Anda di halaman 1dari 5

Forensic Science International 162 (2006) 108112

Solid-phase extraction and analysis of 20 antidepressant

drugs in human plasma by LC/MS with SSI method
Tatsuo Shinozuka a,*, Masaru Terada b, Einosuke Tanaka c
Department of Central Research Laboratory, School of Medicine, Keio University,
35 Shinanomachi, Shinjuku-ku, Tokyo 160-8582, Japan
Department of Legal Medicine, School of Medicine, Toho University,
5-21-16 Oomorinishi, Ota-ku, Tokyo 143-8540, Japan
Department of Legal Medicine, Institute of Community Medicine,
University of Tsukuba, Ibaraki-ken 305-8575, Japan
Received 26 August 2005; received in revised form 1 March 2006; accepted 1 March 2006
Available online 20 July 2006

A simultaneous determination of 20 antidepressant drugs (imipramine, amitriptyline, desipramine, trimipramine, nortriptyline, clomipramine,
amoxapine, lofepramine, dosulepin, maprotiline, mianserin, setiptiline, trazodone, fluvoxamine, paroxetine, milnacipran, sulpiride, tandspirone,
methylphenidate and melitracen) in human plasma was developed using LC/MS with sonic spray ionization (SSI) method. These drugs showed
good separation and sensitivity by LCMS using an Inertsil C-8 column with methanol:10 mM ammonium acetate (pH 5.0):acetonitrile (70:20:10)
as mobile phase at 0.10 mL/min at 35 8C. Solid-phase extraction of these drugs added to the human plasma was performed with an Oasis1 HLB
cartridge column. Recovery and limit of detection of these drugs were between 69 and 102% and between 0.03 and 0.63 mg/mL, respectively. The
present procedure offers an easier and more convenient screening method for antidepressants, and will be useful for forensic toxicology
# 2006 Elsevier Ireland Ltd. All rights reserved.

Keywords: Antidepressant drug; LCMS; Solid-phase extraction

1. Introduction tandspirone, methylphenidate and melitracen) has not been

reported. Furthermore, electrospray ionization (ESI), sonic spray
Antidepressant drugs are becoming increasingly widely used ionization (SSI) developed by Hirabayashi et al. [8] is a very soft
for the treatment of depression and these drugs are frequently ionization method.
encountered in emergency toxicology screening, drug-abuse The present paper describes a sensitive LCMS with SSI
testing and forensic medical examinations [1]. Various methods method for simultaneous determination of 20 antidepressant
for determination of antidepressant drugs have been reported, drugs, and a simple procedure for solid-phase column
including high-performance liquid chromatography (HPLC) extraction of the drugs from human plasma with an Oasis1
[2,3], gas chromatography (GC) [4,5], gas chromatography/mass HLB cartridge column.
spectrometry (GCMS) [6] and high-performance liquid
chromatographymass spectrometry (LCMS)[7]. However, a
2. Experimental
sensitive LCMS method for the simultaneous determination of
20 antidepressant drugs (imipramine, amitriptyline, desipra-
2.1. Chemicals
mine, trimipramine, nortriptyline, clomipramine, amoxapine,
lofepramine, dosulepin, maprotiline, mianserin, setiptiline,
Imipramine, desipramine, clomipramine, maprotiline and
trazodone, fluvoxamine, paroxetine, milnacipran, sulpiride,
methylphenidate were gifts from Novaltis Pharm Japan (Tokyo,
Japan). Fluvoxamine and sulpiride were gifts from Astellas
* Corresponding author. Tel.: +81 3 5363 3779; fax: +81 3 3353 1920. (Tokyo, Japan). Amoxapine and melitracen were from Takeda
E-mail address: (T. Shinozuka). (Tokyo, Japan). Amitriptyline, trimipramine, nortriptyline,
0379-0738/$ see front matter # 2006 Elsevier Ireland Ltd. All rights reserved.
T. Shinozuka et al. / Forensic Science International 162 (2006) 108112 109

lofepramine, dosulepin, mianserin, setiptiline, trazodone, par- 2.4. Extraction procedure of drugs from human plasma
oxetine, milnacipran and tandospirone were gifts from Banyu using an Oasis HLB1 cartridge column
(Tokyo, Japan), Shionogi (Osaka, Japan), Dainippon (Osaka,
Japan), Daiichi (Tokyo, Japan), Kaken (Tokyo, Japan), Sankyo Human plasma samples were obtained from Wako Co.
(Tokyo, Japan), Mochida (Tokyo, Japan), Organon Japan (Tokyo, (Tokyo, Japan) and kept at 20 8C until analysis. These
Japan), Glaxo SmithKline (Tokyo, Japan), Asahi KASEI Pharma samples were analyzed to determine possible endogenous
(Tokyo, Japan) and Sumitomo (Osaka, Japan), respectively. interferences, and were used as blanks. An Oasis HLB1
Methanol (HPLC grade), water (HPLC grade) and cartridge column with a capacity of 1 mL, was placed in a Vac-
ammonium acetate (analytical grade) were purchased from Elut1 system. The vacuum pressure was adjusted to 15
Kanto Kagaku (Tokyo, Japan). Oasis1 HLB extraction 20 mmHg and each column was activated by washing with
cartridges were purchased from Waters (Etten-Leur, The 1 mL methanol followed by 1 mL water.
Netherlands). All other common chemicals were commercially To 1.0 mL of human plasma containing 1 mg of each of the
available and of reagent grade. 20 antidepressant drugs and 2 mg diazepam (internal
standard) was added 100 mL phosphoric acid and 500 mL
2.2. Standard solution water. The mixture was applied to the Oasis HLB1 cartridge
column and allowed to pass through. The column was then
A stock standard solution containing 20 antidepressant drugs washed with 1 mL 10% methanol2% ammonium hydroxide
(1 mg/mL of each compound in methanol) was stable for at solution. The antidepressant drugs were eluted with 1 mL
least 3 months when stored at 20 8C. Spiked human plasma methanol2% acetic acid (70:30) solution. The eluent was
samples were prepared at concentrations of 1 mg/mL of each dried under N2 gas at 40 8C. The residue was dissolved in
compound by diluting appropriate volumes of stock solution. 100 mL mobile phase and an aliquot (5 mL) was injected into
the LCMS.
2.3. High-performance liquid chromatographymass
spectrometry 2.5. Recovery and accuracy

LCSSI (sonic spray ionization mass spectrometry) was The recovery and the accuracy were calculated by
performed on a HITACHI M-8000 LC/3DQMS system comparing the peak areas as the signal intensities of each
(HITACHI, Tokyo, Japan). The HPLC column was an Inertsil mass fragment for the 20 antidepressant drugs (1 mg/mL) in
C-8 (2.0 mm  150 mm, 5 mm) and the mobile phase was spiked samples after solid-phase extraction from human plasma
methanol10 mM ammonium acetate (pH 5.0)acetonitrile with the peak area as the signal intensity of the mass fragment
(70:20:10, v/v/v). A flow rate of 0.10 mL/min at 35 8C was used. of a series of internal standard samples. Concentrations of each
The qualitative and quantitative analyses were carried out in of the 20 antidepressant drugs were measured five times.
positive ion and full scan mode over the mass range m/z 60500.
The plate, aperture 1 and aperture 2 temperatures were 200, 150
and 120 8C, respectively. The drift voltage was 40 V and the Table 1
Retention time of 20 antidepressant drugs by LCMS
mass spectra of the drugs were recorded by total ion
monitoring. Retention times and characteristic mass fragments Antidepressant Retention
were recorded and the chosen diagnostic mass fragments were drug time (min)
monitored in the mass chromatography mode. 1 Imipramine 6.6
Signal intensities of each of the mass fragments: m/z 281 2 Amitriptyline 7.1
[M + H]+ for imipramine, m/z 278 [M + H]+ for amitriptyline, 3 Desipramine 7.4
4 Trimipramine 5.6
m/z 267 [M + H]+ for desipramine, m/z 295 [M + H]+ for 5 Nortriptyline 5.9
trimipramine, m/z 264[M + H]+ for nortriptyline, m/z 315 6 Clomipramine 7.8
[M + H]+ for clomipramine, m/z 314[M + H]+ for amoxapine, 7 Amoxapine 6.5
m/z 419 [M + H]+ for lofepramine, m/z 296 [M + H]+ for 8 Lofepramine 17.8
dosulepine, m/z 278 [M + H]+ for maprotiline, m/z 265 9 Dosulepine 6.5
10 Maprotiline 5.6
[M + H]+ for mianserin, m/z 262 [M + H]+ for setiptiline, m/ 11 Mianserin 8.7
z 372 [M + H]+ for trazodone, m/z 318 [M + H]+ for 12 Setiptiline 7.1
fluvoxamine, m/z 330 [M + H]+ for paroxetine, m/z 230 for 13 Trazodone 6.4
milnacipran, m/z 242 [M + H]+ for sulpiride, m/z 384 [M + H]+ 14 Fluvoxamine 5.7
for tandspirone, m/z 234 [M + H]+ for methylphenidate and m/z 15 Paroxetine 5.4
16 Milnacipran 4.8
292 [M + H]+ for melitracen were used for quantification. 17 Sulpiride 4.3
The calibration curves (the ratio between the peak areas as 18 Tandspirone 5.8
signal intensities of the drugs analyzed and that of the internal 19 Methylphenidate 5.1
standard (diazepam: m/z 285 [M + H]+) versus the concentra- 20 Melitracen 7.9
tion of each drug) exhibited linearity over the concentration Column, Inertsil C-8; mobile phase, methanol:10 mM ammonium acetate (pH
range 0.11.0 mg/mL human plasma. 5.0):acetonitrile (70:20:10, v/v/v); flow rate, 0.10 mL/min.
110 T. Shinozuka et al. / Forensic Science International 162 (2006) 108112

Fig. 1. Mass chromatograms of 20 antidepressant drugs by LCMS: (1) imipramine; (2) amitriptyline; (3) desipramine; (4) trimipramine; (5) nortriptyline; (6)
clomipramine; (7) amoxapine; (8) lofepramine; (9) dosulepine; (10) maprotiline; (11) mianserin; (12) setiptiline; (13) trazodone; (14) fluvoxamine; (15) paroxetine;
(16) milnacipran; (17) sulpiride; (18) tandspirone; (19) methylphenidate; (20) melitracen.

2.6. Limits of detection and quantification (LOD and LOQ) lowest concentration on the standard curve which can be
measured with acceptable accuracy (relative standard devia-
The LOD was determined from the amount injected which tion, R.S.D. <20%). The LOQ in human plasma was
produced a peak with a signal/noise ratio of at least 3:1. The determined by the LCMS with SSI method for 20
limit of quantification using an Inertsil C-8 column is the antidepressant drugs.
T. Shinozuka et al. / Forensic Science International 162 (2006) 108112 111

3. Results and discussion that the calibration curves were linear over the concentration
range studied 0.11.0 mg/mL. In this method, LOD and LOQ
3.1. Chromatography and mass spectrometry were 0.030.63 and 0.101.0 mg/mL, respectively (Table 2).
The sensitivity for these antidepressants except lofepramine in
When methanol:10 mM ammonium acetate:acetonitrile plasma indicated this method is applicable for the determina-
(70:20:10, v/v/v) was used as the mobile phase, the retention tion at therapeutic concentrations [9]. Tanaka et al. [2] have
times obtained using the Inertsil C-8 column are listed in described a sensitive HPLC method for the simultaneous
Table 1. All drugs were well separated under the conditions determination of 11 cyclic antidepressants in human biological
used, and every peak was detected within 17.8 min (Table 1). samples. The detection limit of these drugs by HPLC was as
The 20 antidepressant drugs all exhibited a pseudomolecular described in previous report [2]. The sensitivity of our method
ion peak except for milnacipran from the mass spectra obtained is identical to that by LCMS with ESI as reported by Gutteck
by the LCMS with SSI (positive mode) method. Fig. 1 shows the and Rentsch [7] in which 13 antidepressant and 5 neuroleptic
mass chromatograms of 20 antidepressant drugs when using the drugs in serum were separated on a reversed-phase C18
Inertsil C-8 column under the nominated LC conditions. No column. The internal standard used in the analysis was
endogenous interferences were observed during the analysis of diazepam according to our previous HPLC analysis [2].
the blank plasma, and a very simple chromatogram was obtained. However, it is the most common benzodiazepine in most parts
of the world and if appropriate may be replaced by another
3.2. Recovery, accuracy and precision data suitable substance.

The recoveries of the drugs ranged from 69 to 102%. 4. Conclusions

Satisfactory relative standard deviation (R.S.D.) values (signal
intensity ratio: I.S. versus amounts of each drugs) of 2.112.6% A simple, rapid and sensitive analytical method was
were obtained. Only intra-day reproducibility was tested, using developed for the simultaneous determination of 20 anti-
1 mL human plasma spiked with 1 mg drug (Table 2). The depressant drugs in plasma using LCMS technique after solid-
method showed very good precision and accuracy. phase extraction with an Oasis1 HLB cartridge column. This
simple and sensitive procedure is suitable from use in forensic
3.3. LOD and LOQ toxicology investigations.

Visual inspection of the plotted calibration curves involving Acknowledgment

triplicate analyses and correlation coefficients >0.99 confirmed
This work was supported in part by a Grant-in-Aid for
Table 2 Scientific Research (No. 16590545) from Japanese Ministry of
Recoveries, accuracies, limits of detection (LOD) and quantification (LOQ) of Education.
20 antidepressant drugs in human plasma by LCMS
Antidepressant Recoverya LODb LOQb References
(mg/mL) (mg/mL)
Added Mean R.S.D. [1] A.H. Dawson, Cyclic antidepressant drugs, in: R.C. Dart (Ed.), Medical
(mg/mL) (%) (%) Toxicology, third ed., Lippincott Williams & Wilkins, Philadelphia, 2004,
pp. 834843.
1 Imipramine 1.0 83.2 8.3 0.05 0.10
[2] E. Tanaka, M. Terada, T. Nakamura, S. Misawa, C. Wakasugi, Forensic
2 Amitriptyline 1.0 80.5 2.1 0.06 0.10
analysis of 11 cyclic antidepressants in human biological samples using a
3 Desipramine 1.0 69.2 6.3 0.08 0.15
new reversed-phase chromatographic column 2 mm porous micropherical
4 Trimipramine 1.0 70.4 6.7 0.05 0.10
silica gel, J. Chromatogr. B 692 (1997) 405412.
5 Nortriptyline 1.0 69.4 5.79 0.08 0.15
[3] C. Frahnert, M.L. Rao, K. Grasmader, Analysis of 18 antidepressants, 4
6 Clomipramine 1.0 72.4 10.0 0.10 0.15
atypical antipsychotics and active metabolites in serum by liquid chroma-
7 Amoxapine 1.0 74.8 8.4 0.16 0.20
tography: a simple tool for therapeutic drug monitering, J. Chromatogr. B
8 Lofepramine 1.0 69.5 6.7 0.63 1.0
794 (2003) 3547.
9 Dosulepine 1.0 77.4 7.9 0.08 0.10
[4] J.J. Nevado, M.J. Llerena, A.M. Salcedo, E.A. Nuevo, Assay validation for
10 Maprotiline 1.0 89.2 9.1 0.15 0.20
three antidepressants in pharmaceutical formations: practical approach
11 Mianserin 1.0 102.4 8.4 0.05 0.10
using capillary gas chromatography, J. Pharm. Biomed. Anal. 38 (2005)
12 Setiptiline 1.0 102.1 12.6 0.10 0.15
13 Trazodone 1.0 76.3 8.6 0.05 0.10
[5] M. Terada, T. Shinozuka, E. Tanaka, K. Osman, Z. Tun, N. Ikawa, Y.
14 Fluvoxamine 1.0 96.0 9.7 0.13 0.20
Fujimoto, K. Honda, J. Yanagida, S. Misawa, Simultaneous determination
15 Paroxetine 1.0 87.6 11.7 0.05 0.10
of 11 tricyclic and tetracyclic antidepressant drugs in human serum by fused
16 Milnacipran 1.0 85.2 8.7 0.08 0.15
silica wide-bore capillary gas chromatography with nitrogenphosphorus
17 Sulpiride 1.0 90.5 7.9 0.08 0.15
detection, Jpn. J. Forensic Toxicol. 15 (1997) 217225.
18 Tandspirone 1.0 82.3 9.6 0.03 0.10
[6] H.H. Maurer, J. Bickeboeller-Friedrich, Screening procedure for determi-
19 Methylphenidate 1.0 83.7 7.2 0.12 0.20
nation for detection of antidepressants of the selective serotonin reuptake
20 Melitracen 1.0 95.9 5.2 0.08 0.15
inhibitor type and the metabolites in urine as part of a modified systematic
Intra-day precision and accuracy of the assay (n = 5). toxicological analysis procedure using gas chromatographymass spectro-
n = 3. metry, J. Anal. Toxicol. 24 (2000) 340347.
112 T. Shinozuka et al. / Forensic Science International 162 (2006) 108112

[7] U. Gutteck, K.M. Rentsch, Therapeutic drug monitering of 13 antidepressant protein solutions a high flow rates, Anal. Chem. 70 (1998) 1882
and five neuroleptic drugs in serum with liquid chromatographyelectrospray 1884.
ionization mass spectrometry, Clin. Chem. Lab. Med. 41 (2003) 15711579. [9] M. Schulz, A. Schmoldt, Therapeutic and toxic blood concentrations of
[8] Y. Hirabayashi, A. Hirabayashi, Y. Takada, M. Sakairi, H. Koizumi, A more than 800 drugs and other xenobiotics, Pharmazie 58 (2003) 447
sonic spray interface for mass analysis of highly charged ions from 474.