International Research Journal of Pharmacy and Pharmacology (ISSN 2251-0176) Vol. 2(12) pp.

299-305, December 2012

Available online http://www.interesjournals.org/IRJPP
Copyright © 2012 International Research Journals

Full length Research Paper

Determination of amoxicillin in human plasma using

cephalexin as an internal standard and HPLC-MS/MS
for the application of bioequivalence studies
*Claudio R. F. Cardoso1,2, Maria Isabel R. Alves1 and Nelson R. Antoniosi Filho1
1
LAMES (Laboratory of Extraction and Separation Methods), Institute of Chemistry, Federal University of Goiás, Campus
Samambaia, C. P. 131, CEP 74001-970 – Goiania – GO - Brazil
2
Pharmacokinetics Integrated Center of the Federal University of Goiás
Accepted November 13, 2012

A method for determining amoxicillin in human plasma by ESI-LC-MS/MS and using cephalexin as an
internal standard was validated. The chromatographic separation was performed in a C18 Shim-pack
CLC-ODS(M) analytical column (150 mm x 4.6 mm i.d. x 5 µm), which provided satisfactory separation of
amoxicillin and cephalexin peaks, with analysis time of 3.1 minutes. A yield of extraction of 96.6% for
amoxicillin and 98.7% for cephalexin was obtained, limit of quantification of 90 ng mL-1, limit of
detection of 500 pg mL-1, and acceptable linearity for the analytical curve and calibration. Hence, the
results show that the present method is appropriate for application in bioequivalence studies of
amoxicillin, and it can also act as basis for the study of cephalexin.

Keywords: Amoxicillin, cephalexin, LC-MS/MS, SPE, bioequivalence.

INTRODUCTION

Amoxicillin (2S,5R,6R)-6-[(R)-(−)-2-amino-2- (p- Several methods have been developed and validated
hydroxyphenyl) acetamido]-3,3-dimethyl-7-oxo-4-thia-1- in order to determine amoxicillin in human plasma.
azabicyclo [3. 2. 0] heptane-2-carboxylic acid trihydrate) Tavakoli et al. (2007) obtained limits of quantification of
-1
(C16H19N3O5S) is an antibiotic of the beta-lactam group, 0.15 µg mL using HPLC and UV detector. Wen et al.
characterized as a broad spectrum penicillin (Goodman (2008) determined simultaneously amoxicillin, ambroxol,
et al., 2007). It may be administered in clinical settings of and clenbuterol as an internal standard using LC-MS/MS
pneumonia (Kabra et al., 2010), urinary tract infections and the plasma samples were subjected to a simple
(Schulman, 1981), otitis (Coles, 1993), among others. protein precipitation with methanol, reaching a limit of
The molecular mass of amoxicillin is 365.10 g mol-1. It detection for amoxicillin of 5 ng mL-1. Pei et al. (2011)
presents characteristics of white powder with mild determined simultaneously amoxicillin, sulbactam, and
aromatic odor and bitter taste, and it is found either in cefadroxil in plasma as an internal standard for HLPC-
anhydrous or in trihydrate preparations. Sodium salt is DAD using acetronile with precipitation of proteins. Limits
used for parenteral preparations. Amoxicillin (Figure 1) is of detection of 0.163 and 0.250 µg mL-1 for amoxicillin
directly obtained from acid 6-aminopenicillanic through and sulbactam were obtained, respectively. However,
chemical or semi-synthetic synthesis (Silva, 1998; there are no reports for the determination of amoxicillin in
Alekseev et al., 2005). The broad commercialization of human plasma using cephalexin as an internal standard.
this active ingredient makes it important the development Thus, the object of this paper was to develop an
of new methodologies for the determination of its analytical methodology for determining amoxicillin in
pharmacokinetics. human plasma using the solid phase extraction (SPE)
technique and high performance liquid chromatography
coupled to mass spectrometry (HPLC-MS/MS) in order to
develop a fast, highly specific and repeatable
methodology for use in pharmaceutical bioequivalence
*Corresponding Author E-mail: isaribeiroalves@yahoo.com.br studies.
300 Int. Res. J. Pharm. Pharmacol.

Figure 1. Chemical structure of amoxicillin.

Experimental monitored mass (Dwell) of 0.5 seconds and reading time

between the end of a screening and the beginning of the
Chemicals and reagents next one (Delay) of 0.1 seconds.
The quasi-molecular ion (precursor ion) for amoxicillin
All the processes used ultrapure water generated by the was 366.1 Da and for cephalexin, 348.0 Da. The
ultra-purification system Mille-Q® Gradient A10, by monitored fragment (product ion) for amoxicillin was
Millipore. The HPLC-grade reagents used in the present 114.0 Da and for cephalexin, 157.9 Da.
study were purchased by Merck®. Reference standards
were used upon authentication of origin and indication of
a ‘Certificate of Analysis (CA)’. The primary standards for Preparation of stock and standard solutions
amoxicillin and cephalexin were obtained at United
Stated Pharmacopeia – USP. A stock solution of 1.00 mg mL-1 amoxicillin in
water:acetonitrile solution (50:50 v/v) was prepared. The
amoxicillin stock solution was diluted for the preparation
Instrumentation and Chromatographic Conditions of standard solutions in the concentrations of 1.80; 5.00;
10.0; 25.0; 40.0; 60.0; 75.0; and 100 µg mL-1. A work
The samples were analyzed in a HPLC SHIMADZU AD solution for the cephalexin as an internal standard in the
Vp chromatograph (Shimadzu, Japan), equipped with a concentration of 20 µg mL-1 was prepared from the stock
degasser and autosampler, coupled with a triple solution of 1.00 mg mL-1 in water:acetonitrile (50:50 v/v).
quadrupole (Micromass, UK) Quattro LC mass All solutions were packed in polypropylene tubes with
spectrometer (MS-MS), equipped with ion sources Z- screw cap and stored at -20 °C.
spray type. The chromatographic analysis were
performed using a C18 Shim-pack CLC-ODS(M) (150 mm
x 4.6 mm i.d. x 5 µm) analytical column (Shimadzu, Sample preparation
Japan) with an outflow of 700 µL min-1 for the mobile
phase, comprised of a mixture of aqueous formic acid The samples of human plasma were defrosted at room
solution 0.1% and acetonitrile (77:23 v/v), with a post- temperature, homogenized in vortex, and centrifuged at
column split of 1:3 proportion. The system pressure 3400 rpm for 4 minutes to precipitate the suspended
operated between 85 to 95 bar and the column and materials. As for the extraction, it used solid-phase (SPE)
sampler were kept at a controlled temperature of 22°C. Waters Oasis HLB extraction cartridges of 1cc and 30
Injection volume was 20 µL. mg of adsorbent previously stored with 1 mL of methanol
Quattro LC (MS-MS) mass spectrometer operated on and 1 mL of ultrapure water. The amount of 250.00 µL of
positive electrospray ionization (ESI), operating in 2- plasma was pipette in the cartridges, and then, adding
channel MRM mode. As for the electrospray ionization, 5.00 µL from the cephalexin 20.00 µg mL solution,
-1

the following parameters were used: capillary in 3.00 kV; followed by the addition of 450.00 µL of ultrapure water.
cone in 16.0 V for amoxicillin and in 17.0 V for The SPE cartridges were centrifuged at 3400 rpm for 4
cephalexin; extractor in 2.00V; RF lenses in 0.1V; source minutes for the elution of plasma in the test-tube and
temperature of 110 oC; desolvation temperature of 250°C; retention of amoxicillin in the cartridge. The elution of
cone gas outflow of 0.0 L/h; desolvation gas outflow of amoxicillin and the internal standard were performed by
530 L h-1. As for the quadrupoles of Quattro LC mass adding 950.00 µL of extraction solvent acetonitrile:water
spectrometer, it was established the following conditions: (50:50 v/v) and, afterwards, the cartridges and their
resolution in LM 1 of 14.8; resolution in HM 1 of 14.8; ion respective test-tubes were centrifuged once again for two
1 energy of 1.0 V; second hexapole input of 2.0 V; minutes, at 3400 rpm, for the complete elution of the
collision energy for amoxicillin of 21.0 eV; collision energy extraction solvent. Then, 50 µL from the aqueous 2%
for cephalexin of 10.0 eV; output of 12 V; resolution in LM formic acid:acetonitrile (50:50 v/v) was added to the
2 of 12.0; resolution in HM 2 of 12.0; ion 2 energy of 1.0 eluted solution, and the test-tubes were shaken in vortex
V; multiplier of 650 V; reading time for each of the
 Cardoso et al. 301

for 5 seconds. After that, the analysis was conducted acceptance criteria were deviation in nominal
through LC-MS/MS. concentration equal to or lesser than 15%, accuracy and
precision equal to or lesser than 15% (Anvisa RE n°899-
2003).
Linearity The long-term stability of the standard solution of the
drug and internal standard in the biological fluid, at room
The analytical curve was prepared at concentration levels temperature, was assessed after six hours from
of 0.0900; 0.250; 0.500; 1.25; 2.00; 3.00; 3.75; and 5.00 preparation. Such solutions were frozen at -20ºC for
µg mL-1. Each SPE cartridge received 250 µL of human seven days, before analyses were performed.
plasma, 50 µL of work solution related to each level of the
analytical curve, and 50 µL of cephalexin as the internal
standard solution. The solutions have been distributed to Biological Sample
a final volume of 1.00 mL with 0.1% formic
acid:acetonitrile (77:23 v/v). After that, 400 µL of ultrapure Samples for bioequivalence tests were collected at the
water was added in each cartridge and they were Cardiology Institute of Anápolis, in the state of Goiás,
submitted to centrifugation. The sample quality controls Brazil. The study was comprised of 24 volunteers, and
(QC) were prepared in concentrations of 0.090 µg mL-1 performed after approval by the Research Ethics
(LQ); 0.250 µg mL-1 (CQB); 2.00 µg mL-1 (CQM); and Committee of the Federal University of Goiás.
3.75 µg mL-1 (CQA) through the fortification of blank The blood samples were collected in a total volume of
plasma and subsequent extraction process. 8 mL. The blood was centrifuged and the plasma was
removed. Plasma samples were frozen at -20ºC. Total
sample volumes were 720, 1080, and 1440 ml samples
Recovery of extraction for two, three, and four hospitalizations of volunteers,
respectively, both for the study of amoxicillin in pills and
The recovery of the extraction method was determined by oral suspension. Time for collecting samples was: 0.25;
the comparison of the areas obtained for amoxicillin peak 0.50; 0.75; 1.00; 1.25; 1.50; 2.00; 2.50; 3.00; 4.00; 5.00;
related to the samples that were not submitted to the 6.00; 8.00; and 12.00 hours.
extraction process, considering yield of extraction of
100%, and to the samples submitted to the extraction
process, in concentrations of 0.090; 0.250; 0.500; 1.250; RESULTS AND DISCUSSIONS
2.00; 3.00; 3.75; 5.00 µg mL-1. The internal standard
showed a yield of extraction in replicates in the Analyses using HPLC-MS/MS
concentration of 1.00 µg mL-1.
The chromatographic methodology applied in the present
study allowed the amoxicillin and internal standard
Precision and accuracy (cephalexin) to elute with a resolution of 1.66, with
analysis time of 3.1 minutes (Figure 2), which represents
The precision and accuracy of the methodology was satisfactory separation in a reduced period. Pei et al.
determined by the extraction of four samples containing (2011) and Tavakoli et al. (2007) obtained analysis time
cephalexin as the internal standard and amoxicillin in the greater than 8 minutes, and in the study by Wen et al.
concentrations LQ, CQB, CQM, and CQA, analyzed in (2008) the chromatographic analysis was performed in
replicates, in a procedure performed intraday and 4.2 minutes.
interdays. The criteria for approval for the coefficient of Figure 3 presents the mass spectrum obtained for the
variation (CV) and relative standard deviation (RSD) were standards of amoxicillin and cephalexin. The spectrum
15% (Anvisa RE n°899-2003). obtained in the positive electrospray mode, for the
establishment of quasi-molecular ion (precursor ion) and
fragments (product ion) of amoxicillin and cephalexin,
Stability were used to determine the ions monitored in MRM, that
is, during monitoring of the ionized molecule and pre-
The studies of amoxicillin and cephalexin stability in established fragments, making the analysis even more
plasma were performed under three conditions. First, it sensitive (Straub and Voyksner, 1993). The amoxicillin
assessed stability during three 24-hour freezing cycles at was protonated and analyzed in the first mass spectrum,
-20°C, followed by defrosting. After that, it assessed the in which the most abundant ion was 366.1 Da, resulting
stability of amoxicillin and cephalexin kept at room from the addition of a proton to its structure, [M+H]+. The
temperature for seven hours. The long-term stability fragmentation ion (product ion), obtained in the se-
was assessed for a storage period of fifteen days. The cond mass spectrum for amoxicillin, presented a greater
302 Int. Res. J. Pharm. Pharmacol.

Figure 2. Total ion chromatogram for amoxicillin (2.17 min) and cephalexin (2.49 min).

amoxi scan 3 1 (2.065) Scan ES+ amox1 1 (2.065) Daughters of 366ES+

366 3.18e9 100 114 4.89e7
100
H+
H H CH3
H H CH3 S
S HO CHCONH
HO CHCONH CH3
CH3 NH2 N
NH2 N O
O COOH
COOH
F

% % [F -COOH]+ = 114.0 Da
349
m/z= 366.1 Da 367

amoxicillin (A) 134 amoxicillin (B)

368

160 165 190

83 123 407 70
0 m/z 0 m/z
50 100 150 200 250 300 350 400 450 500 50 75 100 125 150 175 200 225 250 275 300 325 350 375 400
cefa scan 1 (2.065) Scan ES+ cefa1 1 (2.065) Daughters of 348ES+
348 1.59e9 158 6.39e7
100 100
H+
H H 106
S 174
CHCONH H H
S
NH2 N CHCONH
O CH3 140
COOH NH2 N
O CH3
COOH

m/z= 348,0 Da F
% %
cephalexin (A) [F + H]+
m/z = 157.9 Da
158
349
114 cephalexin (B)
191 192
174 370 68
106 91
0 m/z 0 m/z
50 100 150 200 250 300 350 400 450 500 50 100 150 200 250 300 350 400 450 500

Figure 3. (A) Full scan mass spectrum; (B) tandem mass spectrum for the amoxicillin standard.

intensity ion of 114.0 Da. This fragment was originated by quasi-molecular ion m/z of 348.0 Da, as well as ionization
+
breaking the β-lactam ring and from the release of COOH as [M+H] , which is typical for ESI. The cephalexin
group (Riediker and Stadler, 2001). fragment ion (product ion) used was the m/z ion of
The cephalexin showed protonation originating one greater intensity (157.9 Da), obtained from the breaking
 Cardoso et al. 303

blank plasma blank plasma

022001plasma01 MRM of 2 Channels ES+ 042001VOL01e02-01 MRM of 2 Channels ES+
100
2.09 366.1 > 114 100 0.03 348 > 157.9
42.8 103
amoxicillin cephalexin

1.52
2.41
0.75
0.25 1.72 2.81
0.49
0.03 0.75
% %
1.60
2.93

Tempo Tempo
(min) (min)
0 0
0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00

Figure 4. Chromatogram of blank plasma with mass spectrum monitoring amoxicillin (366.1 Da and
114.0 Da) and cephalexin (348.0 Da and 157.9 Da).

Table 1. Intraday and interday precision and accuracy for the determination of amoxicillin in plasma.

Concentration (µg mL-1) n=8

Intraday Interdays
Nominal Determined Accuracy (%) RSD (%) Determined RSD (%) Accuracy (%)
0.090 0.0970 99.7 1.78 0.094 4.18 104.5
0.250 0.241 102.6 1.27 0.250 3.30 99.9
2.00 2.007 108.5 1.44 2.081 3.94 104.5
3.75 3.673 101.9 1.96 3.756 1.92 100.2

of the beta-lactam ring and the addition of another proton The limit of quantification (LQ) obtained was 80.36 ng
to the released fragment (Figure 3). mL-1 and the limit of detection (LD) was 500 pg mL-1,
concentration in which signal/noise relation was 5.82.
The relative error for LQ was determined through 8
Validation of Method replicated and three stages of validation, obtaining RSD
ranging from 1.92% to 3.13%. Therefore, precision
Specificity and selectivity values are adequate and provide analytical reliability for
the determination and validation of LQ.
This study evaluated the presence of interfering agents,
as well as amoxicillin and cephalexin, in the samples of
normal blank plasma. Thus, each sample of blank plasma Precision and accuracy
was assessed according to the extraction process and
the chromatographic and spectrometric conditions The results of intraday and interday precision and
proposed. Hence, the blank plasma sample used in the accuracy for determination of amoxicillin in human
present study did not show interfering agents for the plasma are presented in Table 1. The results indicate that
retention time of the drug and the internal standard the proposed methodology is accurate and shows
(Figure 4). intraday and interdays preciseness, considering the
parameters desired for an analytical methodology applied
to the bioequivalence studies of medicines.
Calibration curve

The calibration curve for amoxicillin presented an Recovery

average linear correlation coefficient of 0.9990 (n=3), in
-1
which linearity ranged from 0.090 to 5.00 µg mL . For all The absolute recovery of amoxicillin was evaluated for
points in the calibration curve, the values of relative error eight concentrations with n=6 for each one of them. The
were below 8%, which indicates that the calibration curve recovery obtained was 99.5 ±7.77% for a nominal
-1
obtained provided analytical reliability to the quantification concentration of 0.090 µg mL ; 93.1 ±2.4% for 0.250 µg
-1 -1
process. Hence, the method showed satisfactory linearity mL ; 94.4 ±2.3% for 0.500 µg mL ; 95.4 ±2.8% for 1.250
-1 -1
and a wide linear dynamic range. µg mL ; 98.2 ±2.3% for 2.000 µg mL ; 95.9 ±1.4% for
304 Int. Res. J. Pharm. Pharmacol.

Table 2. Stability of amoxicillin and cephalexin under different conditions (n=8 for
amoxicillin and n=16 for cephalexin).

Conditions Drug Concentration (µg mL-1)

Nominal Determined mean
0.250 0.243 ± 1.73%
amoxicillin
Frost-defrost cycles 3.750 3.767 ± 3.57%
cephalexin 1.000 1.009 ± 2.16%
0.250 0.228 ± 3.24%
amoxicillin
Room temperature for 7 h 3.750 3.571 ± 1.95%
cephalexin 1.000 0.996 ± 4.24%
0.250 0.248 ± 2.68%
amoxicillin
Storage at -20°C for 15 days 3.750 3.762 ± 2.83%
cephalexin 1.000 0.923 ± 3.78%

3.000 µg mL-1; 100.0 ±2.4% for 3.750 µg mL-1; and 96.2 stable in all processes used to assess the stability of
±4.2% for 5.000 µg mL-1. The recovery of internal these drugs in biological matrices, making it possible to
standard was 98.7 ±2.85% and average recovery for analyze such drugs even after long periods of storage at
plasma amoxicillin by SPE was 96.6%, results that 0ºC.
demonstrate the efficacy of this method in bioequivalence Therefore, the proposed methodology is completely
studies. applicable to the bioequivalence and bioavailability
studies for amoxicillin, and it can be the basis for the
application in bioequivalence and bioavailability studies
Stability for cephalexin, using amoxicillin as an internal standard.

Table 2 shows stability data for amoxicillin and

cephalexin under different conditions. Criteria for ACKNOWLEDGEMENTS
acceptance were deviation from nominal concentration
and precision equal to or lesser than 15% (Anvisa RE To MCT, FINEP, FUNAPE and CNPq by the financial
n°899-2003). It was observed that both amoxicillin and support, to CAPES and CNPq by the productivity
cephalexin did not degrade by influence of several fellowship to Nelson Roberto Antoniosi Filho (Process
oscillations in storage temperature, which assures that 309832/2010-1).
the samples may go through several frost-defrost cycles
without loses in analytes concentration. Besides, both
amoxicillin and cephalexin did not degrade in the period REFERENCES
of time that the sample was kept in the equipment to be Alekseev VG, Demskaya EV, Milashs EA, Igolkin VV (2005). Acid-Base
analyzed. The evaluation of long-term stability allowed Properties of Amoxycillin. Russ. J. Gen. Chem, (75)7: 1147-1150.
the observation that the samples remained stable for 15 Anvisa Resolution n. 899 (2003). Guide for Analytical and Bioanalytical
days. Methods. Available at
http://www.anvisa.gov.br/legis/resol/2003/re/899_03re.htm.
Coles SJ (1993). A comparative study of clarithromycin and amoxycillin
suspensions in the treatment of pediatric patients with acute otitis
CONCLUSIONS media. Infection 4: 272-278.
Goodman LS, Gilman A, Brunton LL (2007). As bases farmacológicas
th
da terapêutica, 11 ed., McGraw-Hill, Rio de Janeiro.
The analytical methodology developed presents high Kabra SK, Lodha R, Pandey RM (2010). Antibiotics for community
recovery, satisfactory linearity, precision and accuracy, acquired pneumonia in children (Review). The Cochrane
as well as limit of quantification that are appropriate for Collaboration. John Wiley & Sons Ltd, 110-125.
the verification of plasma concentrations in amoxicillin in Pei Q, Yang GP, Li ZJ, Penga XD, Fan JH, Liu ZQ (2011).
Simultaneous analysis of amoxicillin and sulbactam in human plasma
concentrations of up to five half-lives. Besides, the by HPLC-DAD for assessment of bioequivalence. J. Chromatogr. B
procedure is of quick chromatographic analysis, with 879: 2000– 2004.
acceptable resolution for amoxicillin and cephalexin as Riediker S, Stadler RH (2001) Simultaneous determination of five beta-
lactam antibiotics in bovine milk using liquid chromatography coupled
the internal standard.
with electrospray ionization tandem mass spectrometry. Anal. Chem.
The present methodology is appropriate for the 73: 1614-1621.
evaluation of amoxicillin and cephalexin stability under Schulman CC (1981). A randomised clinical study of bacampicillin and
different conditions. Amoxicillin and cephalexin were amoxycillin in urinary tract infections. Infection (9)1: 25-28.
 Cardoso et al. 305

th
Silva P (1998). Farmacologia, 4 ed., Guanabara Koogan, Rio de simultaneous in vitro determination of amoxicillin and metronidazole
Janeiro. at single wavelength. J. Pharmaceut. Biomed. 43: 325–329.
Straub RF, Voyksner RD (1993). Determination of penicillin G, Wen A, Hang T, Chena S, Wang Z, Ding L, Tian Y, Zhang M, Xu X
ampicillin, amoxicillin, cloxacillin and cephapirin by high-performance (2008). Simultaneous determination of amoxicillin and ambroxol in
liquid chromatography-electrospray mass spectrometry. J. human plasma by LC–MS/MS: Validation and application to
Chromatogr. 647: 167-181. pharmacokinetic study. J. Pharm. Biomed. 48: 829–834.
Tavakoli N, Varshosaz J, Dorkoosh F, Zargarzadeh MR (2007).
Development and validation of a simple HPLC method for