Forensic Science International 145 (2004) 123–130

Determination of bromazepam, clonazepam and metabolites after

a single intake in urine and hair by LC–MS/MS
Application to forensic cases of drug facilitated crimes
Marjorie Chèzea,*, Marion Villainb, Gilbert Pépina
a
Laboratory of Toxicology, TOXLAB, 7 rue Jacques Cartier, 75018 Paris, France
b
Forensic Institute, 11 rue Humann, 67000 Strasbourg, France

Available online 15 June 2004

Abstract

The number of reports on drug facilitated crimes is increasing these last years. Apart from ethanol and cannabis,
benzodiazepines (BZD) and analogs are the most common drugs reported to be used probably due to their amnesic and
sedative properties. We have developed a rapid and sensitive method using LC–MS/MS triple stage quadrupole (TSQ) for the
determination of single exposure to bromazepam (Lexomil1, 6 mg) and clonazepam (Rivotril1, 2 mg) in urine and hair of
healthy volunteers. Chromatography was carried out on a Uptisphere ODB 5 mm, 2.1 mm
150 mm column (Interchim) with a
gradient of acetonitrile and formate 2 mM buffer, pH 3. Urine was extracted with Toxitube A1 (Varian) and allowed the
detection of bromazepam, 3-hydroxy-bromazepam, clonazepam and 7-Aminoclonazepam for more than 6 days. Head hair,
collected 1 month after the exposure, was treated by incubation with Soerensen buffer pH 7.6, followed by liquid–liquid
extraction with dichloromethane for common BZD. A specific pre-treatment for amino-BZD, with an incubation of 15 min at
95 8C in 0.1 N NaOH before liquid–liquid extraction with dichloromethane, gave better recoveries and repeatability. After single
exposure, bromazepam was present in powdered hair at 28 pg/mg and 7-Aminoclonazepam at 22 pg/mg in the first 1-cm
segment, while no clonazepam was detectable. This method was applied in two forensic cases. It allowed us to determine
bromazepam in urine 3 days after the alleged offense and in cut head hair at a concentration of 6.7 pg/mg only in the 2-cm
proximal segment. The other case showed the presence of clonazepam and 7-Aminoclonazepam in urine a few hours after the
offense and the presence of 7-Aminoclonazepam at about 3.2 pg/mg in axillary hair 4 months later.
# 2004 Elsevier Ireland Ltd. All rights reserved.

Keywords: Bromazepam; Clonazepam; LC–MS/MS; Hair

1. Introduction facilitated crimes [3,4]. Sedative and anterograde amnesia

The number of drug facilitated crimes seems to have
increased during the last few years, may be due to a better
knowledge of side effects of drugs like anxiolytic or hyp-
notics by common people, and also by an improved sensi-
tivity of apparatus used for forensic toxicology [1,2].
Apart from ethanol, cannabis and ecstasy, benzodiaze-
pines and analogs are the most common drugs used in drug
*
Corresponding author. Tel.: þ33 158 59 2808;
fax: þ33 158 59 2801.
E-mail address: toxlab@wanadoo.fr (M. Chèze).
properties of benzodiazepines are widely used for this
purpose.
Bromazepam was synthesized in 1963 and is widely
prescribed in European countries as an anxiolytic; but it
also exhibits sedative and hypnotic properties. The French
pharmaceutical called Lexomil1 (Roche) contains broma-
zepam at a concentration of 6 mg per tablet. The literature
shows that following a single administration of 12 mg to 10
subjects, an average peak plasma concentration of broma-
zepam of 131 ng/ml (107–173 ng/ml) was achieved between
1 and 4 h, declining with an average half-life of 11.9 h (7.9–
19.3 h) [5]. 3-hydroxy-bromazepam is reported to be its
main metabolite in urine.

0379-0738/$ – see front matter # 2004 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.forsciint.2004.04.066
124 M. Chèze et al. / Forensic Science International 145 (2004) 123–130

Clonazepam (Rivotril1, Roche) is an anti-convulsivant 2.2. Instrumentation

prescribed as anti-epileptic agent. The principal adverse
effect of clonazepam is drowsiness, which occurs in about The detection was performed on a triple stage quadrupole
50% of the patients on starting the therapy [6]. A single oral (TSQ) Quantum (ThermoElectron) with an electrospray
administration of 2 mg of clonazepam leads to an average ionization (ESI) orthogonal source and allowed the simul-
plasma concentration of 17 ng/ml (7–24 ng/ml) between 1 taneous detection of benzodiazepines in positive polarity and
and 4 h after ingestion, with a half-life of 19–42 h (8 SRM mode. To each pseudo-molecular ion (M þ H)þ,
subjects). During chronic therapy, plasma concentrations three product ions were acquired (Table 1) at a scan time
reach from 29 to 75 ng/ml. Its major metabolite is the 7- of 0.1 s with a width of 3–1 a.m.u. and a source collision
Aminoclonazepam [5]. induced dissociation (CID) energy of 10 V. For the optimi-
In the case of a suspected drug facilitated crime, urine and zation of detection conditions, direct infusions of standard
blood analysis must be done prior to hair analysis, if solutions (10 ml/ml) were done with mobile phase at 100 ml/
possible, for the interpretation of quantitative and qualitative min. Best results were obtained with a spray voltage of
results [3,4]. As the single administration of benzodiazepine 4800 V, a capillary temperature at 350 8C and an offset at
tablets leads to low levels of the drug and metabolites in 35 V. Tube lens was near 120 V. Sheath gas pressure and
urine and hair, since steady state concentrations are usually auxiliary gas flow were optimized at 50 and 10 (arbitrary
reached only after some days of treatment, classical tech- units), respectively. Collision cell pressure was at 1.5 m Torr
niques like GC/MS and LC/DAD are often not sufficient to of argon and collision energies were optimized for a max-
document these cases. Moreover, when the victim lodges a imum detection of each product ion.
complaint several days after the offense, hair analysis may The chromatographic separation was carried out with an
still the only proof of exposure. In addition, segmental hair HPLC Surveyor (ThermoElectron) on an Uptisphere ODB
analysis also permits to show the victim’s drug abstinence C18 5 m–150 mm
2 mm column (Interchim, France) ther-
without the period of the offense. mostated at 30 8C. Mobile phase initial conditions were 85%
The detection of benzodiazepines in segmented hair has formate buffer 2 mM pH 3 and 5% ACN (v/v) delivered at a
been documented previously [7–9] but rarely after a single flow of 200 ml/min and held for 0.5 min, then increased to
dose [10–12]. 10–90 for 10 min and re-equilibrated for 5 min at initial
To give an analytical response for these emerging cases, conditions. The total run time was 15 min.
we have developed an LC–MS/MS method for the determi-
nation of low levels of benzodiazepines in urine and hair 2.3. Sample collection
after a single oral dosage (one tablet).
In order to evaluate the reliability of our method, we We applied this technique to the analysis of urine and hair
have participated to a drug screening test for benzodiaze- extracts of healthy volunteers for the determination of
pine, in urine and hair, after a single intake by healthy
volunteers.
Table 1
SRM transitions for the detection of BZD by LC–MS/MS
2. Experimental
BZD [M þ H]þ Daughter Collision
ions (m/z) energy (%)
2.1. Standards and reagents
Bromazepam 316.03 288 26
Bromazepam, clonazepam, clonazepam-d4, 7-Amino- 261 30
clonazepam and 7-Aminoclonazepam-d4 were obtained 209 32
from Cerilliant (Promochem, France). Hydroxy bromaze- 3-OH-bromazepam 332 314.9 20
pam was obtained by Roche Laboratories. Ammonium 303 24
287 24
formate (minimum 98%), formic acid (99%), methanol
Clonazepam 316.0 270 30
and sodium hydroxide 0.1 N for analysis were obtained 241 40
from Carlo Erba, France. Acetonitrile, HPLC grade RS 214 42
Plus was obtained from Carlo Erba, France. Dichloro- Clonazepam-d4 319.9 274 30
methane Chromanorm HPLC grade was obtained from 245 42
Prolabo, France. PTFE filters (0.2 mm
25 mm) were 218 42
obtained from Alltech, France. Soerensen buffer was done 7-Aminoclonazepam 286.0 250 26
with a mixture of 38.8 ml of potassium dihydrogenopho- 222 30
sphate (KH2PO4) at 9.07 g/l and 61.2 ml sodium hydro- 195 38
genophosphate (Na2HPO4) (Sigma, France) at 11.87 g/L 7-Aminoclonazepam-d4 290.3 254 24
226 28
adjusted at pH 7.6 by 1 N NaOH (for analysis, Carlo Erba,
199 36
France).
 M. Chèze et al. / Forensic Science International 145 (2004) 123–130 125

Table 2 Table 4
Limits of detection and quantitation (LOD and LOQ) of BZD in Limits of detection and quantitation (LOD and LOQ) of BZD in
urine by LC–MS/MS hair by LC–MS/MS

BZD LOD (ng/ml) LOQ (ng/ml)– BZD Hair LOD LOQ (pg/mg)–
accuracy (%) (n ¼ 5) (pg/mg) accuracy (%) (n ¼ 5)

Bromazepam 0.1 1 (14.6) Bromazepam 1–2 5 (2.3)

3-OH-bromazepam 0.5 5 (5.4) Clonazepam 0.5 2 (10.2)
Clonazepam 0.05 1 (9.8) 7-Aminoclonazepam 2 > 10
7-Aminoclonazepam 0.1 1 (15.6) specific 5 (5.0)
pre-treatment: 1

Table 3
Repeatability of BZD of 1 and 5 ng/ml, in urine, by LC–MS/MS Table 5
(n ¼ 5) Repeatability of BZD at 5 and 10 pg/ mg, in hair, by LC–MS/MS
(n ¼ 5)
BZD 1 ng/ml, 5 ng/ml,
intraday CV (%) intraday CV (%) BZD 5 pg/mg– 10 pg/mg–
intraday CV (%) intraday CV (%)
Bromazepam 14.6 7.0
3-OH-bromazepam 42.0 5.2 Bromazepam 6 7.3
Clonazepam 5.6 3.4 Clonazepam 6.8 5.9
7-Aminoclonazepam 24.2 5.3 7-Aminoclonazepam 12.8 5.4

2.4. Urine extraction

bromazepam, clonazepam and metabolites after a single oral
dose of Lexomil1 6 mg tablet and Rivotril1 2 mg tablet.
Urine was collected in plastic containers of 30 ml before
intake, then every 12 h during 6 days, and was frozen at
20 8C prior to analysis.
Strands of hair were collected from the vertex posterior 3
weeks to one month after ingestion and were stored at an
ambient temperature until analysis.
Urine was extracted with Toxitube A1 (Varian) and with
1 ng/ml of clonazepam-d4 as internal standard (IS). After
horizontal agitation for 10 min and centrifugation at
3500 rpm for 5 min at an ambient temperature, the extract
was evaporated to dryness at 45 8C under a gentle stream of
air. The residues were reconstituted by 50 ml of ACN/MeOH
(50/50) and transferred in glass vials. 10 ml were injected

Urinary excretion

12,0

10,0
bromazepam
8,0 OH-bromazepam
(ng/ml)

6,0

4,0

2,0

0,0
0 0,08 12 24 36 48 60 72 84 96 108 120 132 144
Hours

Fig. 1. Urine excretion curve of bromazepam and 3-hydroxybromazepam after single oral dose of Lexomil1, 6 mg.
126 M. Chèze et al. / Forensic Science International 145 (2004) 123–130

Urinary excretion

60,0

50,0
7-aminoclonazepam
40,0 Clonazepam
ng/ml

30,0

20,0

10,0

0,0
0 0,08 12 24 36 48 60 72 84 96 108 120 132 144
Hours

Fig. 2. Urine excretion curve of clonazepam and 7-Aminoclonazepam after single oral dose of Rivotril1, 2 mg.

RT: 0.00 - 15.01 SM: 7G

RT: 11.31 NL: 1.41E5
MA: 1682930 TIC F: + c sid=-10.00 SRM
100
ms2 319.90@-30.00 [
210.72-274.66] MS
Clonazepam-d4 (ISTD)
50 290703-bzd04

0.38 1.23 2.87 3.67 5.47 6.06 7.54 9.03 10.25 11.84 13.80
0 NL: 2.04E3
RT: 10.10
MA: 21765 TIC F: + c sid=-10.00 SRM
100
ms2 316.03@-26.00 [
Bromazepam 180.68-288.53] MS
50 13.92 290703-bzd04
13.12
1.13 2.46 3.04 4.37 5.86 6.49 8.51 8.77 12.38
0 NL: 4.98E2
RT: 10.10
MA: 5287 m/z= 180.60-181.60 F: + c
100
sid=-10.00 SRM ms2
F1 316.03@-26.00 [
50 12.38 180.68-288.53] MS
8.99 14.19 290703-bzd04
0.39 1.67 2.46 3.20 4.32 5.11 7.18 8.51 10.79
0 NL: 9.75E2
RT: 10.10
MA: 10503 m/z= 208.66-209.66 F: + c
100
sid=-10.00 SRM ms2
F2 316.03@-26.00 [
50 180.68-288.53] MS
1.13 8.77 14.93 290703-bzd04
2.62 3.36 4.85 5.49 6.49 8.46 11.16 13.97
0 NL: 8.26E2
13.92 m/z= 287.50-288.50 F: + c
100 RT: 10.10
F3 sid=-10.00 SRM ms2
MA: 5128 13.39
316.03@-26.00 [
50 12.86 180.68-288.53] MS
11.53 290703-bzd04
0.13 1.51 3.04 4.37 5.86 7.55 9.30
0
0 2 4 6 8 10 12 14
Time (min)

Fig. 3. Chromatogram of extracted hair cut at 3.5 pg/mg bromazepam after single intake.
 M. Chèze et al. / Forensic Science International 145 (2004) 123–130
into the chromatographic system with a partial loop injection
mode.
2.5. Hair extraction
Because of oral route administration of drugs tested, hair
was washed only two times 5 min with dichloromethane for
cleaning fat residues. Hair was powdered or finely cut with
scissors depending on available quantities.
2.5.1. Common benzodiazepines
After decontamination, 100 pg/mg of clonazepam-d4 as
internal standard was added to 20 mg of cut hair. Matrix
extraction was performed for 14 h at 56 8C with 1 ml of
Sorensen buffer at pH 7.6. Liquid–liquid extraction was
carried out at ambient temperature for 10 min with 2 ml
of dichloromethane. After centrifugation, the organic layer
was filtered with PTFE 0.2 mm then evaporated until dry.
The residues were reconstituted by 60 ml of ACN/MeOH
(50/50) and transferred in glass vials.
2.5.2. 7-Amino-benzodiazepines
To improve the recoveries and repeatability for 7-Ami-
nobenzodiazepines in hair, the matrix extraction was done
Fig. 4. Chromatogram of extracted powdered hair at 22 pg/mg 7-Aminoclonazepam after single intake.
127
with a 15 min incubation at 95 8C in 0.1 N NaOH 7-Ami-
noclonazepam-d4 (100 pg/mg) was added before incuba-
tion, then a liquid–liquid extraction was performed with 2 ml
dichloromethane. After centrifugation, the organic layer was
filtered with a PTFE 0.2 mm filter, and then evaporated to
dryness. The residues were reconstituted by 50 ml of ACN/
MeOH (50/50) and transferred in glass vials.
3. Results
3.1. Urine
A calibration curve was prepared with blank urine at
concentrations ranging 0.5–50 ng/ml for all benzodiaze-
pines. Curves were linear in that range with equal weighting
and r2 values of 0.9950 for bromazepam, 0.9939 for 3-
hydroxy-bromazepam, 0.9958 for clonazepam and 0.9943
for 7-Aminoclonazepam. Detection limits were determined
by the analysis of successive dilutions. Quantitation limits
were determined from the repeatability and accuracy
(n ¼ 5). Detection and quantitation limits (LOD and
LOQ) are listed in Table 2. Repeatability at 1 and 5 ng/
ml (n ¼ 5) are listed in Table 3.
128 M. Chèze et al. / Forensic Science International 145 (2004) 123–130
3.2. Hair
Calibration curves were done for clonazepam, 7-Amino-
clonazepam and bromazepam with blank hair supplemented
at concentrations ranging 1100 pg/mg. Curves were linear in
this range with equal weighting and r2 values of 0.9975 for
clonazepam, 0.9996 for 7-Aminoclonazepam and 0.9995 for
bromazepam. Detection limits are listed in Table 4. Repeat-
ability (n ¼ 5) at 5 and 10 pg/mg are given in Table 5. The
improvement obtained with specific extraction for 7-Ami-
nobenzodiazepines led to a detection limit of 1 pg/mg for 7-
Aminoclonazepam.
3.3. Healthy volunteers
The LC–MS/MS analysis of urine of two healthy volun-
teers for the determination of bromazepam, clonazepam and
metabolites after a single oral dose of Lexomil1 6 mg tablet
and Rivotril1 2 mg tablet clearly shows that these drugs
could be detected more than 144 h (6 days) after the intake.
The excretion curves in urine are showed Figs. 1 and 2. Hair
analysis showed concentrations of 28 pg/mg of bromazepam
in powdered hair and about 3.5 pg/mg (<LOQ) when only
cut for subject 1 (Fig. 3), and 22 pg/mg of 7-Aminoclona-
zepam (Fig. 4) in the 1-cm proximal segment of subject 2,
Fig. 5. Chromatogram of extracted urine at 11.8 ng/ml bromazepam and 41.8 ng/ml 3-hydroxybromazepam.
while clonazepam couldnot be detected. No drugs were
found in the following segments.
4. Discussion
Comparison of these results to those obtained with con-
ventional routine techniques for urine analysis such as
Axsym (no detection), GC-MS (detection during 3 days)
and HPLC-DAD (detection during 1.5 days) and hair ana-
lysis (no detection by any technique) demonstrate the power
of LC–MS/MS as well as hair analysis for the documentation
of drug facilitated crimes.
Although the result of 3.5 pg/mg of bromazepam in hair
cut of subject 1 is inferior to the LOQ, the accuracy
calculated at 2 pg/mg (n ¼ 5) is 39 % and may be sufficient
to give a semi-quantitative result.
5. Forensic cases
5.1. Case 1
A young woman spent her Friday evening in a bar with
some friends. During the party she drank too much; but she
 M. Chèze et al. / Forensic Science International 145 (2004) 123–130
Fig. 6. Chromatogram of extracted hair cut at 3.2 pg/mg 7-Aminoclonazepam in axiliary hair.
was also abnormally tired and fainted, had incoherent beha-
vior and excessive irritability during 3 days, after which she
went with her mother to the hospital. She said later that she
couldnot remember how she came back home from the bar.
No drugs were found by classical analyses. After this, the
laboratory was requested to analyze the victim’s hair (32 cm
long) and urine with more sensitive apparatus. Urine was
found to be positive by LC–MS/MS at levels of 11.8 ng/ml
for bromazepam and 41.8 ng/ml for 3-hydroxy-bromazepam
(Fig. 5). Hair was collected 3 weeks after the offense. Three
segments of 2 cm long from the root were analyzed. Prox-
imal segment was positive for bromazepam at a concentra-
tion of 6.7 pg/mg of cut hair. The other segments were
negative. These results indicated a single drug exposure.
5.2. Case 2
After an evening out in Pigalle (Paris), a 40-year old man
took a rendez-vous, for the next evening, with a man in a gay
night club. After he had drunk one beer with him, they went
to his apartment where he felt asleep. When he woke up, he
realized that he had been robbed (money, computer, watch,
mobile phone, etc. . .). He went to the police station just a
few hours after. Urine was sampled at that time, but no
requisition was done. Four months after, the laboratory was
129
requested for urine and hair analysis. Urine analysis showed
the presence of clonazepam at 17.8 ng/ml and 7-Amino-
clonazepam at 561.7 ng/ml, indicating a recent exposure to
that drug, near the time of the offense. Head hair was too
short (about 2–3 cm) and pubic hair was recently shaved.
Axillary hair was sampled since its growth is similar to pubic
hair [12] and its length could reach the suspected period of
exposure. Hair analysis showed the presence of 7-Amino-
clonazepam at a concentration of about 3.2 pg/mg (<LOQ)
of cut axillary hair (Fig. 6), clearly indicating a single
exposure which is consistent with previous published data
of 4.8 pg/mg 7-Aminoclonazepam after single oral dosage
of 2 mg [11]. The offender was arrested with tablets in his
wallet. The, white and round, tablets seized were analyzed
by LC/DAD and showed the presence of clonazepam at
2.1 mg per tab, corresponding to commercially available
tablets of Rivotril1.
6. Conclusion
Among analysis techniques used in forensic toxicology,
LC–MS/MS is a powerful one for the determination of low
levels of benzodiazepines in urine and hair after a single oral
dose. In the case of a suspected drug facilitated crime, head
130 M. Chèze et al. / Forensic Science International 145 (2004) 123–130
hair testing should be done, if possible, about one month
after exposure, especially for drugs with limited window of
detection in biological fluids and when the victim lodges a
complaint several days after the offense since hair analysis
still be the only evidence of exposure [13]. Segmental hair
analysis also permits to show the victim’s drug abstinence
without the period of the offense. In cases of pubic or
axillary hair analysis, since no segmentation is possible,
the amount found must be in accordance with those found
after a single exposure. These cases must be further docu-
mented by inquiry.
We have recently improved our method for the screening
of 13 benzodiazepines, with double injection in the chro-
matographic system, dedicated to document drug facilitated
crimes and to give a response to the increasing demand of
justice.
References
[1] M.J. Mc Gregor, M. Lipowska, S. Shah, J. Du Mont, C. De
Siato, An exploratory analysis of suspected drug-facilitated
sexual assault seen in a hospital emergency department,
Woman Hlth. 37 (2003) 71–80.
[2] A. Negrusz, R.E. Gaensslen, Analytical developments in
toxicological investigation of drug-facilitated sexual assault,
Anal. Bioanal. Chem. 376 (2003) 1192–1197.
[3] M.H. Ghysel, G. Pépin, P. Kintz, Drug submission,
Toxicorama. 10 (1998) 126–127.
[4] P. Kintz, Soumission chimique: quels produits, quels
pélèvements, quelles analyses? J. Med. Leg. Droit Med. 45
(2002) 158–160.
[5] R.C. Baselt, Disposition of Toxic Drugs and Chemicals in
Man, Chemical Toxicology Institute, Foster City, California,
2000. pp. 94, 194.
[6] K. Parfitt (Ed.), Martindale—The Complete Drug Reference,
Parmaceutical Press, 1999, p. 343.
[7] R. Kronstrand, I. Nystrom, M. Josefsson, S. Hodgins, Segmental
ion spray LC–MS/MS analysis of benzodiazepines in hair of
psychiatric patients, J. Anal. Toxicol. 26 (2002) 479–484.
[8] M. Yegles, F. Mersh, R. Wenning, Detection of benzodiaze-
pines and other psychotropic drugs in human hair by GC/MS,
Forensic Sci. Int. 84 (1997) 211–218.
[9] A. El Mahjoub, C. Staub, Determination of benzodiazepines
in human hair by on-line high-performance liquid chromato-
graphy using a restricted access extraction columns, Forensic
Sci. Int. 123 (2001) 17–25.
[10] A. Negrusz, C.M. Moore, K.B. Hinkel, T.L. Stockham, M.
Verma, M.J. Strong, P.G. Janicak, Deposition of 7-amino-
flunitrazepam and flunitrazepam in hair after a single dose of
Rohypnol, J. Forensic Sci. 46 (2001) 1143–1151.
[11] A. Negrusz, C.M. Moore, J.L. Kern, P.G. Janicak, M.J.
Strong, N.A. Levy, Quantitation of clonazepam and its major
metabolite 7-Aminoclonazepam in hair, J. Anal. Toxicol. 24
(2000) 614–620.
[12] P. Mangin, Drug Testing in Hair, P. Kintz, CRC Press, 1996,
p. 280.
[13] G. Pépin, M. Chèze, G. Duffort, F. Vayssette, Interest of hair
and tandem mass spectrometry for chemical submission:
about several cases, Ann. Toxicol. Anal. 14 (2002) 395–406.