BIOMEDICAL CHROMATOGRAPHY

Biomed. Chromatogr. 22: 702711 (2008)

Published
702 online 3 MarchRESEARCH
ORIGINAL 2008 in Wiley InterScience ORIGINALH.
RESEARCH
Umezawa et al.
(www.interscience.wiley.com) DOI: 10.1002/bmc.987

Simultaneous determination of -blockers in human plasma

using liquid chromatographytandem mass spectrometry

Hironobu Umezawa, Xiao-Pen Lee, Yoshiko Arima, Chika Hasegawa, Hikaru Izawa, Takeshi Kumazawa*
and Keizo Sato
Department of Legal Medicine, Showa University School of Medicine, 1-5-8 Hatanodai, Shinagawa-ku, Tokyo 142-8555, Japan

Received 5 September 2007; revised 22 November 2007; accepted 28 November 2007

ABSTRACT: A detailed procedure for the analysis of four -blockers, acebutolol, labetalol, metoprolol and propranolol, in
human plasma by high-performance liquid chromatography (LC)tandem mass spectrometry (MS-MS) using an MSpak GF
column, which enables direct injection of crude plasma samples, is presented. Protein and/or macromolecule matrix compounds
were eluted rst from the column, while the drugs were retained on the polymer stationary phase of the MSpak GF column. The
analytes retained on the column were then eluted into an acetonitrile-rich mobile phase using a gradient separation technique. All
drugs showed base peak ions due to [M + H]+ ions by LC-MS with positive ion electrospray ionization, and the product ions were
produced from each [M + H]+ ion by LC-MS-MS. Quantication was performed by selected reaction monitoring. The recoveries
of the four -blockers spiked into plasma were 73.589.9%. The regression equations for all compounds showed excellent linearity
in the range 10 1000 ng/mL of plasma, with the exception of propranolol (10800 ng/mL). The limits of detection and quantica-
tion for each drug were 13 and 10 ng/mL, respectively, of plasma. The intra- and inter-day coefcients of variation for all drugs in
plasma were not greater than 10.9%. Copyright 2008 John Wiley & Sons, Ltd.

KEYWORDS: LC-MS-MS; -blockers; human plasma; direct injection

INTRODUCTION methods coupled to mass spectrometry, including tandem

-Adrenoceptor antagonists, also known as -blockers,
are prescription drugs mainly used for the treatment of
hypertension, congestive heart failure and abnormal
heart rhythms, to relieve angina, and to prevent cardiac
infarctions in humans (Johnsson and Regrdh, 1976;
Prichard, 1978). The most commonly prescribed -blockers
in Japan include acebutolol (Sectral), labetalol (Trandate),
metoprolol (Lopresor) and propranolol (Inderal). These
drugs are frequently encountered in clinical or forensic
toxicological practice, because of their relatively narrow
safe therapeutic dose ranges (Tracqui et al., 1992; Reith
et al., 1996; Johnson and Lewis, 2006; Unverir et al.,
2007). Therefore, monitoring of these drugs is required
in order to allow optimum dosage adjustments, and for
clinical and forensic toxicological analysis.
Several gas chromatographic (Quaglio et al., 1993; Black
et al., 1996; Hartonen and Riekkola, 1996; Amendola
et al., 2000) and liquid chromatographic (Naidong et al.,
2002; Maurer et al., 2004; do Carmo Borges et al., 2005;
Josefsson and Sabanovic, 2006; Keski-Rahkonen et al.,
2007; Li et al., 2007; Ding et al., 2007; Fanali et al., 2007)
*Correspondence to: T. Kumazawa, Department of Legal Medicine,
Showa University School of Medicine, 1-5-8 Hatanodai, Shinagawa-
ku, Tokyo 142-8555, Japan.
E-mail: kumazawa@med.showa-u.ac.jp
Abbreviations used: LLE, liquidliquid extraction; SPE, solid-phase
extraction.
mass spectrometry (MS-MS), have been developed in
order to determine the levels of acebutolol, labetalol,
metoprolol and propranolol in biological samples. These
methods utilized liquidliquid extraction (LLE; Amendola
et al., 2000; do Carmo Borges et al., 2005; Keski-
Rahkonen et al., 2007; Li et al., 2007) or off-line solid-
phase extraction (SPE; Quaglio et al., 1993; Black et al.,
1996; Hartonen and Riekkola, 1996; Naidong et al., 2002;
Maurer et al., 2004; Josefsson and Sabanovic, 2006)
for specimen preparation and concentration prior to
chromatography. However, LLE and off-line SPE meth-
ods are laborious and time-consuming.
Recently, a direct-injection liquid chromatography
(LC)mass spectrometry (MS) method using size exclusion
chromatography has been introduced as a new technique
for the determination of drug levels in biological samples
(Arinobu et al., 2001). An MSpak GF column, size-
exclusion chromatography column, consisting of a highly
cross-linked hard gel of polyvinyl alcohol, was developed
for use in LC-MS in Japan. The MSpak GF column
combines on-line extraction with chromatographic
separation in a single step. In a previous study, we
reported on the use of an MSpak GF column in com-
bination with LC-MS or LC-MS-MS for the analysis
of benzodiazepines (brotizolam, etizolam, lorazepam
and triazolam; Lee et al., 2003, 2006) and antiallergics
(cetirizine, ibudilast, ketotifen and olopatadine; Fujimaki
et al., 2006), and demonstrated that this column was

Copyright 2008 John Wiley & Sons, Ltd. Biomed. Chromatogr. 22: 702711 (2008)
DOI: 10.1002/bmc
Simultaneous determination of -blockers in human plasma
effective for the separation of directly injected biologi-
cal uids containing these drugs, with good linearity and
reproducibility. In the present study, we have estab-
lished a detailed procedure for LC-MS-MS analysis
of four -blockers, acebutolol, labetalol, metoprolol and
propranolol, by direct injection of human plasma using
an MSpak GF column. To the best of our knowledge,
this is the rst report dealing with the use of LC-MS-
MS for the determination of -blocker levels following
direct injection method for crude biological samples.
EXPERIMENTAL
Materials. Acebutolol, labetalol, metoprolol, propranolol and
pindolol as an internal standard (IS) were purchased from
Sigma (St Louis, MO, USA). Syringe lters with a polyvinyliden
uoride membrane (Whatman 13 mm GD/X syringe lter,
13 mm diameter and 0.2 m pore size) and LC-MS grade
acetonitrile were obtained from Whatman (Clifton, NJ, USA)
and Wako Pure Chemical (Osaka, Japan), respectively. Other
common chemicals used were of the highest purity commer-
cially available. A water purier WG222 (Yamato Scientic,
Tokyo, Japan) and an Autopure ultraviolet water purication
system WR600G (Yamato Scientic) were used to obtain
puried water for the LC solvent.
Preparation of standard solutions and quality control
(QC). Stock standard solutions of acebutolol, labetalol,
metoprolol, propranolol and IS were prepared separately by
dissolving an accurately weighed amount of each compound
in methanol to achieve a concentration of 1 mg/mL. All stock
standard solutions were stored at 4C. Working standard
solutions of these compounds were prepared by serial dilution
of the stock standard solutions using the initial LC mobile
phase (10 mM ammonium acetate in puried water). All work-
ing standard solutions were freshly prepared every week and
stored at 4C. Calibration standards were prepared by mixing
appropriate amounts of the working standard solutions and
drug-free plasma to achieve eight different concentrations
from 10 to 1000 ng/mL for acebutolol, labetalol, metoprolol
and propranolol, and 1000 ng/mL for IS. QC samples were
prepared by the same procedure as the calibration standards,
and concentrations were 101000 ng/mL for acebutolol,
labetalol, metoprolol and propranolol, and 50 1000 ng/mL
for IS.
Preparation of plasma samples. Drug-free whole blood
samples from healthy volunteers, who were recruited from
the laboratory personnel, were taken intravenously in the pres-
ence of EDTA-2Na as an anticoagulant. To prepare drug-free
plasma samples, EDTA-whole blood was centrifuged at 4000 rpm
for 15 min at 4C, and plasma was decanted into a clean cen-
trifuge tube. Drug-free plasma samples were stored at 80C
until use. Human plasma (1 mL) samples containing each of
the four drugs and IS were mixed with 3 mL of puried water
containing 13.3 mM ammonium acetate. After centrifugation
of the mixture at 9000 rpm for 10 min, the supernatants were
ltered through a syringe lter, and a 100 L aliquot of clear
supernatant was directly injected into the LC system.
Copyright 2008 John Wiley & Sons, Ltd.
ORIGINAL RESEARCH 703
Evaluation of recovery, quantification and linearity.
Recovery was calculated by comparing the chromatographic
peak areas obtained from extracts of QC samples with those
obtained by direct injection of standard compounds dissolved
in 10 mM ammonium acetate, and determined at three differ-
ent concentrations (low, medium, high) for each compound.
Regression equations for the four drugs were obtained by
plotting the peak-area ratios of the analytes to the IS (y-axis)
vs the analyte concentration (ng/mL; x-axis) in spiked plasma
samples. The equations for acebutolol, labetalol, metoprolol
and propranolol were constructed using pindolol as IS. The
concentrations of the calibrators ranged from 10 to 1000 ng/
mL for acebutolol, labetalol and metoprolol (eight calibrators:
10, 20, 40, 80, 100, 200, 500 and 1000 ng/mL) and from 10 to
800 ng/mL for propranolol (eight calibrators: 10, 20, 40, 80,
100, 200, 500 and 800 ng/mL). Each point was determined
in duplicate. The concentration of IS was xed at 1000 ng in
1 mL of plasma. The regression equations of the calibration
curves were then used to calculate the analyte concentrations
in QC samples. Intra-day coefcient of variation (CV) was
determined by replicate analysis of QC samples spiked with
three different concentrations (low, medium, high) of acebutolol,
labetalol, metoprolol and propranolol on the same day. The
same procedure was repeated for 5 days in order to deter-
mine inter-day CV. Accuracy was expressed as the mean
calculated concentration as a percentage of the nominal
concentration.
The limit of detection (LOD) was dened as the lowest
concentration of analyte spiked in plasma that could be
detected with a signal-to-noise ratio of at least 3. The limit
of quantication (LOQ) was dened as the lowest concentra-
tion on the calibration curve that could be measured with
a signal-to-noise ratio of at least 10, an intra-day CV 20%,
and an accuracy of 80 120%. The acceptance criterion for the
correlation coefcient was >0.999.
Matrix effect . The matrix effect was investigated by analyzing
ve individual drug-free blank plasma samples, and evaluated
at three different concentrations (5, 50 and 1000 ng/mL) of
analyte standards. The matrix effect was calculated by com-
paring the chromatographic peak areas for analytes in blank
plasma samples spiked after sample preparation with those
obtained by direct injection of standard compounds dissolved
in 10 mM ammonium acetate.
LC conditions. The LC system consisted of an Agilent 1100
Series (Agilent Technologies, Inc., Palo Alto, CA, USA)
equipped with a G1315A diode-array detector (DAD; Agilent)
set at 260 nm, and a G1313A autosampler (Agilent). The
mobile phase was degassed on-line using a G1322A vacuum
membrane degasser (Agilent). Chromatographic separations
were performed using a Shodex MSpak GF-310 4B column
(50 4.6 mm i.d., particle size 6 m, Showa Denko, Tokyo).
The mobile phases were 10 mM ammonium acetate in puried
water (solvent A) and 100% acetonitrile (solvent B). The
following gradient program was used: 100% solvent A 0%
solvent B for 3 min, changing to 0% solvent A100% solvent
B between 3 and 4 min, and constant 0% solvent A100%
solvent B until 9.5 min. A further 5.5 min at 100% solvent A
0% solvent B was used to equilibrate the column for the next
injection. A constant ow rate of 0.55 mL/min, a column
Biomed. Chromatogr. 22: 702711 (2008)
DOI: 10.1002/bmc
704 ORIGINAL RESEARCH
temperature of 30C, and a sample size of 100 L were used
for all injections. An in-line lter (SUMIPAX lter PG-ODS,
2 m pore size, Sumika Chemical Analysis Service, Osaka) was
installed between the autosampler and the MSpak GF column
for protection of the column and the mass spectrometer.
MS-MS conditions. An API 2000 triple-stage quadrupole mass
spectrometer (Applied Biosystems/MDS SCIEX, Ontario,
Canada), equipped with a TurboIonSpray ion source and an
electric ten-port diverter valve, was used for mass spectral
recording and determination of the analyte concentrations.
Ionization of the analytes was carried out using electrospray
ionization (ESI) in the positive mode: TurboIonSpray tem-
perature, 490C; ion source voltage, 3 kV; ring voltage, 390 V;
and nebulizer gas (high-purity air), heater gas (high-purity
air), and curtain gas (high-purity nitrogen) pressures of 20,
80 and 40 psi, respectively. Orice voltage was optimized for
the best response of [M + H]+ ions for each analyte: 41 V
for acebutolol, 25 V for labetalol, 40 V for metoprolol, 51 V
for propranolol and 40 V for the IS. Full-scan data were
obtained with a mass range of m /z 50500, a dwell time of
200 ms, and a step-size of 0.1 amu. MS-MS product ion spec-
tra were produced by collision-activated dissociation of the
protonated molecule ion [M + 1]+ using nitrogen as collision
gas at a setting of 4 (arbitrary units). Peak widths of precur-
sor ions were maintained at 0.60.75 amu in selected reaction
monitoring (SRM). Quantication was performed using SRM
of precursor product ion transitions at m /z 337 116
(collision energy, 29 eV) for acebutolol, m /z 329 311
(collision energy, 22 eV) for labetalol, m /z 268 116 (colli-
sion energy, 25 eV) for metoprolol, m /z 260 116 (collision
energy, 25 eV) for propranolol and m /z 249 116 (collision
energy, 25 eV) for IS. Instrument control, data acquisition,
peak integration and calculations were accomplished using
MassChrom 1.1.1 software. Mass tuning was performed by
infusing a 0.1 mM polypropylene glycol solution into the ion
source.
Analytical system. The instrumental schema of the proposed
direct-injection LC-MS-MS method with the MSpak GF
column is shown in Fig. 1. A 100 L volume of the diluted
plasma sample was injected automatically into the MSpak
GF column at a consistent ow rate of 0.55 mL/min. The
Figure 1. Schematic diagram of the present direct-injection LC-MS-MS system.
DAD, diode array detector; ESI, electrospray ionization.
Copyright 2008 John Wiley & Sons, Ltd.
H. Umezawa et al.
eluent from the MSpak GF column was connected directly
to the mass spectrometer for detection via an electric 10-port
diverter valve. To avoid contamination of the mass spectro-
meter with matrix molecules, the eluent was directed to the
waste pool by the diverter valve during the early stages. After
6 min, the diverter valve was switched to the injection posi-
tion, and the analytes were introduced into the ESI-MS-MS
system to detect the test compounds. After an analysis time
of 9.5 min, the diverter valve was switched back to the waste
position and the mobile phase was changed from 0% solvent
A100% solvent B to 100% solvent A0% solvent B.
RESULTS AND DISCUSSION
Mass spectra
Structurally, most -blockers have a common CH(OH)
CH2NH side chain. In the ve compounds used in this
study, this chain is either bonded via a CH2O bond
(acebutolol, metoprolol, propranolol and IS) or directly
to an aromatic ring (labetalol). The mass spectra for
the analyte compounds obtained by LC-MS and LC-
MS-MS are shown in Table 1 together with the pro-
posed fragmentation pathways for each compound.
Acebutolol, labetalol, metoprolol, propranolol and IS
showed protonated molecular ions [M + H]+ at m /z
337, 329, 268, 260 and 249, respectively, in the full-scan
mode by LC-MS. These protonated molecular ions
were subjected to product ion formation by LC-MS-
MS. Acebutolol, metoprolol, propranolol and IS gave
a base peak or prominent peak ions at m /z 116, which
corresponded to cleavage of the carbonoxygen bond
in the dimethylaminopropoxyl side chain. Labetalol
produced a base peak ion with an m /z of 311 corre-
sponding to [M + H H2O]+. Product ions with m /z
values of 191, 183 and 172 corresponding to [M H2O
C3H7NH]+ were obtained from the protonated ion
[M + H]+ of metoprolol, propranolol and IS, respec-
tively. Propranolol also gave an intense product ion at
m /z 157, probably corresponding to [M C5H11ONH]+.
Biomed. Chromatogr. 22: 702711 (2008)
DOI: 10.1002/bmc
Simultaneous determination of -blockers in human plasma ORIGINAL RESEARCH 705

Table 1. Chemical structures of the four -blockers and the IS used in the present study and their mass spectra obtained by LC-
MS and LC-MS-MS

MS spectrum MS/MS spectrum

Product ion
m /z (relative Precursor m /z (relative
Compound Structure MW intensity, %) ion m /z intensity, %)
Acebutolol 336 337 (100) 337 337 (34)
116 (100)

Labetalol 328 329 (100) 329 329 (14)

311 (49) 311 (100)

Metoprolol 267 268 (100) 268 268 (100)

191 (21)
116 (44)

Propranolol 259 260 (100) 260 260 (75)

183 (61)
157 (35)
116 (100)

Pindolol (IS) 248 249 (100) 249 249 (44)

172 (27)
116 (100)

MW, molecular weight.

alcohol phase. This principle enables direct injection

Optimization of conditions
Separation by the MSpak GF column is basically size-
exclusion chromatography associated with the slight ac-
tion of partition and adsorption. This column is suitable
for eliminating macromolecular compounds such as
proteins from biological samples, because their molecu-
lar size is too large to enter the pores of the stationary
phase, whereas drugs with small molecular weights
can enter the pores and be retained by the polyvinyl
of crude biological samples. The polymer support is
chemically and structurally stable, and can be used with
a wide pH range (between 2 and 9). Another advan-
tage of the present polymer support is that both water
and various organic solvents can be used for elution.
Several mobile phases have been tested for the
separation of -blockers (Naidong et al., 2002; Maurer
et al., 2004; Johnson and Lewis, 2006; Josefsson and
Sabanovic, 2006). Mobile phases based on acetonitrile

Copyright 2008 John Wiley & Sons, Ltd. Biomed. Chromatogr. 22: 702711 (2008)
DOI: 10.1002/bmc
706 ORIGINAL RESEARCH
or on methanol and ammonium acetate or water as
proposed in the literature (Naidong et al., 2002; Johnson
and Lewis, 2006; Fanali et al., 2007; Ding et al., 2007),
were initially optimized for the separation of -blockers.
The results indicated that the response was higher with
acetonitrile than with methanol in the mobile phase;
therefore, the former was used in subsequent experi-
ments. An increase in the acetonitrile content of the
mobile phase caused shorter retention time and poor
separation of the ve analytes; a decrease in the
acetonitrile content resulted in longer retention times.
The addition of ammonium acetate to the mobile phase
caused remarkable increases in the intensities of the
protonated ion peaks [M + H]+ of the compounds
by single MS. Therefore, the mobile phase composition
of 10 mM ammonium acetate in puried water and
acetonitrile was nally adopted and the gradient mode
was also optimized to achieve good analyte separation.
The direct-injection method is generally dirtier than
conventional extraction methods. We had previously
conducted various preliminary experiments for the opti-
mization of drug separation using the present direct-
injection LC-MS-MS method with the MSpak GF-310
4B column. Plasma samples were diluted four-fold with
the mobile phase, and 100 L of the diluted samples
was introduced into LC-MS-MS after ltration with
a syringe lter. Figure 2 shows LC chromatograms
obtained using a DAD (at 260 nm), for monitoring
proteins plus matrix compounds as well as the four -
blockers and IS, which had been spiked into the plasma
Figure 2. LC-DAD chromatograms with the MSpak GF column recorded at 260 nm for the
four -blockers and IS in human plasma. The amount of metoprolol, acebutolol, labetalol,
propranolol and IS spiked into 1 mL of plasma was 30 g. Peaks: 1, labetalol; 2, metoprolol;
3, acebutolol; 4, propranolol; 5, IS.
Copyright 2008 John Wiley & Sons, Ltd.
H. Umezawa et al.
samples at concentrations of 30 g/mL. As shown in the
chromatograms, a broad matrix of peaks appeared
around 1.0 and 5.5 min. Therefore, the eluates from
the MSpak GF column were discarded up to 6 min and
then were automatically introduced into the MS using
the diverter valve to detect the analytes, ensuring the
cleanliness of the ion source of the mass spectrometer.
Method validation
An IS is necessary for quantitative analysis of drug
compounds in biological samples. There are two types
of IS that can be used for drug monitoring by LC and/
or LC-MS: a stable isotope-labeled analog of the drug,
and a structural analog (or homolog) of the drug.
Although the best IS is a stable isotope-labeled version
of the analyte, disadvantages exist, including the neces-
sity for custom synthesis, high cost, lack of availability,
and lack of chromatographic resolution of the isotopic
IS and its parent drug. On the other hand, the advan-
tages of a structural analog IS are low cost and high
availability. The structural analog IS compound should
match the chromatographic retention, recovery and
ionization properties of the four -blockers. Pindolol,
a structural analog of -blockers, was found to fulll
these criteria sufciently. Moreover, the coexistence of
pindolol in an actual sample may be rare, because the
compound is not popular in clinical practice in Japan.
Hence, pindolol was chosen as the IS in the quantita-
tive analysis of the four -blockers in this study.
Biomed. Chromatogr. 22: 702711 (2008)
DOI: 10.1002/bmc
Simultaneous determination of -blockers in human plasma ORIGINAL RESEARCH 707

The recoveries of acebutolol, labetalol, metoprolol,

Figure 3. SRM chromatograms for acebutolol, labetalol,
metoprolol, propranolol and IS in human plasma. The
amount of each drug spiked into 1 mL of plasma was 20 ng.
Figure 3 shows chromatograms with SRM proles
obtained from human plasma in the presence of test
compounds (20 ng each compound). Distinct peaks
appeared for all compounds, and the retention times
of labetalol, metoprolol, acebutolol, propranolol and
IS were 6.6, 6.9, 7.2, 7.8 and 7.9 min, respectively.
propranolol and IS from QC samples were determined
at concentrations of 50, 100 and 1000 ng/mL in tripli-
cate. The results are shown in Table 2. Recovery of all
compounds was in the range of 73.5 89.9%, and was
considered satisfactory for our protocol. These values
were superior to those obtained in human serum using
LLE and LC-MS-MS (Keshi-Rahkonen et al., 2007),
and comparable to those obtained in human plasma
using off-line SPE and LC techniques, such as LC-
ultraviolet detection (Musch et al., 1989), LC-uorometric
detection (Albers et al., 2005), LC-MS (Mauer et al.,
2004) and LC-MS-MS (Naidong et al., 2007; Li et al.,
2007).
One signicant drawback of electrospray mass spectro-
metry is that the ionization source is highly susceptible
to co-eluting matrix component (King et al., 2000).
This matrix effect typically results in the suppression
or, less frequently, the enhancement of the analyte
signal. Furthermore, suppression or enhancement of the
analyte response is accompanied by diminished preci-
sion of subsequent measurements (Matuszewski et al.,
2003). The matrix effect of plasma for acebutolol,
labetalol, metoprolol, propranolol and IS in this study
showed 11.529.8% suppression (Table 3). However,
this matrix effect did not cause quantication bias as
evidenced by the CV values (3.7 13.8%); this variabil-
ity is considered acceptable for method validation based
on current criteria (Viswanathan et al., 2007). There-
fore, attempts to further improve the matrix effect were
not pursued.
Table 4 shows the regression equations and the LOD
of the present method for the four -blockers, which
were investigated using spiked plasma samples. Regres-
sion equations of the drugs showed good linearity
in the range 101000 ng/mL for acebutolol, labetalol
and metoprolol and 10 800 ng/mL for propranolol.

Table 2. Recoveries of the four -blockers and IS from QC samples using the present method

Nominal concentration Found concentration Recovery

Compound (ng/mL) (ng/mL) (%)
Acebutolol 50 37.2 1.12a 74.4
100 89.8 7.24 89.9
1000 879 62.8 88.0
Labetalol 50 43.9 2.94 87.9
100 87.9 3.75 88.0
1000 813 47.3 81.3
Metoprolol 50 36.7 1.60 73.5
100 77.0 12.3 77.1
1000 834 105 83.4
Propranolol 50 41.7 1.17 83.6
100 85.8 2.82 85.8
800 643 33.7 80.4
Pindolol (IS) 50 41.7 1.17 84.6
100 77.2 3.48 77.2
1000 817 40.8 81.7
a
Data are given as means SD of three experiments.

Copyright 2008 John Wiley & Sons, Ltd. Biomed. Chromatogr. 22: 702711 (2008)
DOI: 10.1002/bmc
708 ORIGINAL RESEARCH H. Umezawa et al.

Table 3. Matrix effect of the four -blockers and IS in ve individual lots of human plasma

Nominal Found Matrix

concentration concentration effect CV
Compound (ng/mL) (ng/mL) (%) Type of effect (%)
Acebutolol 5 4.0 0.24a 80.2 4.8a 19.8% suppression 6.0
50 35.1 3.35 70.2 6.7 29.8% suppression 9.7
1000 793 25.6 79.3 2.6 20.7% suppression 3.7
Labetalol 5 4.4 0.45 88.5 8.9 11.5% suppression 11.9
50 37.8 4.93 75.7 9.9 24.3% suppression 13.8
1000 864 30.7 86.4 3.1 13.6% suppression 4.0
Metoprolol 5 4.3 0.28 85.5 5.7 14.5% suppression 7.5
50 38.5 4.19 77.0 8.4 23.0% suppression 10.4
1000 792 54.4 79.2 5.4 20.8% suppression 6.9
Propranolol 5 3.8 0.30 76.9 6.0 23.1% suppression 7.2
50 38.3 5.04 76.5 10.1 23.5% suppression 13.1
800 657 23.1 82.1 2.9 17.9% suppression 4.1
Pindolol (IS) 5 4.3 0.28 71.6 3.8 28.4% suppression 5.4
50 37.1 3.90 74.2 7.8 25.8% suppression 10.5
1000 808 53.2 80.8 5.3 19.2% suppression 7.5
a
Mean value SD.

Table 4. Regression equations and LOD for the four -blockers from human plasma samples using the present method

Correlation Concentration LOD

Compound Equationa coefcient (r) range (ng/mL) (ng/mL)
Acebutolol y = 0.0007x 0.0020 0.9992 101000 1
Labetalol y = 0.0006x + 0.0056 0.9990 101000 1
Metoprolol y = 0.0001x + 0.0004 0.9991 101000 1
Propranolol y = 0.0017x + 0.0076 0.9993 10800 3
a
The data were subjected to linear regression analysis of the peak area ratio (y) of analytes/IS against the spiking analyte concentrations (ng/
mL) (x). Eight plots (each point represents the mean of duplicate determinations) with different concentrations for each compound were used
to obtain the equations.

The LOD of all drugs under optimal conditions was
1 ng/mL, with the exception of propranolol (3 ng/mL).
The LOQ, which corresponds to the lowest level of the
concentration range, was 10 ng/mL in plasma, for all
drugs. As a comparison, the best previously reported
LOQ values in human plasma samples were 10 ng/mL
for acebutolol and labetalol using SPE-LC-MS (Mauer
et al., 2004), 5 and 4.9 ng/mL for metoprolol using
LLE-LC-MS-MS (Li et al., 2007) and SPE-LC-MS-MS
(Naidong et al., 2002), respectively, and 2 ng/mL for
propranolol using LLE-LC-MS-MS (Li et al., 2007).
Therapeutic blood levels of the drugs were reported to
be 90 250 ng/mL for labetalol (0.5 h after a 100 mg oral
dose; Moffat et al., 2004), 30284 ng/mL for metoprolol
(2.5 h after a 100 mg oral dose; Kirch et al., 1983)
and 49 ng/mL for propranolol (2.3 h after an 80 mg oral
dose; Karol et al., 2000), while the toxic blood levels of
these drugs have been reported to be 500 1000 ng/mL
(Schulz and Schmoldt, 2003). For acebutolol, therapeu-
tic and postmortem blood concentrations were 1587 ng/
mL (3.6 h after a 400 mg oral dose; Kendall et al., 1984)
and 13149 g/mL (Tracqui et al., 1992; Love, 2000),
respectively. Therefore, therapeutic and toxic blood
levels of acebutolol, labetalol, metoprolol and pro-
pranolol can be determined using the present method.
Intra- and inter-day CVs and accuracy were evaluated
by assessing QC samples prepared from human plasma,
and are summarized in Table 5. Intra-day CVs at all con-
centrations examined were <2.9% for acebutolol, <5.5%
for labetalol, <4.7% for metoprolol and < 7.4% for pro-
pranolol, whereas inter-day CVs at all concentrations
examined were <10.9% for the four drugs. Accuracy
was in the range of 89.4 120% for all concentrations.
These CV values and accuracy are close to those found
for acebutolol, labetalol, metoprolol and propranolol
using SPE-LC-MS (Mauer et al., 2004) and for metoprolol
and propranolol using LLE- and SPE-LC-MS-MS
(Naidong et al., 2002; Li et al., 2007). Thus, the data
indicated that the method was suitable for the quanti-
cation of -blocker levels in human plasma.
Stability
The stock standard solutions containing 1 mg/mL of
acebutolol, labetalol, metoprolol, propranolol and IS
prepared in methanol were stable for at least 3 months

Copyright 2008 John Wiley & Sons, Ltd. Biomed. Chromatogr. 22: 702711 (2008)
DOI: 10.1002/bmc
Simultaneous determination of -blockers in human plasma ORIGINAL RESEARCH 709

Table 5. Precision and accuracy for the four -blockers in QC samples measured using the present method

Intra-day (n = 5) Inter-day (n = 5)
Nominal Found Found
concentration concentration CV Accuracy concentration CV Accuracy
Compound (ng/mL) (ng/mL) (%) (%) (ng/mL) (%) (%)
Acebutolol 10 11.2 0.25a 2.3 111 10.4 0.49a 4.7 104
100 105 2.40 2.3 105 98.0 5.50 5.6 98.0
500 525 15.2 2.9 105 484 26.6 5.5 96.8
1000 941 19.9 2.1 94.0 957 46.8 4.9 95.7
Labetalol 10 8.9 0.49 5.5 89.4 9.3 0.32 3.4 92.7
100 104 2.44 2.3 104 101 1.67 1.6 101
500 527 16.4 3.1 105 507 11.7 2.3 101
1000 1067 21.3 2.0 106 1022 35.9 3.5 102
Metoprolol 10 12.1 0.45 3.8 120 11.0 1.20 10.9 109
100 106 4.98 4.7 106 100 4.09 4.1 100
500 472 19.7 4.2 94.5 499 10.4 2.1 99.9
1000 994 45.6 4.6 99.5 1008 15.1 1.5 100
Propranolol 10 9.6 0.70 7.3 95.9 9.0 0.64 7.1 90.0
100 90.4 1.0 1.1 90.4 93.2 7.81 8.4 93.2
500 456 6.02 1.3 91.4 471 16.5 3.5 94.4
800 782 57.6 7.4 97.8 785 27.2 3.5 98.2
a
Values are means SD.

when kept at 4C. The working standard solutions of
the four drugs and IS were also investigated over a
period of 2 weeks at 4C, and no signicant change was
observed.
The stability of acebutolol, labetalol, metoprolol and
propranolol in plasma was tested using QC samples
under various conditions, including storage at 4C for
up to 48 h and storage at 80C for up to 8 weeks.
Prior to analyses, QC samples were brought to room
temperature and vortex-mixed thoroughly. The stabilities
of the analytes in QC samples were expressed as the
percentage remaining of the initial values which were
determined immediately after sample preparation. The
analytes were considered stable in plasma when 85
115% of the initial amount remains (van den Broek
et al., 2006). The results of stability assays for the four
-blockers in QC samples are shown in Table 6. The
drugs in plasma were found to be stable for at least

Table 6. Stability of the four -blockers in QC samples at 4 and 80C

Nominal
concentration CV
Compound (ng/mL) Stability (%)a Mean SD (%)
Storage at 4C
6h 12 h 24 h 48 h
Acebutolol 20 95.7b 97.7 98.2 93.8 96.3 2.00 2.08
1000 100 98.4 95.1 97.6 97.9 2.35 2.40
Labetalol 20 103 98.5 93.8 94.6 97.6 4.56 4.67
1000 107 98.8 97.7 98.3 100 4.70 4.67
Metoprolol 20 99.8 94.9 93.3 92.6 95.1 3.26 3.42
1000 106 101 102 100 102 2.35 2.29
Propranolol 20 101 104 99.6 93.5 102 2.08 2.03
800 104 100 104 103 103 2.03 1.96
Storage at 80C
1 week 2 weeks 4 weeks 8 weeks
Acebutolol 20 101b 107 110 109 107 3.99 3.72
1000 88.3 93.3 96.8 94.1 93.1 3.56 3.81
Labetalol 20 95.3 97.6 99.7 96.1 97.2 1.93 1.98
1000 89.2 96.9 106 107 99.9 8.54 8.54
Metoprolol 20 103 105 106 105 105 1.01 0.96
1000 97.6 96.6 98.7 96.8 97.4 0.94 0.97
Propranolol 20 93.9 84.2 90.1 88.2 89.1 4.04 4.53
800 94.2 91.3 95.9 98.7 95.0 3.10 3.26
a
Each freshly prepared QC sample was analyzed before the beginning of storage and was set at 100%.
b
Each value represents the means of duplicate experiments.

Copyright 2008 John Wiley & Sons, Ltd. Biomed. Chromatogr. 22: 702711 (2008)
DOI: 10.1002/bmc
710 ORIGINAL RESEARCH
48 h at 4C (95.1103% of the nominal concentration;
1.96 4.67% CV) and for at least 8 weeks at 80C
(93.1107% of the nominal concentration; 0.968.54%
CV). The MSpak GF column showed excellent stability
in this system; one column could be repeatedly used for
at least 500 injections with good reproducibility.
CONCLUSION
We have established a detailed procedure for the simple
determination of four -blockers, acebutolol, labetalol,
metoprolol and propranolol, in human plasma by direct-
injection LC-MS-MS method using an MSpak GF
column. The levels of these four drugs in plasma can be
determined by SRM of LC-MS-MS without tedious
pretreatments. The present method is expected to be
very useful in therapeutic drug monitoring, clinical
toxicology and forensic toxicology.
REFERENCES
Albers S, Elshoff JP, Vlker C, Richter A and Ler S. HPLC quanti-
cation of metoprolol with solid-phase extraction for the drug
monitoring of pediatric patients. Biomedical Chromatography 2005;
19: 202207.
Amendola L, Molaioni F and Botr F. Detection of beta-blockers in
human urine by GC-MS-MS-EI: perspectives for the antidoping
control. Journal of Pharmaceutical and Biomedical Analysis 2000;
23: 211221.
Arinobu T, Hattori H, Seno H, Ishii A and Suzuki O. Comparison of
SSI with APCI as an interface of HPLC-mass spectrometry for
analysis of a drug and its metabolites. American Society for Mass
Spectrometry 2001; 13: 204 208.
Black SB, Stenhouse AM and Hansson RC. Solid-phase extraction
and derivatisation methods for beta-blockers in human post mortem
whole blood, urine and equine urine. Journal of Chromatography
B 1996; 685: 67 80.
Ding L, Zhou X, Guo X, Song Q, He J and Xu G. LE-ESI-MS
method for the determination of bisoprolol in human plasma. Journal
of Pharmaceutical and Biomedical Analysis 2007; 44: 520 525.
do Carmo Borges NC, Mendes GD, de Oliveira Silva D, Rezende
VM, Barrientos-Astigarraga RE and De Nucci G. Quantication of
carvedilol in human plasma by high-performance liquid chromato-
graphy coupled to electrospray tandem mass spectrometry applica-
tion to bioequivalence study. Journal of Chromatography B 2005;
822: 253 262.
Fanali S, Aturki Z, DOrazio G and Rocco A. Separation of basic com-
pounds of pharmaceutical interest by using nano-liquid chromato-
graphy coupled with mass spectrometry. Journal of Chromatography
B 2007; 1150: 252258.
Fujimaki K, Lee XP, Kumazawa T, Sato J and Sato K. Determina-
tion of some antiallergic drugs in human plasma by direct-injection
high-performance liquid chromatographytandem mass spectrometry.
Forensic Toxicology 2006; 24: 816.
Hartonen K and Riekkola ML. Detection of beta-blockers in urine
by solid-phase extraction-supercritical uid extraction and gas
chromatographymass spectrometry. Journal of Chromatography B
1996; 676: 4552.
Johnson RD and Lewis RJ. Quantitation of atenolol, metoprolol,
and propranolol in postmortem human uid and tissue specimens
via LC/APCI-MS. Forensic Science International 2006; 156: 106117.
Johnsson G and Regrdh CG. Clinical pharmacokinetics of beta-
adrenoreceptor blocking drugs. Clinical Pharmacokinetics 1976; 1:
233263.
Copyright 2008 John Wiley & Sons, Ltd.
H. Umezawa et al.
Josefsson M and Sabanovic A. Sample preparation on polymeric
solid phase extraction sorbents for liquid chromatographictandem
mass spectrometric analysis of human whole blooda study on a
number of beta-agonists and beta-antagonists. Journal of Chromato-
graphy A 2006; 1120: 112.
Karol MD, Locke CS and Cavanaugh JH. Lack of interaction
between lansoprazole and propranolol, a pharmacokinetic and
safety assessment. Journal of Clinical Pharmacology 2000; 40: 301
308.
Kendall MJ, Jack DB, Quarterman CP, Smith SR and Zaman R. -
adrenoceptor blocker pharmacokinetics and the oral contraceptive
pill. British Journal of Clinical Pharmacology 1984; 17: 87S89S.
Keski-Rahkonen P, Prssinen O, Leppnen E, Mauriala T, Lehtonen
M and Auriola S. Determination of tamsulosin in human aqueous
humor and serum by liquid chromatographyelectrospray ioniza-
tion tandem mass spectrometry. Journal of Pharmaceutical and
Biomedical Analysis 2007; 43: 606 612.
King R, Bonglio R and Fernandez-Metzler C. Mechanistic investiga-
tion of ionization suppression in electrospray ionization. Journal of
American Society for Mass Spectrometry 2000; 11: 942950.
Kirch W, Spahn H, Ohnhaus EE, Khler H, Heinz U and Mutschler E.
Inuence of inammatory disease on the clinical pharmacokinetics of
atenolol and metoprolol. Biopharmaceutics and Drug Disposition
1983; 4: 7381.
Lee XP, Kumazawa T, Sato J, Shoji Y, Hasegawa C, Caribe C,
Arinobu T, Seno H and Sato K. Simple method for determination
of benzodiazepines in human body uids by high-performance
liquid chromatography/mass spectrometry. Analytica Chimica Acta
2003; 492: 223231.
Lee XP, Kumazawa T, Fujishiro M, Hasegawa C, Marumo A, Shoji
Y, Arinobu T, Seno H and Sato K. Simple method for determina-
tion of triazolam in human plasma by high-performance liquid
chromatography/tandem mass spectrometry. Journal of Pharmaceu-
tical and Biomedical Analysis 2006; 41: 6469.
Li S, Liu G, Jia J, Liu Y, Pan C, Yu C, Cai Y and Ren J. Simultane-
ous determination of ten antiarrhythic drugs and a metabolite in
human plasma by liquid chromatographytandem mass spectro-
metry. Journal of Chromatography B 2007; 847: 174 181.
Love JN. Acebutolol overdose resulting in fatalities. Journal of Emergency
Medicine 2000; 18: 341344.
Matuszewski BK, Constanzer ML and Chavez-Eng CM. Strategies
for the assessment of matrix effect in quantitative bioanalytical
methods based on HPLC-MS/MS. Analytical Chemistry 2003; 75:
3019 3030.
Maurer HH, Tenberken O, Kratzsch C, Weber AA and Peters FT.
Screening for library-assisted identication and fully validated
quantication of 22 beta-blockers in blood plasma by liquid
chromatography-mass spectrometry with atmospheric pressure
chemical ionization. Journal of Chromatography A 2004; 1058: 169
181.
Moffat A, Osselton MD and Widdop B. Clarkes Analysis of Drugs
and Poisons. Pharmaceutical Press: London, 2004; 11591160.
Musch G, Buelens Y and Massart DL. A strategy for the determination
of beta blockers in plasma using solid-phase extraction in combina-
tion with high-performance liquid chromatography. Journal of
Pharmaceutical and Biomedical Analysis 1989; 7: 483 497.
Naidong W, Shou WZ, Addison T, Maleki S and Jiang X. Liquid
chromatography/tandem mass spectrometric bioanalysis using
normal-phase columns with aqueous/organic mobile phase-a novel
approach of eliminating evaporation and reconstitution steps in
96-well SPE. Rapid Communications in Mass Spectrometry 2002;
16: 19651975.
Prichard BNC. -adrenergic receptor blockade in hypertension, past,
present and future. British Journal of Clinical Pharmacology 1978;
5: 379399.
Quaglio MP, Bellini AM, Minozzi L, Frisina G and Testoni F. Simul-
taneous determination of propranolol or metoprolol in the pres-
ence of butyrophenones in human plasma by gas chromatography
with mass spectrometry. Journal of Pharmaceutical Sciences 1993;
82: 8790.
Reith DM, Dawson AH, Epid D, Whyte IM, Buckley NA and Sayer
GP. Relative toxicity of beta blockers in overdose. Clinical Toxico-
logy 1996; 34: 273278.
Schulz M and Schmoldt A. Therapeutic and toxic blood concentrations
Biomed. Chromatogr. 22: 702711 (2008)
DOI: 10.1002/bmc
Simultaneous determination of -blockers in human plasma
of more than 800 drugs and other xenobiotics. Pharmazie 2003; 58:
447 474.
Tracqui A, Kintz P, Wendling P, Ritter-Lohner S, Mangin P and
Jaeger A. Toxicological ndings in a fatal case of acebutolol
self-poisoning. Journal of Analytical Toxicology 1992; 16: 398 400.
Unverir P, Topacoglu H, Bozkurt S and Kaynak F. Cardiovascular
toxicity due to metoprolol poisoning in a patient with coronary
artery disease. British Jounal of Clinical Pharmacology 2007; 64:
694697.
van den Broek I, Sparidans RW, Huitema ADR, Schellens JHM and
Copyright 2008 John Wiley & Sons, Ltd.
ORIGINAL RESEARCH 711
Beijnen JH. Development and validation of a quantitative assay
for the measurement of two HIV-fusion inhibitors, enfuvirtide
and tifuvirtide, and one metabolite of enfuvirtide(M-20) in human
plasma by liquid chromatographytandem mass spectrometry.
Journal of Chromatography B 2006; 837: 49 58.
Viswanathan CT, Bansal S, Booth B, DeStefano AJ, Rose MJ,
Sailstad J, Shah VP, Skelly JP, Swann PG and Weiner R. Workshop/
conference reportquantitative bioanalytical methods validation
and implementation: Best practices for chromatographic and ligand
binding assays. The AASP Journal 2007; 9: E30E42.
Biomed. Chromatogr. 22: 702711 (2008)
DOI: 10.1002/bmc