Journal of Chromatography B, 993-994 (2015) 26–35

Contents lists available at ScienceDirect

Journal of Chromatography B
journal homepage: www.elsevier.com/locate/chromb

Analysis of drugs in plasma samples from schizophrenic patients by

column-switching liquid chromatography-tandem mass spectrometry
with organic–inorganic hybrid cyanopropyl monolithic column
Diego Soares Domingues, Israel Donizeti de Souza, Maria Eugênia Costa Queiroz ∗
Departamento de Química, Faculdade de Filosofia Ciências e Letras de Ribeirão Preto, Universidade de São Paulo, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: This study reports on the development of a rapid, selective, and sensitive column-switching liquid
Received 6 March 2015 chromatography-tandem mass spectrometry (LC-MS/MS) method to analyze sixteen drugs (antide-
Received in revised form 24 April 2015 pressants, anticonvulsants, anxiolytics, and antipsychotics) in plasma samples from schizophrenic
Accepted 26 April 2015
patients. The developed organic-inorganic hybrid monolithic column with cyanopropyl groups was used
Available online 5 May 2015
for the first dimension of the column-switching arrangement. This arrangement enabled online pre-
concentration of the drugs (monolithic column) and their subsequent analytical separation on an XSelect
Keywords:
SCH C18 column. The drugs were detected on a triple quadrupole tandem mass spectrometer (multiple
Column-switching
Organic–inorganic hybrid monolith
reactions monitoring mode) with an electrospray ionization source in the positive ion mode. The devel-
Liquid chromatography-tandem mass oped method afforded adequate linearity for the sixteen target drugs; the coefficients of determination
spectrometry (R2 ) lay above 0.9932, the interassay precision had coefficients of variation lower than 6.5%, and the rela-
Plasma samples tive standard error values of the accuracy ranged from −14.0 to 11.8%. The lower limits of quantification
Schizophrenia in plasma samples ranged from 63 to 1250 pg mL−1 . The developed method successfully analyzed the
target drugs in plasma samples from schizophrenic patients for therapeutic drug monitoring (TDM).
© 2015 Elsevier B.V. All rights reserved.

1. Introduction to adjust doses, minimize adverse effects, and check adherence to

Schizophrenia is a severe, chronic, and debilitating neuropsy-
chiatric disorder that affects one percent of the general world
population. This disorder generally appears in late adolescence or
early adulthood, and involves cognitive, emotional, and behavioral
abnormalities [1–3]. The reintroduction of clozapine in the market
in the 1990s and the advent of the second-generation or atypical
antipsychotics (olanzapine, quetiapine, risperidone, and ziprasi-
done) have benefitted patients with schizophrenia [4].
Physicians frequently recommend that patients switch from one
antipsychotic to another before exploring full dose ranges [5,6].
Another common strategy in management of these difficult-to-
treat patients is the combination of psychoactive drugs [7]. Apart
from antipsychotics, most schizophrenic patients use other drug
classes such as antidepressants, anxiolytics, and anticonvulsants,
to lessen the symptoms associated with the disease [8]. Therefore,
therapeutic drug monitoring of schizophrenic patients is important
∗ Corresponding author. Tel.:/fax: +55 16 36034465/+55 16 36024838.
E-mail address: mariaeqn@ffclrp.usp.br (M.E.C. Queiroz).
therapy.
Column-switching LC-MS/MS has become a powerful technique
to determine drugs in biological samples. It enables direct injec-
tion of the samples into the analytical system or sample injection
into the system after a simple treatment. This system uses two
columns with different stationary phases; fractions from one col-
umn are selectively transferred to a secondary column for further
separation. Direct on-line injection methods reduce the sample
preparation steps, effectively pre-concentrate the target drugs, and
remove endogenous compounds from biological samples. Thereby,
these methods minimize the exposition to biological fluids. More-
over, the column-switching technique can improve the sensitivity,
the resolution of complex samples, and the sample throughput
by means of heart-cutting, front- or end-cutting, and straight- or
back-flushing modes [9–13].
The monolithic column has been developed as an alternative
to conventional packed columns. It displays a continuously porous
structure characterized by macropores and mesopores that results
in a good column permeability and high mass transfer efficiency
[14–17]. In recent years, organic–silica hybrid monolithic columns
have attracted great attention, because they are biocompatible
with biological samples, do not shrink or swell upon exposure

http://dx.doi.org/10.1016/j.jchromb.2015.04.040
1570-0232/© 2015 Elsevier B.V. All rights reserved.
 D.S. Domingues et al. / J. Chromatogr. B 993-994 (2015) 26–35
to organic solvents, and exhibit good mechanical and chemical
stability [18].
The literature brings a wide variety of reports on organic-
inorganic hybrid monoliths with different organic moieties and
distinct macro-mesoporous structures applied as stationary phases
in chromatographic separations [19–21], electrophoreses [22],
solid-phase extraction (SPE) [23,24], stir bar sorptive extraction
(SBSE) [25], solid-phase microextraction (SPME) [26–28], and in-
tube SPME-LC [29].
The present work describes the preparation of an organic-
inorganic hybrid cyanopropyl monolithic capillary column func-
tionalized with cyanopropyl groups for the first dimension of
the column-switching LC-MS/MS system. This method was suc-
cessfully applied to determine sixteen drugs (antidepressants,
anticonvulsants, anxiolytics, and antipsychotics) in plasma samples
from schizophrenic patients for therapeutic drug monitoring.
2. Materials and methods
2.1. Standards and reagents
Haloperidol, olanzapine, clonazepam, mirtazapine, paroxetine,
citalopram, sertraline, chlorpromazine, imipramine, clomipramine,
quetiapine, diazepam, fluoxetine, clozapine, carbamazepine, and
lamotrigine standards were purchased from Cerilliant (Round
Rock, TX, USA). Stable labeled (internal standards) haloperidol-
d4, clonazepam-d4, paroxetine-d6, citalopram-d6, sertraline-d3,
imipramine-d3, clomipramine-d3, quetiapine-d8, diazepam-d5,
fluoxetine-d6, clozapine-d4, and carbamazepine-d10 were also
acquired from Cerilliant (Round Rock, TX, USA). Tetraethyl
orthosilicate (TEOS) (98%), 3-cyanopropyltriethoxysilane (CN-
TEOS) (98%), (3-aminopropyl) triethoxysilane (APTES) (98%), and
N-dodecylamine (99%) were obtained from Sigma Aldrich (St.
Louis, USA). Acetonitrile, ethanol, and methanol (HPLC grade) as
well as ammonium acetate and formic acid were supplied by
JT Baker (Phillipsburg, USA). Cetyltrimethylammonium bromide
(CTAB, 95%), sodium hydroxide (NaOH), and hydrochloric acid (HCl)
were purchased from Sigma–Aldrich (St. Louis, USA). Aqueous solu-
tions were prepared by using ultrapure water obtained from a
Milli-Q (18 ) system (Millipore, São Paulo, Brazil).
2.2. Preparation of the organic-silica hybrid cyanopropyl
monolithic capillary
The fused-silica capillary (530 ␮m × 4.5 cm), purchased from
Ohio Valley (Ohio, USA), was activated with NaOH (1 mol L−1 ),
followed by HCl (1 mol L−1 ). After rinsing with ultrapure water,
the capillary was dried under a N2 stream at 160 ◦ C, for 8 h. The
hybrid monolithic capillary was synthesized by hydrolysis (at low
pH value) and polycondensation (higher pH value) of precursors
via a two-step catalytic sol-gel process, according to previously
described procedures [29], with some modifications. The precur-
sors, 110 ␮L of CN-TEOS and 110 ␮L of TEOS, were added to a
solution consisting of 180 ␮L of ethanol and 25 ␮L of acetic acid
(2 mol L−1 ) in an Eppendorf vial (1.5 mL). Hydrolysis was performed
at 60 ◦ C, for 5 h. After that, 10 mg of N-dodecylamine was added
to the solution, at room temperature. This mixture was manu-
ally injected into a pre-treated capillary of an appropriate length
with the aid of a syringe. Both ends of the capillary were sealed
with two pieces of rubbers. Then, the capillary was incubated at
40 ◦ C, for 15 h, for condensation and polymerization. Subsequently,
the capillary was extensively rinsed with ethanol, to remove N-
dodecylamine and soluble hydrolysis products, and dried at 60 ◦ C,
for 30 h.
27
2.3. Characterization of the organic–inorganic cyanopropyl
hybrid silica monolithic capillary
The morphological structural aspects of the monolithic cap-
illary were examined by Scanning Electron Microscopy. The
samples were coated with gold in a Bal-Tec SCD050 Sputter
coater instrument (Fürstentum Liechtenstein), for 180 s. Then, they
were analyzed in a Zeiss EVO 50 scanning electron microscope
(Cambridge UK). The chemical groups were identified by Fourier
Transform Infrared Spectroscopy (FTIR), which was conducted on
the Shimadzu- IRPrestige-21 equipment (using KBr pellets).
2.4. Plasma samples
To optimize and validate the developed LC-MS/MS method,
drug-free plasma samples from volunteers that had not been
exposed to any drug for at least 72 h (blank plasma) were used.
These blank plasma samples as well as the plasma samples from
schizophrenic patients were supplied by the Psychiatric Nursing
staff of Hospital das Clínicas de Ribeirão Preto, University of São
Paulo, Brazil. This study has been approved by Research Ethics Com-
mittee, process n. 591/2011-2011.1.1664.59 2 .from Faculdade de
Filosofia Ciências e Letras de Ribeirão Preto–University of São Paulo.
Bench top and autosampler stability tests showed that plasma
samples containing the target drugs were stable for 24 h at room
temperature. Patients’ plasma samples containing the target drugs
were stable over three freeze-thaw cycles or when stored at −20 ◦ C
for six months [30–33].
2.5. Standard Solutions
The working standard solutions were prepared by diluting
the stock standard solutions (200 ␮g mL−1 ) to concentrations
ranging from 0.075 to 40.5 ng mL−1 for haloperidol and olanza-
pine, from 0.625 to 155.0 ng mL−1 for clonazepam, from 0.125
to 155.0 ng mL−1 for mirtazapine, from 0.250 to 155.0 ng mL−1
for paroxetine, from 0.125 to 290.0 ng mL−1 for chlorpromazine
and imipramine, from 0.250 to 510.0 ng mL−1 for clomipramine,
from 0.125 to 510.0 ng mL−1 for quetiapine, from 1.250 to
405.0 ng mL−1 for citalopram, from 0.625 to 405.0 ng mL−1 for
sertraline, from 0.313 to 1050.0 ng mL−1 for diazepam and
fluoxetine, from 0.188 to 1550.0 ng mL−1 for clozapine, from
0.063 to 10500.0 ng mL−1 for carbamazepine, and from 1.250
to 10500.0 ng mL−1 for lamotrigine. These solutions were sta-
ble for two months, at −20 ◦ C [34,35]. The internal standard
solutions were prepared in methanol at the following concen-
trations: haloperidol-d4 (20.5 ng mL−1 ), clonazepam-d4 and
paroxetine-d6 (80.0 ng mL−1 ), imipramine-d3 (185.0 ng mL−1 ),
citalopram-d6 and sertraline-d3 (205.0 ng mL−1 ), clomipramine-
d3 and quetiapine-d8 (260.0 ng mL−1 ), diazepam-d5 and
fluoxetine-d6 (6550.0 ng mL−1 ), clozapine-d4 (850.0 ng mL−1 ),
and carbamazepine-d10 (5.5 ␮g mL−1 ).
2.6. Pre-treatment of the plasma samples
The proteins of the plasma samples (200 ␮L) were pre-
cipitated with acetonitrile (400 ␮L). The supernatant (500 ␮L)
was dried in a vacuum concentrator (Eppendorf, Brazil), and
the dried extract were reconstituted (100 ␮L) with ammonium
acetate (5 mmol L−1 )/ammonium hydroxide (5 mmol L−1 ) (1:1 v/v)
solution. Then, 10.0 ␮L of this solution was injected into a two-
dimensional system. Different pH values (3.0, 7.0, 10.0) of the
ammonium solution were evaluated to establish the sorption capa-
bility of the monolith capillary.
28 D.S. Domingues et al. / J. Chromatogr. B 993-994 (2015) 26–35

Fig. 1. Scheme of column-switching LC-MS/MS system. (a) Position 1, and (b) Position 2 in straight-flushing mode. QSM: quaternary pump, BSM: binary pump, MS: mass
spectrometry, 1st column: monolithic capillary, 2nd column: analytical column.

2.7. LC-MS/MS conditions
LC-MS/MS analyses were carried out on a Waters ACQUITY UPLC
H-Class system coupled to the Xevo® TQ-D tandem quadrupole
(Waters Corporation, Milford, MA, USA) mass spectrometer
equipped with a Z-spray source operating in the positive mode.
The drugs were separated in an XSelect CSH C18 XP (Waters)
column (2.1 × 100 mm, 2.5 ␮m), at 40 ◦ C. The source and oper-
ating parameters were optimized as: capillary voltage, 0.50 kV;
source temperature, 150 ◦ C; desolvation temperature, 600 ◦ C; des-
olvation gas flow, 500 L h−1 (N2 , 99.9% purity); and cone gas flow,
5 L h−1 (N2 , 99.9% purity). All the drugs were analyzed in the mul-
tiple reaction monitoring (MRM) mode; Argon (99.9999% purity)
was used as the collision gas. The dwell time was established for
each transition separately, and the interscan delay was set to the
automatic mode. Data were acquired using the MassLynx V4.1
software.
in Position 1, 10 ␮L of the sample was injected. The mobile phase,
which was composed by water (poor solvent) was percolated
through the monolithic capillary at a flow rate of 100 ␮L min−1 .
This condition aided drugs sorption and removal of endogenous
components from the plasma samples. After 5.00 min, the six-port
valve was switched to Position 2. In this arrangement, the initial
mobile phase consisted of (A) ammonium acetate 5 mmol L−1
(with 0.1% formic acid) and (B) acetonitrile (70:30 v/v) eluted the
analytes from the monolithic capillary to the analytical column
at a flow rate of 100 ␮L min−1 . At 7.00 min, the six-port valve was
switched back to Position 1. From 7.01 to 10.00 min, the drugs
were separated at a flow rate of 300 ␮L min−1 . After this period,
the composition of the mobile phase was restored to the initial
condition, to re-equilibrate the analytical column for the next
sample injection. Evaluation of distinct mobile phases at different
time conditions helped to establish appropriate drugs sorption
and analytes resolution within a minimum analysis time, Table 1.

2.7.1. Automatic procedures for column-switching LC-MS/MS 2.8. Analytical validation

The capillary monolithic column and the analytical column
were connected at the six-port valve, Fig. 1. The quaternary pump
(QSM) was connected to the monolithic column (first column), and
the binary pump (BSM) was connected to the analytical column
(second column). In the first step conducted with the six-port valve
In agreement with the therapeutic intervals of the drugs, the
analytical validation of this method was based on current interna-
tional guidelines issued by EMA (European Medicines Agency) and
FDA (Food and Drug Administration).

Table 1
Chromatography conditions for column switching LC-MS/MS analysis.

QSM Pump BSM Pump

A = Water A = Ammonium acetate 5 mmol L−1 (0.1% formic acid)

B = Acetonitrile B = Acetonitrile

t (min) Pump Flow rate %B Valve Comments

(␮L min−1 ) Position

0.00 QSM 100 0 1 Pre-concentration of the drugs and remotion of endogenous

components from monolith capillary (first column)
0.00 BSM 100 30 1 Analytical column conditionation
5.00 QSM 100 30 2 Elution the drugs from monolith capillary (first column) to the
analytical column (second column)
7.01 BSM 300 30 1 Chromatographic separation on the analytical column (second column)
10.00 BSM 300 80 1
10.01 QSM 100 100 1 Clean up of the capillary column (first column)
10.01 BSM 300 30 1 Columns re-equilibrium for the next sample injection
15.01 QSM 100 0 1
17.01 BSM 100 30 1
20.00 All – – 1 End of cycle

QSM: quaternary pump, and BSM: binary pump.

 D.S. Domingues et al. / J. Chromatogr. B 993-994 (2015) 26–35 29

Fig. 2. Scanning electron microscope images of the cross-section of cyanopropyl-hybrid silica monolithic capillary.

3. Results and discussion for the ammonium solution, the sample diluted at pH 10 pre-
sented adequate sorption capability, mainly in the case of drugs
3.1. Preparation and characterization of organic-inorganic hybrid with pKa values above 8. At pH 10, the target drugs were in their
cyanopropyl monolithic capillary partially ionized or non-ionized form, which favoured their interac-
tion (dipole–dipole and induced dipole–dipole) with the monolith
Pretreatment of the fused silica capillary is important to clean phase [29]. Pre-treatment of the plasma sample and clean-up of the
and increase the concentration of silanol groups on the inner sur- monolith phase after drugs sorption removed the endogenous com-
face, because these groups serve as binding sites for the monolith pounds from the plasma samples. Those procedures avoid blockage
during the sol–gel synthesis [36]. of the capillary and minimize the matrix effect on the LC analysis.
Fig. 2 illustrates the cross-sectional SEM images of the hybrid
silica monolithic capillary at 200x and 30,000x magnification. The 3.3. Analytical validation
developed hybrid monolith exhibited a continuously macroporous
structures interconnected with selective particles. This decreased In this study, the analytes were first detected by ESI-MS/MS in
not only the back pressure, but also mass-transfer resistance of the the full-scan mode in positive mode. The protonated molecules
analytes from the plasma samples [37]. [M + H]+ of target the analytes were submitted to experimental
The monolith structure (in situ polymerization) remained con- collision-induced dissociation (CID). In order to select a proper
fined inside the capillary, closely attached to the inner surface transition for the MS/MS detection of the analyte, at least two pre-
(Fig. 2). This covalent anchoring provided strong chemical and cursor/product ion pairs for each analyte were examined in this
mechanical stability, preventing the monolith from leaching from study (Table 2).
the capillary and dismissing the use of frits [18]. It was possible It was possible to verify the selectivity of the method by rep-
to reuse the developed capillary over one hundred times, without resentative LC-MS/MS (MRM mode) chromatograms of a blank
significant changes in its sorption capability. This reusability indi- plasma sample, and by LC-MS/MS chromatograms of a blank
cated that this phase is robust. The enrichment factors obtained plasma sample spiked with drugs at concentrations corresponding
with the cyanopropyl monolithic capillary were at least tenfold the to LLOQ (Fig. 4).
enrichment factors achieved with direct injections of the standard The LC-MS/MS method was linear within the adopted calibra-
solution. tion ranges. For all the studied analytes, linearity ranged from the
The organic moieties attached to the capillary via silica–carbon
(Si C) linkages. These linkages were more hydrolytically stable
than the monomeric siloxane bonds typically used in chromato-
graphic stationary phases attached onto silica matrix surfaces.
Fig. 3 depicts the FTIR spectra of the monolith capillary. The
bands located at 800, 1100, and 3500 cm−1 corresponded to
the Si O Si symmetric stretching, the Si O Si antisymmetric
stretching, and the O H stretching bands of the silica network,
respectively [38,39]. The band at 1650 cm−1 referred to scissor
bending vibration of molecular water adsorbed onto the monolith
structure [40]. The bands at 1341 and ∼2960 cm−1 were due to
the CH2 deformational and stretching vibrational motion of the
propyl group, respectively [39,41]. In fact, the band at 2259 cm−1 ,
related to -CN stretching, confirmed the successful incorporation
of the cyanopropyltriethoxysilane monomers into the monolith
structure [42].

3.2. Effect of pre-treatment of the plasma sample on the sorption

capability of the monolith capillary

After the simple pre-treatment of the plasma samples (pro-

tein precipitation), the dried extracts were reconstituted with
ammonium solution. Among the different pH values evaluated Fig. 3. FTIR spectra of cyanopropyl-hybrid silica monolithic capillary.
30 D.S. Domingues et al. / J. Chromatogr. B 993-994 (2015) 26–35

Table 2
Ion transitions, instrument settings, and retention times for each drug.

Analyte Precursor Product DP CE Qualifier rt

ion (m/z) ion (m/z) (V) (eV) ion (m/z) (min)

Haloperidol 376.1 123.0 40 45 165.0 8.56

Olanzapine 313.3 84.1 45 20 256.2 7.40
Clonazepam 316.0 270.1 50 25 214.1 9.23
Mirtazapine 266.2 72.1 40 20 195.0 7.51
Paroxetine 330.2 70.0 50 25 192.2 8.64
Citalopram 325.3 109.1 45 25 262.0 8.44
Sertraline 306.2 159.0 25 30 275.1 9.01
Chlorpromazine 319.2 86.1 40 20 58.1 9.01
Imipramine 281.2 86.0 30 15 57.9 8.77
Clomipramine 315.2 86.0 35 15 57.9 9.11
Quetiapine 384.3 253.2 45 20 221.0 8.37
Diazepam 285.1 193.1 50 35 154.2 9.89
Fluoxetine 310.1 43.9 25 10 148.2 8.95
Clozapine 327.2 192.2 40 40 270.2 8.29
Carbamazepine 237.1 194.1 35 20 – 8.94
Lamotrigine 256.1 58.0 55 40 144.9 7.47
Internal Standards
Haloperidol-d4 380.2 127.1 45 45 169.1 8.54
Clonazepam-d4 320.1 274.1 50 30 218.1 9.22
Paroxetine-d6 336.2 76.1 30 35 198.2 8.63
Citalopram-d6 331.2 108.9 35 25 262.1 8.44
Sertraline-d3 309.1 159.0 20 25 275.0 9.00
Imipramine-d3 284.3 89.0 35 15 208.2 8.77
Clomipramine-d3 318.2 89.0 40 20 60.9 9.11
Quetiapine-d8 392.3 258.1 50 30 286.1 8.36
Diazepam-d5 290.1 198.1 55 30 153.9 9.87
Fluoxetine-d6 316.2 44.1 25 15 154.1 8.94
Clozapine-d4 331.2 192.2 50 50 272.4 8.27
Carbamazepine-d10 247.2 204.1 35 20 – 8.90

DP: Declustering potential, CE: collision energy, and rt: retention time.

Fig. 4. LC-MS/MS chromatograms of a drug-free plasma sample (subscript on the left), and drug-free plasma sample spiked with target drug at lower LLOQ concentration.
Table 3
Linearity, accuracy (%RSE), precision (%CV), and matrix effect (%CV) of the column switching LC-MS/MS method.

Analyte Linearity Amount spiked Accuracy Precision Matrix effects

(ng mL−1 )

Linear Equation R2 Lack of Fit Test Internal Standard Intra-assay Inter-assay Intra-assay Inter-assay IS-normalized MF
p (%RSE) n = 5 (%CV) n = 5 (%CV) n = 3
Haloperidol y = 0.077x + 0.0173 0.9985 0.987 Haloperidol-d4 0.075a 4.9 4.6 1.1 1.1
0.225b −3.3 −3.6 3.2 2.8 2.9
20.5c −2.9 −3.1 2.1 1.9
32.5d −0.2 −0.5 3.7 3.3 2.9
40.5e −0.8 −0.5 5.4 4.7
Olanzapine y = 0.0121x − 0.0003 0.9974 0.965 Haloperidol-d4 0.075a 10.4 9.2 4.3 4.5
0.225b −3.0 −3.9 4.7 4.7 7.9
20.5c 3.1 4.2 2.2 3.0
32.5d 4.6 2.9 5.9 6.4 11.8
40.5e −1.1 −1.0 6.1 5.3
Clonazepam y = 0.0227x − 0.013 0.9993 0.953 Clonazepam-d4 0.625a 10.7 11.8 2.3 3.2
1.875b −6.0 −5.4

D.S. Domingues et al. / J. Chromatogr. B 993-994 (2015) 26–35

4.3 4.1 4.7
80.0c −2.9 −2.3 3.2 3.1
125.0d −2.4 −1.8 1.8 2.1 3.4
155.0e 1.4 0.5 5.5 5.2
Mirtazapine y = 0.6038x + 0.5485 0.9985 0.995 Clonazepam-d4 0.125a −12.9 −11.9 1.5 2.3
0.375b −2.8 −2.0 4.2 4.0 6.3
80.0c 1.2 2.9 5.1 5.6
125.0d −3.1 −3.3 6.5 5.6 6.0
155.0e 1.0 1.2 4.8 4.1
Paroxetine y = 0.0165x + 0.0036 0.9945 0.901 Paroxetine-d6 0.250a 2.1 2.1 1.0 0.8
0.750b −1.6 −1.3 2.3 2.1 6.1
80.0c 2.4 2.6 1.0 1.0
125.0d 5.9 6.1 3.9 3.4 1.5
155.0e −3.0 −3.1 4.2 3.6

Analyte Linearity Amount Accuracy Precision Matrix effects

spikedtd:paraenter
(ng mL−1 )
Linear Equation R2 Lack of Fit Test Internal Standard Intra-assay Inter-assay Intra-assay Inter-assay IS-normalized MF
p (%RSE) n = 5 (%CV) n = 5 (%CV) n = 3
a
Citalopram y = 0.0006x + 0.0006 0.9988 0.979 Citalopram-d6 1.250 4.2 3.7 2.6 2.6
3.750b −5.1 −4.2 3.0 3.0 4.3
205.0c −3.0 −2.9 1.9 1.9
325.0d 2.0 2.3 3.5 3.5 4.5
405.0e −0.6 −0.7 3.7 3.7
Sertraline y = 0.0043x + 0.0068 0.9973 0.898 Sertraline-d3 0.625a −11.5 −10.7 3.1 3.0
1.875b −8.9 −8.2 1.8 2.1 3.0
205.0c −5.5 −4.5 2.9 3.4
325.0d 2.9 2.0 0.8 2.1 1.6
405.0e −1.0 −0.7 4.9 4.2
Chlorpromazine y = 0.0028x − 0.0049 0.9983 0.922 Imipramine-d3 0.125a 4.9 4.7 2.7 2.4
0.375b −1.1 −0.6 2.7 2.6 1.8
185.0c −0.6 −0.2 3.2 2.9
290.0d 3.5 2.9 5.1 4.6 7.2
360.0e −1.2 −2.0 2.3 2.6
Imipramine y = 0.0073x + 0.0141 0.9982 0.901 Imipramine-d3 0.125a −6.1 −6.6 3.2 3.0
0.375b −5.2 −5.5 3.0 2.7 6.2
185.0c 0.1 0.3 2.0 1.8
290.0d 0.6 0.6 1.7 1.5 2.0
360.0e 0.5 0.8 2.7 2.4

31
 32
Table 3 (Continued)

Analyte Linearity Amount spiked Accuracy Precision Matrix effects

(ng mL−1 )

Linear Equation R2 Lack of Fit Test Internal Standard Intra-assay Inter-assay Intra-assay Inter-assay IS-normalized MF
p (%RSE) n = 5 (%CV) n = 5 (%CV) n = 3
Clomipramine y = 0.0046x + 0.0098 0.9961 0.865 Clomipramine-d3 0.250a −2.4 -0.5 4.0 4.9
0.750b −7.0 −7.8 3.2 3.3 3.8
260.0c 0.4 0.2 4.3 3.7
410.0d −2.8 −3.4 3.4 3.3 4.0
510.0e 2.9 2.3 1.4 1.8

Analyte Linearity Amount Accuracy Precision Matrix effects

spikedtd:paraenter
(ng mL−1 )
Linear Equation R2 Lack of Fit Test Internal Standard Intra-assay Inter-assay Intra-assay Inter-assay IS-normalized MF
p (%RSE) n = 5 (%CV) n = 5 (%CV) n = 3
Quetiapine y = 0.0104x + 0.0142 0.9976 0.805 Quetiapine-d8 0.125a 4.1 6.3 1.9 4.6
0.375b −5.4 −4.4 2.8 3.3 4.5
−0.7 −0.2

D.S. Domingues et al. / J. Chromatogr. B 993-994 (2015) 26–35

260.0c 2.7 2.5
410.0d −3.7 −3.5 1.7 1.5 3.9
510.0e 1.2 1.4 4.4 3.9
Diazepam y = 0.0021x + 0.0367 0.9972 0.923 Diazepam-d5 0.313a −11.5 −10.3 2.3 3.0
0.939b 0.7 1.6 3.1 3.3 3.9
550.0c −2.2 −1.8 1.4 1.4
850.0d 2.1 2.1 2.4 2.1 3.5
1050.0e −1.3 −1.6 4.4 3.9
Fluoxetine y = 0.003x + 0.027 0.9932 0.956 Fluoxetine-d6 0.313a −1.3 −0.9 2.3 2.1
0.939b 6.9 7.4 2.1 2.0 5.6
550.0c 2.1 2.6 2.7 2.5
850.0d −4.4 −4.2 4.0 3.5 3.8
1050.0e 3.0 2.6 3.0 2.7
Clozapine y = 0.0016x + 0.128 0.9966 0.943 Clozapine-d4 0.188a −4.5 −4.6 3.2 2.8
0.564b −3.6 −2.8 2.2 2.4 3.0
850.0c 1.5 1.8 4.6 4.0
1270.0d 0.8 1.5 3.4 3.3 2.1
1550.0e −1.5 −1.7 2.5 2.2
Carbamazepine y = 0.0005x + 0.0413 0.9940 0.876 Carbamazepine-d10 0.063a −14.0 −13.7 3.4 3.0
0.189b −6.5 −6.5 2.6 2.3 5.0
5500.0c 1.3 1.5 0.8 0.9
8500.0d 1.9 1.8 1.2 1.1 3.4
10500.0e 2.0 2.1 1.4 1.2

Analyte Linearity Amount Accuracy Precision Matrix effects

spikedtd:paraenter
(ng mL−1 )
Linear Equation R2 Lack of Fit Test Internal Standard Intra-assay Inter-assay Intra-assay Inter-assay IS-normalized MF
p (%RSE) n = 5 (%CV) n = 5 (%CV) n = 3
Lamotrigine y = 2E-05x + 5E-05 0.9994 0.995 Carbamazepine-d10 1.250a 3.8 4.8 2.0 2.9
3.750b 10.3 10.1 0.7 0.6 2.7
5500.0c 9.4 9.1 3.5 3.1
8500.0d 9.7 9.7 2.4 2.1 3.8
10500.0e 10.0 9.9 2.7 2.3
p: p-value at a significance level of 0.05.
a
LLOQ.
b
Low QC.
c
Medium QC.
d
High QC.
e
ULOQ.
 D.S. Domingues et al. / J. Chromatogr. B 993-994 (2015) 26–35 33

Table 4
Comparison of the column switching LC-MS/MS method with other counterparts for the determination of drugs in biological samples.

Analytes Matrix Sample amount Total time analyses Capilar type and dimensions Analytical LOQ Range Reference
(␮L) (min) technique (ng mL−1 )

Antidepressants Human 500 15 OV-1701 (14% in-tube 20-50 Silva et al.

Plasma cyanopropylphenyl SPME/LC-UV (2008) [43]
methylpolysiloxane)
80 cm x 250 ␮m ID
Antipsychotic and Human 700 12 poly(N-isopropylacrylamide- in-tube 41.1-238.1 Ma et al. (2009)
Antidepressants Urine co-ethylene dimethacrylate) SPME/LC-UV [44]
(poly(NIPAAm-co-EDMA))
monolithic
15 cm x 250 ␮m ID
S-Norfluoxetine Human 200 21 Polypyrrole (PPY) in-tube 10-15 Silva et al.
Plasma 60 cm x 250 ␮m ID SPME/LC-UV (2009) [45]
R-Norfluoxetine
S-Fluoxetine
R-Fluoxetine
Antidepressants Human 500 30 Cyanoethyl-functionalized in-tube 0.23-9.83 Zheng et al.
Plasma hybrid silica monolith SPME/LC- (2010) [29]
15 cm x 250 ␮m ID MS/MS
Antipsychotics Human 200 20 Cyanopropyl-functionalized column 0.075-0.188 This paper
Plasma hybrid silica monolith switching
4.5 cm x 530 ␮m ID LC-MS/MS
Antidepressants 0.125-1.250
Anxiolytics 0.313-0.630
Anticonvulsants 0.063-1.250

lower limit of quantification (63.0 pg mL−1 to 1250.0 pg mL−1 ) to
the upper limit of quantification (40.5 ng mL−1 to 10.5 ␮g mL−1 ),
with a coefficient of determination higher than 0.9932. This lin-
ear range was established on the basis of the therapeutic interval.
The test of Lack of Fit also confirmed the linearity of this method
(Table 3). The calibration standards (n = 5) presented CV lower than
14%.
Table 3 illustrates the intra- and inter-assay accuracy of the
method evaluated between the LLOQ and the ULOQ. The accu-
racy of the method presented RSE values ranging from −14 to 10%
(intra-assay accuracy) and from −13 to 11% (inter-assay accuracy).
The precision of the method gave CV values ranging from 0.7 to
6.5% (intra-assay precision) and 0.6 to 6.4% (inter-assay precision)
(Table 3).
Carryover in the blank samples following the ULOQ analysis was
not higher than 0.5% of the analytes signal of the LLOQ, or 0.1% of
the IS signal of the LLOQ. The CV of the IS-normalized MF (matrix
effects) calculated from the five lots of matrix was not greater than
15% (Table 3).
The proposed method presented LLOQ values lower than the
LLOQ values obtained from in-tube SPME-LC methods described
in the literature for determination of drugs in plasma samples
using different capillaries, such as hybrid monolith with cya-
noethyl (0.23–9.83 ng mL−1 ) [29], OV-1701 (20–50 ng mL−1 ) [43],
poly (N-isopropylacrylamide-co-ethylene dimethacrylate) (poly
(NIPAAm-co- EDMA)) (41.1–238.1 ng mL−1 ) [44], and polypyr-
role (10–15 ng mL−1 ) [45]. Moreover, compared with the method
described by Zheng et al. [29] for determination of antidepressants
in plasma samples by in-tube SPME-LC-MS/MS (cyanoethyl-hybrid
silica monolithic capillary), the method proposed here provided: (i)
shorter analysis time (20 min), (ii) smaller plasma volume (200 ␮L),
and (iii) simultaneous determination of a large number of drugs
(five anticonvulsants, seven antidepressants, two anticonvulsi-
vants, and two anxiolytics) co-administered during the treatment
of schizophrenia (Table 4).
3.4. Determination of the target drugs in plasma samples from
schizophrenic patients
Analyses of plasma samples from schizophrenic patients under
treatment with psychotropic drugs proved the applicability of
the developed method. Fig. 5 illustrates the chromatograms of
plasma samples obtained from patients undergoing polytherapy
with citalopram, clozapine, and lamotrigine. Table 5 lists the
drugs plasma levels in the ten schizophrenic patients; the levels
were in agreement with the therapeutic drug intervals [46]. The
plasma levels had CV values lower than 5.0%; the chromatograms
did not present interferences at the retention times of the
drugs.

Table 5
Plasma levels of psychotropics drugs in schizophrenic‘s plasma samples.

Drugs Therapeutic Range Plasma Levels (ng mL−1 )

(ng mL−1 ) [46]

1 2 3 4 5 6 7 8 9 10

Haloperidol 1–10 – – – – – 1 – – – –
Olanzapine 20–80 – – 23 – – – – – – 4
Citalopram 50–110 – – – – 149 18 35 – – –
Sertraline 10–150 – 11 – 28 – – – – – –
Imipramine 175–300 – – 240 – – – – – – –
Clomipramine 230–450 – – – – – – – – 402 –
Fluoxetine 120–500 – – – – – – – – – 99
Clozapine 350–600 819 434 – 1152 1267 688 1507 1253 752 –
Lamotrigine 3000–14000 1136 – – 5666 3326 – – 2897 – –

–: not detected.
34 D.S. Domingues et al. / J. Chromatogr. B 993-994 (2015) 26–35
Fig. 5. LC-MS/MS chromatograms of the plasma sample from schizophrenic patient. Plasma levels: citalopram: 149.0 ng mL−1 ; clozapine: 1267.0 ng mL−1 ; and lamotrigine:
3326.0 ng mL−1 .
4. Conclusion
The selective organic–inorganic hybrid cyanopropyl monolithic
capillary presented suitable sorption capability that enabled the
analysis of drugs (antidepressants, anticonvulsants, anxiolytics,
and antipsychotics) in plasma samples at sub-therapeutic levels.
The low backpressure and the high permeability of the monolithic
capillary facilitated biological sample percolation. The chemical
and mechanical stability of this capillary allowed for its reuse over
one hundred times.
The column-switching LC-MS/MS arrangement improved the
sample throughput (in the straight-flushing mode), as well
as the sensitivity (drugs pre-concentration) and the selectivity
(cyanopropyl groups) of the method. This method was successfully
applied for therapeutic drug monitoring of schizophrenic patients
under treatment with psychotropic drugs.
Acknowledgements
This work was supported by grants from Fundação de Amparo à
Pesquisa do Estado de São Paulo (FAPESP, n. 2012/10705-9) and
Conselho Nacional de Desenvolvimento Científico e Tecnológico
(CNPq).
References
[1] J. Addington, K.S. Cadenhead, T.D. Cannon, B. Cornblatt, T.H. McGlashan,
D.O. Perkins, L.J. Seidman, M. Tsuang, E.F. Walker, S.W. Woods, R. Heinssen,
North American Prodrome Longitudinal Study: A collaborative multisite
approach to prodromal schizophrenia research, Schizophr. Bull. 33 (2007)
665–672.
[2] E. Asor, D. Ben-Shachar, Platelets A possible glance into brain biological pro-
cesses in schizophrenia, World J. Psychiatry. 2 (2012) 124–133.
[3] C. Reeder, N. Smedley, K. Butt, D. Bogner, T. Wykes, Cognitive predictors of social
functioning improvements following cognitive remediation for schizophrenia,
Schizophr. Bull. 32 (Suppl. 1) (2006) 123–131.
[4] S. Leucht, T.R. Barnes, W. Kissling, R.R. Engel, C. Correll, J.M. Kane, Relapse pre-
vention in schizophrenia with new-generation antipsychotics: a systematic
review and exploratory meta-analysis of randomized controlled trials, Am. J.
Psychiatry 160 (2003) 1209–1222.
[5] C.U. Correll, Real-life switching strategies with second-generation antipsy-
chotics, J. Clin. Psychiat. 67 (2006) 160–161.
[6] C. Tsutsumi, H. Uchida, T. Suzuki, K. Watanabe, H. Takeuchi, S. Nakajima, Y.
Kimura, Y. Tsutsumi, K. Ishii, Y. Imasaka, S. Kapur, The evolution of antipsychotic
switch and polypharmacy in natural practice - A longitudinal perspective,
Schizophr. Res. 130 (2011) 40–46.
[7] M. Šagud, B. Vuksan-Ćusa, M. Živković, S. Vlatković, M. Kramarić, Z. Bradaš,
A. Mihaljević-Peleš, Antipsychotics: to combine or not to combine? Psychiatr.
Danubina. 25 (2013) 306–310.
[8] R.W. Buchanan, D.C. Javitt, S.R. Marder, N.R. Schooler, J.M. Gold, R.P. McMa-
hon, U. Heresco-Levy, W.T. Carpenter, The Cognitive and Negative Symptoms
in Schizophrenia Trial (CONSIST): the efficacy of glutamatergic agents for
negative symptoms and cognitive impairments, Am. J. Psychiatry 164 (2007)
1593–1602.
 D.S. Domingues et al. / J. Chromatogr. B 993-994 (2015) 26–35
[9] H. Kataoka, K. Saito, Recent advances in column switching sample preparation
in bioanalysis, Bioanalysis 4 (2012) 809–832.
[10] M. Jayamanne, I. Granelli, A. Tjernberg, P.O. Edlund, Development of a two-
dimensional liquid chromatography system for isolation of drug metabolites,
J. Pharm. Biomed. Anal. 51 (2010) 649–657.
[11] E. Eckert, T. Goen, Rapid determination of four short-chain alkyl mercapturic
acids in human urine by column-switching liquid chromatography-tandem
mass spectrometry, J. Chromatogr. B 965 (2014) 54–60.
[12] V.F. Fagundes, C.P. Leite, G.A. Pianetti, C. Fernandes, Rapid and direct analysis
of statins in human plasma by column-switching liquid chromatography with
restricted-access material, J. Chromatogr. B 947–948 (2014) 8–16.
[13] M. Park, J. Kim, Y. Park, S. In, E. Kim, Y. Park, Quantitative determination of
11-nor-9-carboxy-tetrahydrocannabinol in hair by column switching LC-ESI-
MS(3), J. Chromatogr. B 947-948 (2014) 179–185.
[14] X. Xiong, Z. Yang, Y. Li, L. Xiao, L. Jiang, Y. Chen, M. Ma, B. Chen, Prepara-
tion of a polyhedral oligomeric silsesquioxane-based perfluorinated monolithic
column, J. Chromatogr. A 1304 (2013) 85–91.
[15] F. Svec, Stellan Hjerten’s contribution to the development of monolithic sta-
tionary phases, Electrophoresis 29 (2008) 1593–1603.
[16] J. Ou, Z. Liu, H. Wang, H. Lin, J. Dong, H. Zou, Recent development of hybrid
organic-silica monolithic columns in CEC and capillary LC, Electrophoresis 00
(2014) 1–14.
[17] L.N. Tran, S. Dixit, J.H. Park, Enantioseparation of basic chiral compounds on
a clindamycin phosphate-silica/zirconia hybrid monolith by capillary elec-
trochromatography, J. Chromatogr. A 1356 (2014) 289–293.
[18] M. Wu, R. Wu, Z. Zhang, H. Zou, Preparation and application of organic-
silica hybrid monolithic capillary columns, Electrophoresis 32 (2011)
105–115.
[19] X.F. Liu, N. Sun, Q.F. Zhu, M. Wu, Y. Ye, H.X. Chen, Preparation, characteriza-
tion and application of organic-inorganic hybrid caffeine imprinted monolith,
J. Chromatogr. A 1304 (2013) 10–17.
[20] F. Brothier, V. Pichon, Miniaturized DNA aptamer-based monolithic sorbent for
selective extraction of a target analyte coupled on-line to nanoLC, Anal. Bioanal.
Chem. 406 (2014) 7875–7886.
[21] D.Y. Lee, B.P. Bowen, T.R. Northen, Mass spectrometry-based metabolomics,
analysis of metabolite-protein interactions, and imaging, Biotechniques 49
(2010) 557–565.
[22] H.H. Yang, Y.Z. Chen, Y.X. Liu, L.H. Nie, S.Z. Yao, One-pot synthesis of (3-
sulfopropyl methacrylate potassium)-silica hybrid monolith via thiol-ene click
chemistry for CEC, Electrophoresis 34 (2013) 510–517.
[23] E. Ortiz-Villanueva, F. Benavente, E. Gimenez, F. Yilmaz, V. Sanz-Nebot, Prepa-
ration and evaluation of open tubular C18-silica monolithic microcartridges
for preconcentration of peptides by on-line solid phase extraction capillary
electrophoresis, Anal. Chim. Acta 846 (2014) 51–59.
[24] Y. Zhang, H.Y. Liu, X.Y. Zhang, H. Lei, L.G. Bai, G.L. Yang, On-line solid phase
extraction using organic-inorganic hybrid monolithic columns for the deter-
mination of trace beta-lactam antibiotics in milk and water samples, Talanta
104 (2013) 17–21.
[25] N. Gilart, R.M. Marce, P.A. Cormack, N. Fontanals, F. Borrull, Development of
new polar monolithic coatings for stir bar sorptive extraction, J. Sep. Sci. 37
(2014) 2225–2232.
[26] M. Mei, X. Huang, D. Yuan, Multiple monolithic fiber solid-phase microextrac-
tion: a new extraction approach for aqueous samples, J. Chromatogr. A 1345
(2014) 29–36.
[27] X.F. Liu, Q.F. Zhu, H.X. Chen, L.Z. Zhou, X.P. Dang, J.L. Huang, Preparation of 2,4-
dichlorophenoxyacetic acid imprinted organic-inorganic hybrid monolithic
column and application to selective solid-phase microextraction, J. Chro-
matogr. B 951 (2014) 32–37.
[28] M.M. Zheng, G.D. Ruan, Y.Q. Feng, Hybrid organic-inorganic silica monolith
with hydrophobic/strong cation-exchange functional groups as a sorbent for
micro-solid phase extraction, J. Chromatogr. A 1216 (2009) 7739–7746.
35
[29] M.M. Zheng, S.T. Wang, W.K. Hu, Y.Q. Feng, In-tube solid-phase microextrac-
tion based on hybrid silica monolith coupled to liquid chromatography-mass
spectrometry for automated analysis of ten antidepressants in human urine
and plasma, J. Chromatogr. A 1217 (2010) 7493–7501.
[30] A. Schaefer, F. Piquard, P. Haberey, Plasma amino acids analysis - effects
of delayed samples preparation and of storage, Clin. Chim. Acta 164 (1987)
163–169.
[31] S. Sahai, S. Uhlhaas, Stability of amino acids in human plasma, Clin. Chim. Acta
148 (1985) 255–259.
[32] M. Breier, S. Wahl, C. Prehn, M. Fugmann, U. Ferrari, M. Weise, F. Banning, J.
Seissler, H. Grallert, J. Adamski, A. Lechner, Targeted metabolomics identifies
reliable and stable metabolites in human serum and plasma samples, Plos One
9 (2014) e89728.
[33] O. Gonzalez, M.E. Blanco, G. Iriarte, L. Bartolome, M.I. Maguregui, R.M. Alonso,
Bioanalytical chromatographic method validation according to current reg-
ulations, with a special focus on the non-well defined parameters limit of
quantification, robustness and matrix effect, J. Chromatogr. A 1353 (2014)
10–27.
[34] G. Zhang, A.V.Jr. Terry, M.G. Bartlett, Simultaneous determination of five
antipsychotic drugs in rat plasma by high performance liquid chromatography
with ultraviolet detection, J. Chromatogr. B 856 (2007) 20–28.
[35] J. Bhatt, G. Subbaiah, S. Singh, Liquid chromatography/tandem mass spec-
trometry method for simultaneous determination of risperidone and its active
metabolite 9-hydroxyrisperidone in human plasma, Rapid Commun. Mass
Spectrom. 20 (2006) 2109–2114.
[36] W. Li, D.P. Fries, A. Malik, Sol–gel stationary phases for capillary electrochro-
matography, J. Chromatogr. A 1044 (2004) 23–52.
[37] L. Xu, H.K. Lee, Preparation, characterization and analytical application of a
hybrid organic-inorganic silica-based monolith, J. Chromatogr. A 1195 (2008)
78–84.
[38] A. McKean, J. Vella-Brincat, E. Begg, Prescribing and monitoring clozapine in
christchurch, Australas Psychiatry. 16 (2008) 263–267.
[39] S. Matkó, A. Toldy, S. Keszei, P. Anna, G. Bertalan, G. Marosi, Flame retardancy
of biodegradable polymers and biocomposites, Polym. Degrad. Stab. 88 (2005)
138–145.
[40] P. Innocenzi, Infrared spectroscopy of sol–gel derived silica-based films:
a spectra-microstructure overview, J. Non-Crystalline Solids 316 (2003)
309–319.
[41] R. Al-Oweini, H. El-Rassy, Synthesis and characterization by FTIR spectroscopy
of silica aerogels prepared using several Si(OR)4 and R  Si(OR )3 precursors, J.
Mol. Struct. 919 (2009) 140–145.
[42] M. Wahab, Hybrid periodic mesoporous organosilica materials prepared from
1 2-bis(triethoxysilyl)ethane and (3-cyanopropyl)triethoxysilane, Micropor.
Mesopor. Mater. 69 (2004) 19–27.
[43] B.J.G. Silva, F.M. Lancas, M.E.C. Queiroz, In-tube solid-phase microextraction
coupled to liquid chromatography (in-tube SPME/LC) analysis of nontricyclic
antidepressants in human plasma, J. Chromatogr. B 862 (2008) 181–188.
[44] Q. Ma, M. Chen, Z.G. Shi, Y.Q. Feng, Preparation of a poly(N-isopropyl-
acrylamide-co-ethylene dimethacrylate) monolithic capillary and its appli-
cation for in-tube solid-phase microextraction coupled to high-performance
liquid chromatography, J. Sep. Sci. 32 (2009) 2592–2600.
[45] B.J.G. Silva, F.M. Lancas, M.E.C. Queiroz, Determination of fluoxetine and nor-
fluoxetine enantiomers in human plasma by polypyrrole-coated capillary
in-tube solid-phase microextraction coupled with liquid chromatography-
fluorescence detection, J. Chromatogr. A 1216 (2009) 8590–8597.
[46] C. Hiemke, P. Baumann, N. Bergemann, A. Conca, O. Dietmaier, K. Egberts, M.
Fric, M. Gerlach, C. Greiner, G. Grunder, E. Haen, U. Havemann-Reinecke, E.J.
Sirot, H. Kirchherr, G. Laux, U.C. Lutz, T. Messer, M.J. Muller, B. Pfuhlmann, B.
Rambeck, P. Riederer, B. Schoppek, J. Stingl, M. Uhr, S. Ulrich, R. Waschgler,
G. Zernig, AGNP Consensus Guidelines for Therapeutic Drug Monitoring in
Psychiatry: Update 2011, Pharmacopsychiatry 44 (2011) 195–235.