Toxicology,Vol. 20, January/February1996
12
30. C. Van Dyke, RG. Barash, R Jatlow, and R. Byck. Cocaine: plasma
concentrations after intranasal application in man. Science 197:
859-61 (1976).
31. P. Wilkinson, C. Van Dyke, and R Jatlow. Intranasal and oral
cocaine kinetics. Clin. Pharmacol. Ther. 27:386-94 (1980).
32. Y. Nakahara, T. Ochiai, and R. Kikura. Hair analysis for drugs of
abuse. V. The facility in incorporation of cocaine into hair over its
major metabolites, benzoylecgonine and ecgonine methyl ester.
Arch. Toxicol. 66:446-49 (1992).
33. E.J.Cone, D. Yousefnejad, W.D. Darwin, and T. Maguire. Testing
human hair for drugs of abuse. II. Identification of unique cocaine
metabolites in hair of drug abusers and evaluation of decontamination procedures. J. Anal. Toxicol. 15:250-55 (1991 ).
34. M.R. M~ller, P. Fey, and S. Rimbach. Identification and quantitation of cocaine and its metabolites, benzoylecgonine and ecgonine methyl ester, in hair of Bolivian coca chewers by gas chromatography/mass spectrometry. J. Anal. Toxicol. 16:291-96
(1992).
35. G.L. Henderson, M.R. Harkey, C. Zhou, and R.T. Jones. Cocaine
and metabolite concentrations in the hair of South American
coca chewers. J. Anal. Toxicol. 16:199-201 (1992).
36. K. Graham, G. Koren, J. Klein, J. Schneiderman, and M.G. Greenwald. Determination of gestational cocaine exposure by hair
analysis. J. Am. Med. Assoc. 262(23): 3328-30 (1989).
37. M. Saitoh, M. Usuka, M. Sakamoto, and T. Kobori. Rate of hair
growth. In Advances in the Biology of Skin. Hair Growth. W.
Montagna and RL. Dobson, Eds., Pergamon Press, Oxford, England, 1969, pp 183-201.
38. W. Montagna and T.F. Parakkal. The Structure and Function of
Skin. Academic Press, New York, NY, 1974, pp 83-105.
39. T. Uematsu, M. Nakano, H. Akiyama, and M. Nakashima. The
measurement of a new antimicrobial quinolone in hair as an
index of drug exposure. Br. J. Pharmacol. 35(2): 199-203 (1993).
40. R Kintz and R Mangin. Opiate concentrations in human head,
axillary, and pubic hair. ]. Forensic Sci. 38(3): 657-62 (1993).
41. S. Balabanova and H.U. Wolf. Methadone concentrations in
human hair of the head, axillary and pubic hair. Z. Rechtsmed.
102:293-96 (1989).
42. T. Uematsu and R. Sato. Human scalp hair as evidence of individual dosage history of haloperidol: longer-term follow-up study.
Ther. Drug Monit. 12:582-83 (1990).
43. W.H. Harrison, R.M. Gray, and L.M. Solomon. Incorporation of
L-DOPA, t-alpha-methyldopa and DL-isoproterenol into guinea
pig hair. Acta Dermatovener. 54:249-53 (1974).
44. W.H. Harrison, R.M. Gray, and L.M. Solomon. Incorporation of
D-amphetamine into pigmented guinea pig hair. Br. J. Dermatol.
91:415-18 (1974).
Manuscript received November 28, 1994;
revision received April 25, 1995.
Downloaded from https://academic.oup.com/jat/article/20/1/1/750825 by guest on 20 April 2021
14. A.M. Baumgartner, P.F.Jones, and C.T. Black. Detection of phencyclidine in hair. J. Forensic Sci. 26:576-81 (1981).
15. D.A. Kidwell. Analysis of phencyclidine and cocaine in human
hair by tandem mass spectrometry. J. Forensic Sci. 38:272-84
(1993).
16. J.J.Sramek, W.A. Baumgartner, J.A. Tallos, T.N. Ahrens, J.F. Heiser,
and W.H. Blahd. Hair analysis for detection of phencyclidine in
newly admitted psychiatric patients. Am. J. Psychiatry 142(8)"
950-53 (1985).
17. N.J. Haley and D. Hoffman. Analysis for nicotine and cotinine in
hair to determine cigarette smoker status. Clin. Chem. 31(10):
1598-1600 (1985).
18. R Kintz, B. Ludes, and R Mangin. Evaluation of nicotine and
cotinine in human hair. J. Forensic Sci. 37(1): 72-76 (1992).
19. P. Kintz, I. Kieffer, J. Messer, and P. Mangin. Nicotine analysis in
neonates' hair for measuring gestational exposure to tobacco. J.
Forensic Sci. 38(1): 119-23 (1993).
20. G. Koren, J. Klein, R. Forman, K. Graham, and M.-K. Phan. Biological markers of intrauterine exposure to cocaine and cigarette
smoking. Dev. Pharmacol. Ther. 18:228-36 (1992).
21. A. Mizuno, T. Uematsu, A. Oshima, M. Nakamura, and M.
Nakashima. Analysis of nicotine content of hair for assessing individual cigarette smoking behavior. Ther. Drug Monitor. 15:
99-104 (1993).
22. E.J.Cone. Testing human hair for drugs of abuse. I. Individual dose
and time profiles of morphine and codeine in plasma, saliva,
urine, and beard compared to drug-induced effects on pupils
and behavior. J. Anal. Toxicol. 14:1-7 (1990).
23. M.R. Bell and S. Archer. L(+)-2-tropanone. J. Am. Chem. Soc.
82:4642 (1960).
24. M.R. Harkey, G.L. Henderson, and C. Zhou. Simultaneous quantitation of cocaine and its major metabolites in human hair by gas
chromatography/chemical ionization mass spectrometry. J. Anal.
Toxicol. 15:260-65 (1991).
25. P.I. Jacob, B.A. Elias-Baker, R.T. Jones, and N.L. Benowitz. Determination of benzoylecgonine and cocaine in biological fluids
by gas chromatography. J. Chromatogr. 417:277-86 (1987).
26. G.R. Barnet, R. Hawks, and R. Resnick. Cocaine pharmacokinetics in humans. J. Ethnopharm. 3:353-66 (1981 ).
27. B. Holmstedt, J. Lindgren, L. Rivier, and T. Plowman. Cocaine in
blood of coca chewers. J. Ethnopharm. 1 : 69-78 (1979).
28. P. Jatlow. Cocaine by GLC and a nitrogen detector. In Methodology of Analytical Toxicology. I. Sunshine, Ed., CRC Press,Cleveland, OH, 1975, pp 133-37.
29. D. Paly, P. Jatlow, C. Van Dyke, F. Balieses, and R. Byck. Plasma
levels of cocaine in native Peruvian coca chewers. In InterAmerican Seminar on Medical and Sociological Aspects of Coca and
Cocaine, Lima, Peru, 1980. F.R. Jeri, Ed., Pacific Press, Lima,
Peru, 1980, pp 86-89.