Special issue: review

Received: 11 January 2013, Accepted: 3 February 2013 Published online in Wiley Online Library

(wileyonlinelibrary.com) DOI 10.1002/bmc.2898

A concise review of HPLC, LC-MS and LC-MS/

MS methods for determination of azithromycin
in various biological matrices
Kuldeep Sharma and Ramesh Mullangi*
ABSTRACT: Azithromycin is one of the best selling antibiotics in the world. It belongs to the new macrolide family of azalides.
It is derived from erythromycin and it differs from erythromycin in having a 15-membered ring and a methyl substituted
nitrogen inserted at the 9a position in the aglycone ring. This structural modification confers favourable microbiological
and pharmacokinetic characteristics on azithromycin and greater acid stability compared with erythromycin. It is mainly used
to treat respiratory infections, sexually transmitted diseases, cutaneous and soft-tissue infections, etc. This review provides a
comprehensive overview of various HPLC, LC-MS and LC-MS/MS methods for quantitation of azithromycin in different
biological matrices. In addition, it provides general information on extraction steps, internal standard selection, conditions
for chromatographic separation, brief validation data and applicable conclusions for reported methods in a defined pattern.
Copyright © 2013 John Wiley & Sons, Ltd.

Keywords: azithromycin; HPLC; LC-MS/MS; bioanalytical methods; review

Introduction dose, is reached within 2–3 h (Tmax). It attains Cmax of 0.4–0.45

mg/L at 2.5 h post-dosing of a 500 mg oral dose. It is highly
Azithromycin (Fig. 1, CAS no. 83905-01-5) is the first prototype 15- concentrated in white blood cells and macrophages, undergoes
membered azalide, a subclass of macrolide antibiotics. Azithromycin extensive tissue distribution and achieves high concentrations,
(9-deoxo-9a-aza-9a-methyl-9a-homo-erythromycin) is derived struc- especially in pulmonary and tonsillar tissues. Owing to its high
turally from erythromycin (Fig. 1, CAS no. 114-07-8) by replacing the concentration in phagocytes, azithromycin is actively transported
9a carbonyl in the aglycone ring with a methyl-substituted nitrogen, to the site of infection and, during active phagocytosis, high
as well as expanding the ring to 15 membered (Bright et al., 1988). concentrations of azithromycin are released. It has a high tissue-
This modification confers greater acid stability for azithromycin than to-blood concentration ratio (~50 times higher) with a half-life
erythromycin. In pH 2 solution at 37
C, azithromycin had degraded (t1/2) of 2–4 days in most tissues, which is one of the factors in its
by only 10% after 20 min, whereas erythromycin degraded to the stellar antibiotic performance (Foulds et al., 1990; Shepard and
same extent by intramolecular dehydration after 3.7 s. This modifica- Falkner, 1990). The long half-life is due to its lipophilic nature
tion also expands the spectrum of activity and improves tissue and high volume of distribution (23 L/kg). This allows the adminis-
pharmacokinetic characteristics relative to erythromycin. Two sugar tration of high doses of azithromycin as a single dose and yet
units are linked to the 15-membered macrocylic lactone ring of maintenance of bacetriostatic levels in the infected tissues for
azithromycin. An aminosugar, D-desosamine, is attached through several days. Oral absorption is reduced 2-fold in the presence of
a b-glycosidic bond at C5 position and at C3 position a neutral food. The plasma protein binding is 12–50%. The plasma elimina-
sugar, L-cladinose is attached via a-glycosidic linkage. Currently tion half-life of azithromycin is 10–40 h. Unlike other macrolide
azithromycin is available in all kinds of dosage forms and is mainly antibiotics, the presence of b-lactamases does not affect
used in respiratory infections, sexually transmitted diseases, azithromycin activity. It is mainly excreted as unchanged drug
cutaneous and soft-tissue infections, etc. Azithromycin acts by through faeces (~50%) and to a lesser extent in the urine (4.5%;
inhibiting the protein synthesis by specifically binding to the 50S Lalak and Morris, 1993). The primary route of metabolism,
subunits of ribosomes in the apicoplast and subsequently inhibiting although minimal, is through hepatic demethylation. Most of the
m-RNA-directed polypeptide synthesis (Peters et al., 1992). This metabolites are excreted in bile. Metabolites are devoid of
either causes the death of the bacteria or inhibits their growth,
depending on the organism, its sensitivity to azithromycin and the
concentration of the drug in the infected tissue. The presence of a
* Correspondence to: Ramesh Mullangi, Drug Metabolism and Pharmacokinetics,
methyl-substituted nitrogen in the marcrolide ring of azithromycin Jubilant Biosys Ltd, Industrial Suburb, Yeshwanthpur, Bangalore-560 022, India.
confers an extra positive charge and this causes improved activity Email: [email protected]
against Gram-positive organisms when compared with erythromycin
(Farmer et al., 1992). Drug Metabolism and Pharmacokinetics, Jubilant Biosys Ltd, Industrial
Suburb, Yeshwanthpur, Bangalore-560 022, India
Following oral administration to humans, azithromycin is
rapidly absorbed from the gastrointestinal tract. The peak Abbreviations used: LLE, liquid–liquid extraction; MRM, multiple reaction
plasma concentration (Cmax) of azithromycin, depending on monitoring; SPE, solid-phase extraction.

Biomed. Chromatogr. 2013 Copyright © 2013 John Wiley & Sons, Ltd.
 K. Sharma and R. Mullangi

Sample preparation
In order to achieve good selectivity, sensitivity and ruggedness
in a bioanalytical method, efficient sample preparation is the
key component. Removal of interfering matrix compounds/
components is required to lower or eliminate the risk of matrix
effects in LC-MS procedures. During the pre-treatment, endoge-
nous components such as proteins, salts and lipids are removed
from biological matrix. These endogenous components modify
the sensitivity of method by influencing the ionization efficiency
on MS. Protein precipitation, liquid–liquid extraction (LLE) and
solid-phase extraction (SPE) are the commonly used sample
preparation techniques.
Figure 1. Structural representation of azithromycin and other macrolide
antibiotics Protein precipitation. Azithromycin is practically insoluble in
water and has high molecular weight. As suggested by many re-
searchers, the selection of protein precipitation method is advis-
significant antimicrobial activity (Schentag and Ballow, 1991). The able for compounds with good water solubility and molecular
oral bioavailability is 37%. Azithromycin does not interact with weight <500. Earliest methods reported by Xue-Min et al.
CYP450 isozymes present in liver and is not associated with the (2007) and Liu et al. (2007) included addition of a 3 volume
pharmacokinetic drug interactions seen with erythromycin and of methanol and analysis of supernatant layer for estimation of
other macrolides (Dunn and Barradell, 1996). azithromycin levels in human plasma. Both methods (Xue-Min
et al., 2007; Liu et al., 2007) used mass spectrometry as the detec-
tion mode monitoring molecular ions at 749 amu. The method
Scope reported by Xue-Min et al. (2007) lacked of critical stability
The purpose of this review is to: (a) list the various HPLC, LC-MS and parameters (viz. freeze–thaw and long term) and had issues of
LC-MS/MS methods in different biological matrices for quantitation interferences from endogenous material. These interferences
of azithromycin with other relevant information such as sample were resolved chromatographically from the peaks of interest
processing details, chromatographic conditions and validation using a cyno column, but peak shapes were broad. Both methods
parameters in tabular format; and (b) discuss relevant bioanalytical were applied to clinical studies for a dose level of 500 mg. Re-
strategies and considerations for method development to aid cently, Shen et al. (2010) published a method for azithromycin
in the LC-MS and LC-MS/MS analysis of azithromycin. Tables 1 quantification from rabbit conjunctiva tissue samples using a
and 2 list the chromatographic conditions, validation parameter 3 volume of acetonitrile and analysis of supernatant layer
details and applicability of the HPLC and LC-MS or LC-MS/MS using multiple reaction monitoring detection on MS. From
methods, respectively. the literature data it is apparent that the precipitation method
should not be the primary choice for processing such complex
analytes. Ultrafiltration methodology (Barrett et al., 2005) using a
Sample matrix centrifuge micropartition device with a molecular weight cut-off
For a bioanalyst typically the choice of matrix is blood or its derived 5000 was able to demonstrate a good validation and application
products (serum and plasma) for detecting and quantifying drugs support.
in either preclinical species or humans. Tissue concentrations of
azithromycin can be >50-fold higher than plasma levels, owing Liquid–liquid extraction. In LLE process analytes of interest
to ion trapping and high lipid solubility. The prolonged terminal from the biological matrix (blood/plasma/serum/urine/faeces/
half-life is thought to be due to extensive uptake and subsequent tissue/bile) sample are extracted with a water-immiscible or-
release of drug into plasma/serum from tissues. Owing to the long ganic solvent(s). Although LLE provides excellent sample clean-
half-life and significant plasma/circulating levels, the preferred up for many drugs, a broad range of endogenous compounds
matrix for quantitation of azithromycin is plasma/serum. The levels may be co-extracted during LLE. Following the LLE process, the
of azithromycin show significant variations associated probably organic solvent normally will be evaporated to ensure the
with the fact that serum protein binding of azithromycin is variable maximum analyte concentration. Many reported methods for
in the concentration range approximating human exposure, quantification of azithromycin from biological matrix include
decreasing from 51% at 0.02 mg/mL to 7% at 2 mg/mL. the addition of base to sample matrix to increase the sample
Plasma has been the choice of biological matrix for most of matrix pH close to the pKa of azithromycin (i.e. 8.74) followed
the assays developed for azithromycin (Nirogi et al., 2005; Barrett by extraction using an organic solvent (Fouda and Schneider,
et al., 2005; Wilms et al., 2005; Chen et al., 2006, 2007; Xue-Min 1995; Torano and Guchelaar, 1998; Wilms et al., 2005; Bahrami
et al., 2007; Liu et al., 2007; Yuzuak et al., 2007; Xu et al., 2008; et al., 2005; Bahrami and Mohammadi, 2006; Chen et al., 2006,
Choemunng and Na-Bangchang, 2010; Ebrahimzadeh et al., 2007; Yuzuak et al., 2007; Xu et al., 2008; Choemunng and Na-
2010), while some earlier publications were based on serum Bangchang, 2010). LLE is flexible, and offers opportunities for
azithromycin levels (Fouda and Schneider, 1995; Torano and parallel processing of large numbers of samples. However,
Guchelaar, 1998; Bahrami et al., 2005; Bahrami and Mohammadi, despite its widespread use, it is a tedious multistage operation,
2006). In addition, a few authors have also reported methods in has problems with emulsion formation and invariably consumes
tissue homogenates (Shen et al., 2010) and urine (Ebrahimzadeh large amounts of organic solvents. Off-line methodologies are
et al., 2010). often very tedious and time-consuming, and the risk of sample

wileyonlinelibrary.com/journal/bmc Copyright © 2013 John Wiley & Sons, Ltd. Biomed. Chromatogr. 2013
 Table 1. Summary validation of various published HPLC methods

Authors Samples processing details Chromatographic conditions Validation parameters Applicable conclusions
Ebrahimzadeh Matrix: human plasma and urine System: HPLC with UV detector. Linearity: 0.1–15 mg/mL The authors have discussed in
et al. (2010) (plasma and urine samples are Column: Shimpack CLC-C18 (r 2 > 0.998). detail the nature of the solvent,
diluted in the ratios of 1:4 and 1:1 (250  4.6 mm, 5 mm). Limit of detection: 0.03 mg/mL. compositions of the donor and
with ultrapure water, respectively). Mobile phase: isocratic mobile Precision: intra- and inter-day acceptor phase, ionic strength,

Biomed. Chromatogr. 2013

Extraction: LLLME procedure – phase comprising 0.01 M precisions (RSD) were 5.1 and stirring rate and extraction
diluted plasma/urine (2.5 mL) KH2PO4–ACN (58:42, v/v, final 6.8%, respectively. time on the extraction effi-
sample (donor phase) with pH pH 7.5) delivered at a flow-rate ciency was investigated and
10.5 (adjusted with 1 M potas- of 1 mL/min. optimized. The validated
sium carbonate) was taken into Detection: lmax set at 210 nm. method was used to quantitate
a 5 mL glass vial and over this Volume of injection: 5 mL. levels of azithromycin in
90 mL of n-octane (as an or Retention time: ~6.2 min. human plasma and urine.
Review of azithromycin bioanalytical methods

ganic membrane) was added Total run time: ~14 min.

with a micro-syringe. The vial
was placed in a water bath
(35
C) and contents were stirred
at 800 rpm for 20 min. Then 5 mL
of acceptor phase (0.01 M H3PO4:
KH2PO4) was carefully placed
(with the help of microsyringe)
on the organic membrane. The
pre-extraction equilibrium
between donor phase and
organic membrane was
achieved in 30 min. Then
the plunger of the syringe was

Copyright © 2013 John Wiley & Sons, Ltd.

pushed completely to suspend
the drop in the organic mem-
brane for the back-extraction
procedure. After 20 min the drop
was carefully withdrawn into the
syringe and used for analysis.
Internal standard: no IS used.
Bahrami and Matrix: human serum (1 mL). System: HPLC equipped with six- Linearity: 10–1000 ng/mL The authors have discussed in
Mohammadi Extraction: to an aliquot of serum, port switching valve and fluo- (r 2 > 0.996). detail the on-line derivatization
(2006) 100 mL of IS solution and 5 mL rescence detector. Limit of detection: 5 ng/mL. and optimization with FMOC-
diethyl ether were added, Column: Shimpack CLC-C8 Selectivity: endogenous peaks Cl. The validated method was
vortex-mixed for 30 s and (150  4.6 mm, 5 mm) and excess of the reagent used to quantitate levels of
centrifuged for 5 min at 6000g. maintained at 62
C. eluted within 3.2 min, hence azithromycin following oral ad
The organic layer was evapo- no interference observed at ministration of 400 mg
rated under a gentle stream of the retention times of the azithromycin suspension (200
nitrogen at 40
C and the analyte and IS.

wileyonlinelibrary.com/journal/bmc
(Continues)
 Table 1. (Continued)

Authors Samples processing details Chromatographic conditions Validation parameters Applicable conclusions
residue was reconstituted in Mobile phase: isocratic mobile Absolute recovery: 98  4 and mg/5 mL) to a healthy human
100 mL of 0.5 M phosphate phase comprising MeOH:0.05 M 65  5% for azithromycin and volunteer.
buffer (pH 8.5):ACN (1:3, v/v). sodium phosphate buffer IS, respectively.
Internal standard: amantadine (70:30, v/v) containing 0.1% Accuracy and precision: intra-
(10 mg/mL in water). TEA (v/v, pH 4.3). Before analy- and inter-day accuracies did
sis on-line derivatization with not deviate more than 9.2%
FMOC-Cl (500 mg/mL in ACN) in from 100% accuracy, whereas

wileyonlinelibrary.com/journal/bmc
a steel pipe [500  0.8 mm the precisions (CV) were
(i.d.)  1.6 (i.d.)] delivered at 1.8–16.5 and 1.5–14.4%,
0.1 mL/min. respectively.
Detection: excitation and emis-
sion set at 260 and 315 nm,
respectively.
Volume of injection: 50 mL.
Retention time: 5.8 and 4.2 min
for azithromycin and IS,
respectively.
Total run time: ~14 min.
Wilms et al. (2005) Matrix: human plasma/blood/ System: HPLC with fluorescence Linearity: 0.05–1.5 mg/L Authors have discussed in detail
PMNN (neutrophils) (0.5 mL). detector. (r2 > 0.999). chromatographic conditions
Extraction: to an aliquot of Column: Symmetry Shield C18 Limit of detection: 2 ng/mL (not and optimization of derivatiza-
plasma/blood/PMNN, 50 mL (100  4.6 mm, 3.5 mm) experimental value). tion process. The validated
MeOH, 100 mL IS solution and coupled to a pre-column Selectivity: prednisolone, method was used to quantitate
200 mL saturated disodium (20  4.6 mm) maintained ceftazidime, omeprazole, sulfa- levels of azithromycin in
carbonate (0.4 mg/mL water, at 28
C. methoxazole, trimethoprime, plasma, blood and PMNN

Copyright © 2013 John Wiley & Sons, Ltd.

pH 11) were added and vortex- Mobile phase: isocratic mobile itraconazole and fluconazole following oral administration
mixed for 2 s and to this phase comprising ACN:0.02 M showed no interference at the of 500 mg azithromycin to a
mixture 6 mL diethyl ether was phosphate buffer (pH 7.7, retention times of analyte and cystic fibrosis patient for at
added and shaken for 15 min adjusted with 10% KOH in IS. Ibuprofen interfered with IS least 35 days. The t1/2 for
(200/min) and centrifuged for water) (76:24, v/v). The flow retention time but under diode azithromycin in plasma, blood
10 min at 2550g. The organic rate was 1.5 mL/min. array detection (DAD) no inter- and PMNN was found to be 96,
layer was evaporated under a Detection: excitation and ference was observed so 210 and 340 h, respectively. At
gentle stream of nitrogen at emission set at 267 and authors suggested using a 220 h post-dosing >88% of
40
C and the residue was 317 nm, respectively. combination of DAD and fluo- azithromycin blood content
subjected to derivatization Volume of injection: 50 mL. rescence in patients with high was internalized in PMNNs.
with FMOC-Cl. doses of ibuprofen. No interfer-
Derivatization process: the organic ence was observed from
layer residue was dissolved in plasma and PMNN at retention
100 mL ACN, vortex mixed for times of analyte and IS.
20 s and centrifuged for 10 s at Absolute recovery: for azithromycin
2550g. To the clear aliquot, 100 the recoveries were 103, 94.5 and
K. Sharma and R. Mullangi

Biomed. Chromatogr. 2013

 Table 1. (Continued)

Authors Samples processing details Chromatographic conditions Validation parameters Applicable conclusions
mL of freshly prepared FMOC-Cl Retention time: ~12.5 and 10 min 103.7% from plasma (0.635 mg/L),
solution (1 mg/mL in ACN), for azithromycin and IS, blood (0.635 mg/L) and PMNN
100 mL water and 0.1 M phos- respectively. (2 mg/L), respectively; whereas
phate buffer (pH 7.5 adjusted Total run time: ~17 min. the recovery of IS (1 mg/L) from

Biomed. Chromatogr. 2013

with 10% KOH). The whole con- plasma and blood was 100 and
tents were incubated on a water 99.4%, respectively.
bath (40
C) for 40 min. Follow- Accuracy and precision: intra-
ing derivatization the aliquot and inter-day accuracy and
was used for analysis. precision was found to be
Internal standard: clarithromycin within acceptable limits.
(1 mg/100 mL water). Stability: the derivatized extracts
were stable up to 24 h, plasma
Review of azithromycin bioanalytical methods

and blood samples were stable

up to 18 months at 30
C and
isolated PMNN samples were
stable up to 4 months.
Bahrami et al. Matrix: human serum (1 mL). System: HPLC with fluorescence Linearity: 10–500 ng/mL (r 2 > 0.998). The validated method was used
(2005) Extraction: to an aliquot of serum detector. Limit of detection: 2 ng/mL. to quantitate levels of
5 mL hexane was added, vortex- Column: phenyl column (150  6 Selectivity: endogenous peaks azithromycin following oral
mixed and centrifuged for 5 min mm, 5 mm) maintained at 62
C. eluted within 2 min, hence no administration of 2  250 mg
at 6000g. The organic layer was Mobile phase: isocratic mobile interference was observed tablets of azithromycin to
evaporated under a gentle phase comprising MeOH:0.05 M from serum samples at the healthy human volunteers
stream of nitrogen at 40
C and sodium phosphate buffer retention timess of the analyte (n = 12) in a bioequivalence
the residue was subjected to (71:29, v/v) containing 2 mL/L and IS. The authors have also study.
derivatization with FMOC-Cl TEA and final pH adjusted to tested various commonly used

Copyright © 2013 John Wiley & Sons, Ltd.

(optimized with borate buffer 5.9 with phosphoric acid. The drugs and confirmed that
and glycine to remove excess flow rate was 2.5 mL/min. these drugs were not detected
FMOC-Cl). in the present method.
Internal standard: clarithromycin. Detection: excitation and emis- Gentamycin, topiramte, eryth-
sion set at 260 and 315 nm, romycin and etidronate did
respectively. react with FMOC-Cl but did not
Volume of injection: 20 mL. give interfere with retention
Retention time: 5.3 and 3.9 min times of analyte and IS.
for azithromycin and IS, Absolute recovery: 101  3 and
respectively. 99  5% for azithromycin and
Total run time: ~7 min. IS, respectively.
Accuracy and precision: intra-
and inter-day accuracies were
100.3–107.8 and 101.3–107.2%,
respectively; whereas the
(Continues)

wileyonlinelibrary.com/journal/bmc
 Table 1. (Continued)

Authors Samples processing details Chromatographic conditions Validation parameters Applicable conclusions
precisions (CV) were 1.7–11.9
and 2.0–13.3%, respectively.
Stability: azithromycin was found
to be stable for 60 days at

wileyonlinelibrary.com/journal/bmc
80
C.
Torano and Matrix: human serum (1 mL). System: HPLC with fluorescence Linearity: 0.0988–4.94 mg/L The authors have discussed the
Guchelaar Extraction: to an aliquot of serum, detector. (r2 > 0.9994). conditions required for optimi-
(1998) 20 mL of IS solution and 200 mL Column: Supelcosil C18 (125  4.6 Limit of detection: 2 ng/mL. zation. The validated method
of saturated sodium carbonate mm, 5 mm) coupled to a guard Selectivity: endogenous peaks was used to quantitate levels
solution (0.5 g/mL, pH 12) were column (20  4.6 mm, 5 mm) eluted within 2 min and no in of azithromycin in a human
added, and vortex-mixed for maintained at 50
C. terference were observed from PK study.
10 s. To this 6 mL diethyl ether Mobile phase: isocratic mobile serum samples at the retention
was added and shaken for 15 phase comprising ACN–0.05 M times of the analyte and IS.
min (270/min) and centrifuged potassium diphosphate buffer Absolute recovery: 93–98 and
for 5 min at 2700g. The solu- (60:40, v/v) containing 0.5 mL/L 97–102% for azithromycin and
tion was allowed to freeze at TEA and final pH adjusted to IS, respectively.
20
C in an ethanol bath. The 7.5 with 10% KOH. The flow Stability: the derivatized solution
organic layer was evaporated rate was 2 mL/min. was found to be stable for 4 h
under a gentle stream of nitro- Detection: excitation and emis- and azithromycin was found to
gen at 40
C. The residue was sion set at 255 and 315 nm, be stable for 7 days at 20
C
dissolved in 200 mL of ACN and respectively. in serum.
vortex mixed for 30 s and Volume of injection: 50 mL.

Copyright © 2013 John Wiley & Sons, Ltd.

centrifuged at 2700g. To an al- Retention time: 20.7 and 17.1
iquot, 100 mL of 2.5 mg min for azithromycin and IS,
FMOC-Cl/mL ACN solution and respectively.
75 mL of 0.1 M phosphate buffer Total run time: 25 min.
in water (pH 7.5 with 10% KOH)
were added and incubated at
40
C for 40 min. Following
incubation the solution was
used for analysis.
Internal standard: roxithromycin
(3.12 mg/L in ACN).
ACN, acetonitrile; CV, coefficient of variation; DCM, dichloromethane; FMOC-Cl, 9-fluorenylmethylcarbonyl chloride; F/T, freeze–thaw; H3PO4, phosphoric acid; HPLC, high-perfor-
mance liquid chromatography; IS, internal standard; KH2PO4, potassium dihydrogen phosphate; KOH, potassium hydroxide; LLLME, liquid–liquid–liquid microextraction; MeOH,
methanol; PK, pharmacokinetic; PMNN, polymorpho-nuclear neutrophils; RSD, relative standard deviation; SPE, solid-phase extraction; TEA, triethylamine; UV, ultraviolet
K. Sharma and R. Mullangi

Biomed. Chromatogr. 2013

 Table 2. Summary validation of various published LC-MS and LC-MS/MS methods

Authors Samples processing Chromatographic conditions Validation parameters Applicable conclusions

details
Shen et al. (2010) Matrix: 200 mL of rabbit conjunc- System: LC-ESI-MS/MS by SRM in Regression type: linear fit with The validated method can be
tiva tissue homogenate. positive mode. weighing factor of 1/x2. used routinely for analysis of
Extraction: protein precipitation Column: Shiseido Capcell C18 Calibration range: 10–2000 ng/mL azithromycin in conjunctiva

Biomed. Chromatogr. 2013

with ACN. (75  3 mm, 3 mm) maintained (r2 > 0.999). samples collected during PK
Internal standard: roxithromycin at 30
C. Absolute recovery: mean recover- studies in rabbits.
(20 mL of 2000 ng/mL in ACN). Mobile phase: isocratic mobile ies were found to be 96.0  5.8,
phase comprising 10 mM ammo- 101.3  3.7 and 93.4  6.5% for
nium acetate (pH 5.2 adjusted azithromycin at 20, 400 and 600
with 0.1% acetic acid)–MeOH ng/mL, whereas for IS it was
(18:82, v/v) delivered at a flow 96.6  4.7%.
Review of azithromycin bioanalytical methods

rate of 0.3 mL/min. Specificity: no interference peaks

Injection volume: 5 mL. were observed at the retention
Mass spectrometry detection: time of the analyte and IS from
azithromycin: m/z 375 six different sources of rabbit
[M + 2H]2+ ! 591 and IS: m/z blank conjunctiva samples.
837 [M + H]2+ ! 679. Accuracy and precision: intra-
Retention time: 1.46 and 2.37 and inter-day accuracies were
min for azithromycin and IS, 92.0 and 103.7%, whereas the
respectively. intra- and inter-day RSD were
Total run time: 3 min. 4.30–7.38 and 3.02–8.59%,
respectively.
Stability studies: found to be stable
at room temperature for 24 h, up
to 10 days at 20
C, in auto-

Copyright © 2013 John Wiley & Sons, Ltd.

sampler for 24 h and after three
F/T cycles.
Choemunng and Matrix: 200 mL of human plasma. System: LC-ESI-MS by SRM in Regression type: linear fit with The validated method was suc-
Na-Bangchang Extraction: to an aliquot of positive mode. weighing factor of 1/x2. cessfully used to determine
(2010) plasma, 50 mL IS solution Column: Hypersil Gold C18 Calibration range: 5–1000 ng/mL the PK parameters of
(5 ng/mL in MeOH) was added, (150  4.6 mm, 5 mm) (r2 > 0.999). azithromycin (250 mg), when
thoroughly mixed and to this maintained at 25
C. Absolute recovery: mean recovery given in combination with
MeOH (50 mL) and 0.025 M Mobile phase: isocratic mobile was found to be >85% for fosmidomycin (750 mg) in Thai
carbonate–bicarbonate buffer phase comprising 20 mM ammo- azithromycin at 5, 50 and 200 male patients suffering with
pH 9.5 (250 mL) was added, nium acetate (pH 5.2):ACN: ng/mL, whereas the recovery for malaria. The derived PK param-
vortex-mixed and then extracted MeOH (50: 40:10, v/v/v) delivered IS was 90%. eters, viz., Cmax (303  12 ng/
with diethylether–DCM (70:30, at a flow rate of 0.3 mL/min. Specificity: no interference peaks mL), Tmax (2.0  0.1 h) and t1/2
v/v) by vortex mixing for 30 s. Injection volume: 10 mL. were observed at the retention (33.2  2.3 h, at steady state)
Following this organic layer was time of the analyte and IS, which were in agreement with earlier
separated and centrifuged at
(Continues)

wileyonlinelibrary.com/journal/bmc
 Table 2. (Continued)

Authors Samples processing Chromatographic conditions Validation parameters Applicable conclusions

details
2500g for 5 min at 4
C; the clear Mass spectrometry detection: m/z was evaluated from 10 blank hu reported values, confirming
supernatant was dried under a 749.6 and 837.6 for azithromycin man plasma samples. the suitability of this method.
gentle stream of nitrogen at and IS, respectively. Accuracy and precision: intra- and
40
C. The residue was Retention time: 10.71 and 13.67 inter-day accuracies percentage
reconstituted in 200 mL of min for azithromycin and IS, deviation of mean value from

wileyonlinelibrary.com/journal/bmc
mobile phase. respectively. theoretical value were 4.29–
Internal standard: roxithromycin. Total run time: 15 min. 2.83 and 5.26–6.80%, respec-
tively. The intra- and inter-day
precision CVs were 0.56–1.44
and 0.35–2.06%, respectively.
Stability studies: azithromycin
was found to be stable for
1 month at 20
C.
Xu et al. (2008) Matrix: 200 mL of human plasma. System: LC-ESI-MS by SIM in Regression type: linear fit with Addition of 1 M NaOH in extraction
Extraction: to an aliquot of plasma, positive mode. weighing factor of 1/x2. process helped in dissociation
IS solution (1 mg), 80 mL of 1 M Column: Shimpack VP-ODS C18 Calibration range: 5–2000 ng/mL of analyte from plasma and
NaOH solution and 3 mL of (150  2 mm, 5 mm) (r > 0.99). reduces endogenous interfer
DCM:EtOAc (20:80, v/v) were maintained at ambient room Limit of detection: 2 ng/mL. ence. Addition of TEA in mobile
added and vortex mixed for 3 temperature. phase helped to symmetric
min and centrifuged at 13,800g Mobile phase: isocratic mobile Absolute recovery: mean recoveries peak shapes without tailing.
for 10 min. The organic layer phase comprising ACN–water were found to be 92.71  5.06, The validated method was
was dried under a gentle stream (0.5% TEA, pH 6.2 adjusted 89.07  5.51, 86.64  3.95 and to quantitate the levels of
of nitrogen at 40
C. The residue with acetic acid) delivered at a 87.12  4.25% for azithromycin azithromycin following oral
was reconstituted in 200 mL of flow rate of 0.2 mL/min. at 10, 50, 500 and 2000 ng/mL, administration of 500 mg of

Copyright © 2013 John Wiley & Sons, Ltd.

mobile phase. Injection volume: 10 mL. respectively. disintegrating tablets to 20
Internal standard: roxithromycin Mass spectrometry detection: Specificity: no interference peaks healthy Chinese men in a
(stocks made in ACN: water, m/z 749.35 and 837.35 for were observed at the retention bioequivalence study.
65:35, v/v). azithromycin and IS, time of the analyte and IS evalu-
respectively. ated from six different sources of
Retention time: 4.0 and 4.8 min for human blank plasma samples.
azithromycin and IS, respectively. Accuracy and precision: intra-
Total run time: 6 min. and inter-day accuracies were
94.96–102.32 and 94.22–104.15%,
whereas the intra- and inter-day
precisions (RSD) were 3.48–6.36
and 2.93–5.21%, respectively.
Stability studies: azithromycin
was found to be stable in the
auto-sampler (4
C for 12 h),
K. Sharma and R. Mullangi

Biomed. Chromatogr. 2013

 Table 2. (Continued)

Authors Samples processing Chromatographic conditions Validation parameters Applicable conclusions

details
after 8 weeks storage at 20
C
and after three F/T cycles.
Yuzuak et al. (2007) Matrix: 200 mL of human plasma. System: LC-ESI-MS/MS by MRM in Regression type: linear fit with The authors opined that this

Biomed. Chromatogr. 2013

Extraction: to an aliquot of positive mode. weighing factor of 1/x2. method is suitable for analysis
plasma, 50 mL IS solution (50 mL Column: Sunfire C18 (50  2.1 Calibration range: 2–1000 ng/mL of azithromycin in PK, pharma-
MeOH to blank sample) and mm, 3.5 mm) maintained (r > 0.98). codynamic and BA/BE studies.
250 mL of 0.25 M carbonate– at 30
C. Absolute recovery: mean recovery
bicarbonate buffer solution Mobile phase: 1.54 g ammonium for azithromycin was 81.97%.
(pH 9.5) were added and acetate, 250 mL water, 570 mL Specificity: no interference peaks
vortex-mixed for 10 s. To this 3 ACN, 180 mL MeOH and 0.6 mL were observed at the retention
mL of diethylether:DCM (7:3; of glacial acetic acid. The flow time of azithromycin and IS
Review of azithromycin bioanalytical methods

v/v) was added vortex mixed rate was 0.2 mL/min. evaluated from six different
for 30 s and centrifuged at Injection volume: 10 mL. sources of human blank plasma
4600g for 5 min at 4
C. Then Mass spectrometry detection: m/z samples.
the tubes were stored at 749.58 ! 591.6 and 837.64 ! 158.2 Accuracy and precision: intra- and
70
C for 10 min. Afterwards for azithromycin and IS, respectively. inter-day accuracies were 86.18–
the upper organic layer was Retention time: 0.9 and 1.0 min for 114.75 and 90.50–110.07%,
dried under a gentle stream of azithromycin and IS, respectively. whereas the intra- and inter-day
nitrogen at 40
C. The residue Total run time: 2 min. precisions (RSD) were 5.46–18.76
was reconstituted in 200 mL of and 8.96–14.08%, respectively.
mobile phase. Stability studies: azithromycin
Internal standard: roxithromycin was found to be stable in the
(5 mg/mL in MeOH). auto-sampler (115 h) and after
five F/T cycles.

Copyright © 2013 John Wiley & Sons, Ltd.

Chen et al. (2007) Matrix: 500 mL of human plasma. System: UPLC-ESI-MS/MS by Regression type: linear fit with The validated method was to
Extraction: to an aliquot of plasma, MRM in positive mode. weighing factor of 1/x2. quantitate the levels of
IS solution (100 mL) and 400 mL Column: Acquity UPLC™ BEH C18 Calibration range: 1–1000 ng/mL azithromycin following oral
of 0.05 M sodium carbonate so (50  2.1 mm, 1.7 mm) maintained (r2 > 0.99). administration of 500 mg of
lution were added and vortex at 40
C. Absolute recovery: mean recover- tablets to 20 male healthy
mixed for 1 min. To this 3 mL Mobile phase: gradient elution ies were found to be 92.1  5.0, volunteers in a clinical trial. The
diethyl ether was added vortex mobile phase A (50 mM aqueous 87.9  2.9 and 92.3  4.5 % for observed Cmax and Tmax were
mixed for 1 min then shaken for ammonium acetate) and mobile azithromycin at 2, 200 and 800 565.9  207.6 ng/mL and
10 min on a mechanical shaker. phase B (ACN) delivered at a ng/mL, respectively. For IS the 1.9  0.6 h, respectively, with t1/2
Following this it was centrifuged flow rate of 0.35 mL/min. mean recovery was 80.6  2.7%. of 50.1  5.0 h. The AUC0–1 was
at 3500g for 10 min. The organic Injection volume: 5 mL. Specificity: no interference peaks found to be 5243.3  1257.3 ng.
layer was dried under a gentle Mass spectrometry detection: m/z were observed at the retention h/mL. The obtained PK parame-
stream of nitrogen at 40
C. The 749 ! 591 and 837 ! 679 for time of the analyte and IS eval- ters were in agreement with
residue was reconstituted in azithromycin and IS, respectively. uated from six different earlier reported values.
200 mL of ACN–water (70:30, v/v).
(Continues)

wileyonlinelibrary.com/journal/bmc
 Table 2. (Continued)

Authors Samples processing Chromatographic conditions Validation parameters Applicable conclusions

details
sources of human blank
plasma samples.
Internal standard: roxithromycin Retention time: 1.26 and 0.81 Accuracy and precision: intra-day
(1000 ng/mL in MeOH). min for azithromycin and IS, precisions at low (2 ng/mL),
respectively. mid (200 ng/mL) and high (800
Total run time: 3 min. ng/mL) QC levels were 8.4, 5.2

wileyonlinelibrary.com/journal/bmc
and 4.05%, respectively, and
those of inter-day analysis
were 4.7, 7.3 and 9.4%, respec-
tively, with an accuracy relative
error within 1.3–5.7%.
Stability studies: azithromycin
was found to be stable for 4 h
at room temperature, for 4 h in
auto-sampler, at 4
C for 25
days, at 20
C for 30 days and
after three F/T cycles.
Liu et al. (2007) Matrix: 100 mL of human plasma. System: LC-ESI-MS by SIM in Regression type: linear fit with The validated method was to
Extraction: to an aliquot of positive mode. weighing factor of 1/x. quantitate the levels of
plasma, 10 mL of IS solution Column: Phenomenex CN Luna Calibration range: 4.69-600 ng/mL azithromycin following oral
was added, mixed and to this (100  2.1 mm, 3 mm) (r2 > 0.999). administration of 2  250 mg
300 mL of MeOH was added maintained at 20
C. Absolute recovery: mean recov- of dispersible tablets to 20
vortex mixed for 3 min and Mobile phase: isocratic mobile eries were found to be male healthy volunteers in a PK
centrifuged for 10 min at phase comprising MeOH– 87.41  2.39, 90.30  1.10 and study. Cmax (425.71  184.32

Copyright © 2013 John Wiley & Sons, Ltd.

13,400g. Supernatant was used ACN–water with 0.1% formic 99.97  0.68% for azithromycin ng/mL) was attained at
for analysis. acid and 0.1% acetic acid at 9.38, 37.5 and 500 ng/mL, 3.13  2.18 h (Tmax). The
Internal standard: erythromycin (1.5 (12:30:58, v/v/v) delivered at a respectively. AUC0–120 h and t1/2 were found
mg/mL in MeOH). flow rate of 0.2 mL/min. Specificity: no interference peaks to be 3880.80  1183.56 ng
Injection volume: 10 mL. were observed at the retention h/mL and 26.32  10.25 h,
Mass spectrometry detection: m/z time of the analyte and IS eval- respectively.
749.4 and 734.3 for azithromycin uated from human blank
and for IS, respectively. plasma sample.
Retention time: 3.90 and 2.80 Accuracy and precision: intra-
min for azithromycin and IS, and inter-day accuracy and
respectively. precision values were found to
Total run time: 5 min. be within acceptable limits.
Stability studies: azithromycin
was found to be stable for 4 h
at room temperature and 12 h
in auto-sampler.
K. Sharma and R. Mullangi

Biomed. Chromatogr. 2013

 Table 2. (Continued)

Authors Samples processing Chromatographic conditions Validation parameters Applicable conclusions

details
Xue-Min et al. (2007) Matrix: 200 mL of human plasma. System: LC-ESI-MS by SIM in pos- Regression type: linear fit with The validated method was used to
Extraction: to an aliquot of plasma itive mode. weighing factor of 1/x2. quantitate the levels of

Biomed. Chromatogr. 2013

50 mL IS solution was added, Column: XTerra C18 (100  2.1 Calibration range: 3.048–1016 azithromycin following oral
vortex mixed for 30 s and then mm, 5 mm) maintained at 25
C. ng/mL (r2 > 0.9995). administration of 2  250 mg
extracted with 0.7 mL MeOH by Mobile phase: gradient elution Absolute recovery: the recoveries of tablets to male Chinese volun-
vortex mixing and then using MeOH–ammonium acetate of azithromycin were found teers (n = 20) in a bioequivalence
centrifuged at 14,000g for 10 min. (10 mM, 1% acetic acid delivered to be 104.2  7.3, 107.3  3.9, study. Authors concluded that
To the top organic layer (0.5 mL), at a flow rate of 0.3 mL/min. 97.64  1.2 and 94.86  5.7% at the both the reference and test
70 mL 30 mM ammonium acetate Injection volume: 10 mL. 3.048, 30.48, 203.2 and 1016 formulations were bioequivalent.
was added, mixed and Mass spectrometry detection: for ng/mL, respectively.
Review of azithromycin bioanalytical methods

centrifuged at 14,000 rpm for azithromycin: m/z 749.8 and Specificity: no endogenous inter-
5 min. The upper aliquot was IS: m/z 837.9. ference was observed at the
used for analysis. Retention time: 1.8 and 5.2 min for retention time of the analyte
Internal standard: roxithromycin azithromycin and IS, respectively. and IS evaluated from blank
(1.999 mg/mL in MeOH). Total run time: 9.0 min. human plasma samples.
System: LC-ESI-MS by SIM in Precision: intra- and inter-day
positive mode. precisions (RSD) were 2.1–5.9
Column: Hypersil HyPurity C18 and 3.6–11.2%, respectively.
(150  2.1 mm, 5 mm) Stability studies: azithromycin
maintained at 40
C. was found to be stable for 24 h
Mobile phase: isocratic mobile at room temperature and for
phase comprising 20 mM 50 days at 85
C.
Chen et al. (2006) Matrix: 200 mL of human plasma. ammonium acetate (pH 5.2 Regression type: weighted The chosen column provided the

Copyright © 2013 John Wiley & Sons, Ltd.

Extraction: to an aliquot of plasma, adjusted with acetic acid)–ACN– regression. symmetrical and peak shapes
IS solution (50 mL) and 50 mL of MeOH (50: 40:10, v/v/v) delivered Calibration range: 2–2000 ng/mL and highest intensity for
saturated solution of sodium at a flow rate of 0.2 mL/min. (r2 > 0.9977). azithromycin. The authors have
carbonate were added and Injection volume: 5 mL. Limit of detection: 0.5 ng/mL. discussed the role of mobile
vortex mixed for 1 min. To this Mass spectrometry detection: for Absolute recovery: mean recov- phase pH on the ionization and
1 mL TBME–hexane (50:50, v/v) azithromycin: m/z 375 [M + 2H]2+ eries were found to be 81.2 sensitivity of the analyte and IS.
was added vortex mixed for and IS: m/z 749 [M + H]+. and 75.5% for azithromycin The validated method was
3 min, then centrifuged at and IS, respectively. to quantitate the levels of
14,000g for 5 min. The organic Specificity: no interference peaks azithromycin following oral
layer (1.5 mL) was dried under were observed at the retention administration of 500 mg of
a gentle stream of nitrogen at time of the analyte and IS evalu- tablets to healthy volunteers
40
C. The residue was ated from six different sources of (n = 24) in a bioequivalence study.
reconstituted in 200 mL mobile human blank plasma samples. The key PK parameters viz.,
phase. Accuracy and precision: intra- and AUC0–t, AUC0–1, Cmax, Tmax and
inter-day accuracies relative t1/2 for test formulation were

(Continues)

wileyonlinelibrary.com/journal/bmc
 Table 2. (Continued)

Authors Samples processing Chromatographic conditions Validation parameters Applicable conclusions

details
Internal standard: clarithromycin Retention time: ~2.5 and 5.0 min error (%) were 3.4–4.9 and 2.2– found to be bioequivalent
(500 ng/mL in MeOH). for azithromycin and IS, 7.0, respectively, whereas the to reference formulation.
respectively. precisions (RSD) were 4.8–8.6
Total run time: 7.0 min. and 6.4–10.7%, respectively.

wileyonlinelibrary.com/journal/bmc
Stability studies: azithromycin was
found to be stable for 24 h at
room temperature, for 24 h as
dry extract, at 20
C for 2
months and after four F/T cycles.
Barrett et al. (2005) Matrix: 500 mL of human plasma. System: LC-ESI-MS/MS by MRM in Regression type: linear fit with The validated method was used
Extraction: to an aliquot of plasma, positive mode. weighing factor of 1/x. to quantitate the levels of
25 mL IS solution and 50 mL ACN Column: XTerra RP8 (150  4.6 mm, Calibration range: 2.55-551.43 ng/mL azithromycin following oral
were added and vortex mixed 5 mm) maintained at ambient (r > 0.999). administration of 250 mg of
for 5 s and filtered through room temperature. Limit of detection: 1.25 ng/mL. tablets to healthy volunteers
Centrifree (cut off molecular Mobile phase: isocratic mobile Absolute recovery: mean recover- (n = 46) in a PK study.
weight is 5000) in a centrifuge at phase comprising ACN–MeOH– ies for azithromycin were 96.44– Although the calculated mean
10,000 rpm for 10 min. The 0.05 mol/L ammonium acetate 111.80%, whereas for IS the Cmax (244.65 ng/mL) and t1/2
filterate was used for bioanalysis. (31:19:50, v/v/v) delivered at a mean recovery was 87.96%. (3 h) values from this study were
Internal standard: roxithromycin flow rate of 0.7 mL/min. Specificity: no interference peaks lower than the reported values,
(0.44 mg/mL ACN–water, Injection volume: 7.5 mL. were observed at the retention the PK data obtained are compa-
70:30, v/v). Mass spectrometry detection: time of azithromycin and IS rable with published data.
m/z 749.0 ! 591.6 and evaluated from six different

Copyright © 2013 John Wiley & Sons, Ltd.

837.1 ! 679.6 for azithromycin sources of human blank plasma
and IS, respectively. samples.
Retention time: 3.5 and 5.9 min for Accuracy and precision: intra- and
azithromycin and IS, respectively. inter-day accuracy and preci-
Total run time: 10 min. sion were found to be within
acceptable limits.
Stability studies: azithromycin
was found to be stable for 2 h
at room temperature, in auto-
sampler (10
C) for 24 h, at
70
C for two months and
after four F/T cycles.
Nirogi et al. (2005) Matrix: 500 mL of human plasma. System: LC-ESI-MS/MS by MRM in Regression type: linear fit with The validated method was suc-
Extraction: to an aliquot of positive mode. weighing factor of 1/x2. cessfully applied to quantitate
plasma, 20 mL IS solution and Column: Symmetry C18 (150  4.6 Calibration range: 5–500 ng/mL azithromycin concentrations in
500 mL of 60 mM sodium mm, 5 mm) maintained at 30
C. (r 2 > 0.998). human plasma following
K. Sharma and R. Mullangi

Biomed. Chromatogr. 2013

 Table 2. (Continued)

Authors Samples processing Chromatographic conditions Validation parameters Applicable conclusions

details
carbonate buffer (pH 11) were Mobile phase: isocratic mobile Absolute recovery: mean recover- administration of 200 mg/5 mL
added and vortex-mixed for 30 s phase comprising 80 mM ies were found to be 71.1  2.8 oral dose of azithromycin to
and loaded onto Oasis SPE HLB ammonium acetate (pH 6.6)– and 84.8  1.6% for azithromycin healthy human volunteers.

Biomed. Chromatogr. 2013

cartridge (30 mg), pre-condi- ACN–MeOH (25:57:18, v/v/v) and IS, respectively.
tioned with 1 mL MeOH followed delivered at a flow rate of Specificity: no interference peaks
by 1 mL water. The cartridge was 1.5 mL/min with a split ratio were observed at the retention
rinsed in RapidTrace with 1 mL of 90:10. time of the analyte and IS eval-
water followed by 10% MeOH in Injection volume: 10 mL. uated from five human blank
water and finally eluted with Mass spectrometry detection: m/z plasma samples.
1 mL MeOH. The eluate was dried 749 ! 591 and 734 ! 576 for Accuracy and precision: intra- and
under a gentle stream of nitro- azithromycin and IS, respectively. inter-day accuracies were 98.5–
Review of azithromycin bioanalytical methods

gen at 40
C. The residue was Retention time: 1.0 and 1.2 min 107.0 and 96.9–102.4%, respec-
reconstituted in 250 mL water: for azithromycin and IS, tively; whereas the precisions
ACN: MeOH (25:57:18, v/v/v). respectively. (RSD) were 0.7–4.3 and 1.1–
Internal standard: erythromycin Total run time: 2 min. 7.1%, respectively.
(10 mg/mL in MeOH). Dilution integrity: the upper
concentration limits can be
extended with acceptable ac-
curacy and precision by 4-fold.
Stability studies: azithromycin was
found to be stable for 24 h at
room temperature, for 24 h in
auto-sampler, at ≤ 50
C for 30
days and after three F/T cycles.

Copyright © 2013 John Wiley & Sons, Ltd.

Fouda and Schneider Matrix: 50 mL of human serum. System: LC-APCI -MS by SIM in Regression type: linear fit with The validated method was to quan-
(1995) Extraction: to an aliquot of serum, positive mode. weighing factor of 1/x2. titate the levels of azithromycin
50 mL IS solution and 50 mL of Column: Hypersil ODS (30  4.6 mm, Calibration range: 10–250 ng/mL following oral administration
0.06 M potassium carbonate 3 mm) with a pre-column stainless (r2 > 0.985). to healthy volunteers. These
were added, vortex-mixed and filter (0.5 mm) maintained at Absolute recovery: mean recov- concentrations showed good
to this 200 mL water was added ambient room temperature. ery was found to be 66% for correlation with duplicate aliquot
and contents were vortexed for both azithromycin. concentrations measured by
several seconds. To this 3 mL Mobile phase: isocratic mobile Specificity: no interference peaks HPLC-electrochemical assay.
TBME was added, vortex mixed phase comprising ACN– were observed at the retention
for 20 s and centrifuged at 3300 MeOH–THF–50 mM ammonium time of the analyte and IS.
rpm for 3 min. The clear organic acetate, (44:19:3:34, v/v/v/v) Known antibiotics and metabo-
layer was dried in a vortex evap delivered at a flow rate of lites of azithromycin did not
orator at 40
C. The residue was 1 mL/min. affect the chromatography.
reconstituted in 100 mL mobile Injection volume: 50 mL.
phase.
(Continues)

wileyonlinelibrary.com/journal/bmc
 K. Sharma and R. Mullangi

loss and/or cross contamination is high. Most authors have

liquid chromatography; MeOH, methanol; MRM, multiple reaction monitoring; NaOH, sodium hydroxide; ODS, octadecylsilyl; PK, pharmacokinetic(s); RSD, relative standard devia-
ACN, acetonitrile; APCI, atmospheric pressure chemical ionization; AUC0–t, area under the plasma concentration–time curve from time zero to last measurable time point; AUC0–1,

variation; DCM, dichloromethane; ESI, electrospray ionization; EtOAc, ethyl acetate; IS, internal standard; F/T, freeze–thaw; HLB, hydrophilic–lipophilic based; IS, internal standard; LC,

tion; SIM; selected ion monitoring; SRM, single reaction monitoring; SPE, solid-phase extraction; Tmax, time to reach Cmax; t1/2, terminal half-life; TBME, ter-butyl methyl ether; TEA,
area under the plasma concentration–time curve from time zero to infinity; BA/BE, bioavailability/bioequivalence; Cmax, maximum observed plasma concentration; CV, coefficient of
reported extraction with nonpolar solvents with a low relative

Applicable conclusions
polarity. Torano and Guchelaar (1998), Wilms et al. (2005),
Bahrami and Mohammadi (2006) and Chen et al. (2007) used sin-
gle-solvent diethyl ether for extraction, while others (Yuzuak
et al., 2007; Choemunng and Na-Bangchang, 2010) used a
combination of diethyl ether with dichloromethane for extrac-
tion from biological matrix. Some authors have reported the
use of other solvents, like ter-butyl methyl ether (TBME) by
Fouda and Schneider (1995), hexane by Bahrami et al. (2005)
and ethyl acetate–dichloromethane combination by Xu et al.
(2008). Chen et al. (2006) also reported low recoveries with a
direct precipitation approach, but managed to attain a higher
recovery with a combination of TBME–hexane (50:50, v/v). They
89–108 and 101–106%, respec-
tively, whereas the CVs for preci-

also reported a matrix effect with the ultrafiltration approach

and inter-day accuracies were

sion were 3–14 and 5.1–5.5%,

Accuracy and precision: intra-

suggested by Barrett et al. (2005), but this could be due to their

Validation parameters

detection mode as they were using SIM mode, while Barrett et al.
(2005) used MRM mode for detection.
Addition of a small proportion of polar aprotic solvent (i.e.
dichloromethane) was suggested by a few authors (Yuzuak
et al., 2007; Xu et al., 2008; Choemunng and Na-Bangchang,
respectively.

2010) to attain better selectivity. Such addition may lead to

lower sensitivity, as evident from comparison between two pub-
lications in 2007. Yuzuak et al. (2007) reported a lower limit
of quantification 2-fold higher than the method published by
Chen et al. (2007), with similar strategies apart from the addition
of polar aprotic solvent in combination with nonpolar solvent
(diethyl ether).
Mass spectrometry detection: m/z

azithromycin and IS co-eluted).

749 and 752 for azithromycin
Chromatographic conditions

Retention time: 1.9 min (both

Solid-phase extraction. Currently SPE is a very popular tech-

nique for rapid and selective sample preparation. SPE is an alter-
native to LLE owing to its simplicity, low cost, ease of automation
and IS, respectively.

Total run time: 4 min.

and minimimum use of breakable speciality glassware and large

triethylamine; THF, tetrahydrofuran; UPLC, ultra-performance liquid-chromatography

quantities of organic solvents. SPE offers more advantages over

traditional LLE, such as high recoveries of analyte, concentration
of analyte, highly purified extract and ability to simultaneously
extract a wide range of polarity, as well as reducing the forma-
tion of incomplete phases.
Only one SPE method has been reported for quantification of
azithromycin from biological matrix (plasma). Nirogi et al. (2005)
published a method utilizing OasisW HLB SPE cartridges. OasisW
HLB is a hydrophilic–lipophilic balanced reversed-phase sorbent,
erythromycin (d3) (50 ng/mL in

made from a specific ratio of two monomers: N-vinylpyrrolidone

Internal standard: deuterated

and divinyl benzene. N-Vinylpyrrolidone is hydrophilic in na-

Samples processing

ture, while divinyl benzene is lipophilic. The extraction meth-

ACN–water, 50:50, v/v).

odology involves retaining the analytes on the stationary bed

details

and washing out the plasma interferences with appropriate

washing solvents (water wash followed by 10% methanol in
the water wash). Analytes are eluted with organic solvent,
evaporated and reconstituted in chromatography mobile
phase media. Nirogi et al. (2005) reported low recoveries with
direct protein precipitation with acetonitrile, perchloric acid
and trichloric acid. They also reported inconsistencies in recov-
ery for different organic solvents, such as diethyl ether, ethyl
Table 2. (Continued)

acetate, TBME, hexane, chloroform and dichloromethane,

which subsequently made them opt for an SPE procedure with
a recovery of 71%.
Authors

Internal standard selection

Internal standard (IS) with similar phyisco-chemical properties to
the analyte provide various advantages in the determination of

wileyonlinelibrary.com/journal/bmc Copyright © 2013 John Wiley & Sons, Ltd. Biomed. Chromatogr. 2013
Review of azithromycin bioanalytical methods

drug concentration in biological matrix such as reduction in peak doubling after numerous injections with FMOC-Cl solu-
analysis time, improvement in intra-injection reproducibility, tions. They proposed a strategy of inverting the columns during
reduction of matrix and ionization effects and enabling higher the wash step to avoid chromatographic deterioration. Bahrami
sensitivity. From the published literature, it was evident that sev- et al. (2005) reported the removal of excess FMOC-Cl by addition
eral reported methods include another macrolide as an IS (Nirogi of amino acid, as excess FMOC-Cl leads to band broadening for
et al., 2005; Barrett et al., 2005; Bahrami et al., 2005; Chen et al., the analytes of interest. Most authors opted for acidic mobile
2006, 2007; Xue-Min et al., 2007; Liu et al., 2007; Yuzuak et al., phase in combination with a C18 stationary phase to elute the
2007; Xu et al., 2008; Shen et al., 2010; Choemunng and Na- analytes early, as at higher pH the retention of azithromycin is
Bangchang, 2010). Among these, roxithromycin was used by extended (Chen et al., 2007). Chen et al. (2007) described the
several authors (Barrett et al., 2005; Xue-Min et al., 2007; Chen selection of buffer strength, as lower buffer concentrations led
et al., 2007; Yuzuak et al., 2007; Xu et al., 2008; Shen et al., to poor peak shapes. A similar concern was mentioned by Barrett
2010; Choemunng and Na-Bangchang, 2010) for mass spectrom- et al. (2005), where increasing the buffer strength gave good peak
etry-based methods. Fouda and Schneider (1995) used deuter- symmetry but prolonged retention, and hence they had to opti-
ated azithromycin-d3 (synthesized by reductive alkylation of mize a three-component mobile phase containing acetonitrile,
secondary amine) as an IS. In HPLC-based bioanalytical methods, methanol and buffer for the desired chromatography.
an analogue structurally close to either azithromycin (Wilms
et al., 2005; Bahrami et al., 2005; Torano and Guchelaar, 1998)
Detection
or amantadine (Bahrami and Mohammadi, 2006) was used as
an IS. The HPLC method reported by Ebrahimzadeh et al. The sensitivity of analytical assay depends upon selective isola-
(2010) did not use any IS. tion of analyte from biological constituents into a solvent media
compatible with the detector and analyte nature/response to
the detection method. For the former, we need to develop an
Chromatography
extraction method considering the analyte physiochemical
Chromatography is a physical method used to separate analyte properties, composition and complexity of the intended biolog-
(s) to be measured from other molecules in the mixture based ical matrix and compatibility of the final extract with the
on the differential partitioning effect between the mobile phase required mode of detection. For latter, we need to consider
and stationary phases. Typically the mixture of various compo- analyte properties/response in the detection method.
nents to be separated is dissolved in a mobile phase and passed Azithromycin lacks the presence of any significant chromophore
through the stationary phase. Reverse-phase chromatography in the chemical structure and requires the adoption of either
has been widely chosen, in which the liquid mobile phase is derivatization (for fluorescence detection) or mass spectrometry.
water–buffer (aqueous component) combined with an organic Azithromycin does not have any specific chromophore. Elec-
solvent such as methanol or acetonitrile and the stationary trochemical detection (using amperometric and coulometric de-
phase surface is nonpolar or hydrocarbon-like. The choice of re- tection) and pre-column derivatization followed by fluorescence
verse-phase systems was obvious since azithromycin, a large detection have been applied in order to achieve the required
molecule (i.e. large carbon chain, C38H72N2O12) with high logP quantification levels. Torano and Guchelaar (1998) have pub-
(i.e. 4.02) and polar surface area of 180 Å2, would evidently have lished detailed experiments to attain the optimum derivatization
good retention on conventional reverse-phase columns. C18 sil- conditions. As per their observations, the conditions are highly
ica has the highest degree of hydrophobicity, interacts with dependent on pH, reaction temperature and time. A pH range
the widest range of compounds and the interactions are gener- of 6.5–7.5 was found to be most suitable for derivatization,
ally more pronounced. In quantitation of azithromycin most of with 40
C for 40 min reaction condition. Also a reaction
the researchers have also used octadecyl silane as stationary mixture of 1:3 to 1:4 (v/v) water–acetonitrile ratio was most suit-
phase both on HPLC (Torano and Guchelaar, 1998; Wilms et al., able for reaction. These facts (reaction condition) are important
2005; Ebrahimzadeh et al., 2010) and LC-MS/MS (Fouda and as azithromycin and FMOC-Cl have low water solubility. An
Schneider, 1995; Nirogi et al., 2005; Chen et al., 2006; Xue-Min excess of acetonitrile in reaction mixture resulted in precipita-
et al., 2007; Chen et al., 2007; Yuzuak et al., 2007; Xu et al., tion of phosphate buffer, which was necessary for the removal
2008; Choemunng and Na-Bangchang, 2010). Other columns of hydrogen ions from hydroxyl groups of macrolide. Also, as
reported in the literature are C8 (Barrett et al., 2005; Bahrami mentioned earlier, an excess of FMOC-Cl also led to deterioration
and Mohammadi, 2006), cyno (Liu et al., 2007) and phenyl in chromatographic conditions owing to interaction with free
(Bahrami et al., 2005) columns. Barrett et al. (2005) specifically silanol groups. Fluorescence was attained at an excitation
mentioned their observation of superior elution of azithromycin wavelength of 255–267 nm with an emission wavelength of
on C8 phase vs other tested materials. As per their observation, 315–317 nm.
symmetrical peaks were not attained on C18 stationary phase Azithromycin contains two nitrogen atoms, one locating
owing to high lipophilicity of azithromycin. Xu et al. (2008) the 15-membered ring and the other attaching to the sugar
mentioned the addition of 0.5% triethylamine solution to elimi- (desosamine). These two nitrogen atoms could be simulta-
nate tailing for analytes on the C18 column. Some authors neously protonated easily during the ionization process in acidic
(Torano and Guchelaar, 1998; Bahrami et al., 2005; Wilms et al., conditions. A similar observation was shared by Chen et al.
2005) opted for fluorometric detection, which requires the (2006) in their publication. Base peak intensity was higher for
addition of fluorescent moiety by treatment of extract residue positive mode, as evident in experiments reported by Nirogi
with 9-fluroenylmethyloxy carbonyl chloride (FMOC-Cl). Torano et al. (2005), Chen et al. (2006) and Liu et al. (2007). Liu et al.
and Guchelaar (1998) specifically mentioned the possible inter- (2007) reported a higher signal for [M + H]+, that is, m/z 750,
action of FMOC-Cl with free silanol groups of column material while Chen et al. (2006) reported a higher intensity for
in front, leading to band broadening, front peak tailing and even [M + 2H]+ m/z 375, probably owing to different analytical

Biomed. Chromatogr. 2013 Copyright © 2013 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/bmc
 K. Sharma and R. Mullangi

conditions. Upon fragmentation, sequential loss of 158 and 157 Bright GM, Nagel AA, Bordner J, Desai KA, Dibrino JN, Nowakowska J,
Da produced two abundant product ions at m/z 591 and 434, re- Vincent L, Watrous RM, Sciavolino FC, English AR, Retsema JA,
Anderson MR, Brennan LA, Borovoy RJ, Cimochowski CR, Faiella JA,
spectively. These product ions are generated by subsequent loss Girard AE, Girard D, Herbert C, Manousos M and Mason R. Synthesis,
of cladinose and desosamine moieties. With the cleavage of the in vitro and in vivo activity of novel 9-deoxo-9a-AZA-9a-
glycosidic bond, the oxygen atom is left on the lactone ring as a homoerythromycin A derivatives: a new class of macrolide antibiotics,
hydroxyl function. The desosamine moiety of azithromycin con- the azalides. Journal of Antibiotics 1988; 41: 1029–1047.
tains an ionizable amino function that results in m/z 158 in the Chen BM, Liang YZ, Chen X, Liu SG, Deng FL and Zhou P. Quantitative deter-
mination of azithromycin in human plasma by liquid chromatography–
MS-MS product ion spectrum starting from the ionized molecule mass spectrometry and its application in a bioequivalence study. Journal
or in the MS3 product ion spectrum of the product ion resulting of Pharmaceutical and Biomedical Analysis 2006; 42: 480–487.
from the loss of the cladinose moiety. In SIM mode, there could Chen L, Qin F, Ma Y and Li F. Quantitative determination of azithromycin
be early eluting interferences and a strategy of higher retention in human plasma by ultra performance liquid chromatography–
electrospray ionization mass spectrometry and its application in a phar-
of analytes has to be adopted to resolve the desired peaks from macokinetic study. Journal of Chromatography B 2007; 855: 255–261.
that region (Liu et al., 2007). In MRM mode, almost all studies Choemunng A and Na-Bangchang K. An alternative liquid chromatography
have opted for product ion m/z 591 (resulting from loss of mass spectrometric method for the determination of azithromycin in
cladinose moiety; Nirogi et al., 2005; Barrett et al., 2005; Chen human plasma and its application to pharmacokinetic study of patients
et al., 2007; Yuzuak et al., 2007; Shen et al., 2010). Shen et al. with malaria. The Southeast Asian Journal of Tropical Medicine and Public
Health 2010; 41: 1048–1058.
(2010) selected dual protonated [M + 2H]+ as the precursor ion Dunn CJ and Barradell LB. Azithromycin. A review of its pharmacological
with m/z 591as product ion. The separation and ionization of properties and use as 3-day therapy in respiratory tract infections.
azithromycin and IS used in the bioanalytical method were af- Drugs 1996; 51: 483–505.
fected by the composition of the mobile phase. In most of the Ebrahimzadeh H, Yamini Y, Ara KM, Kamarei F and Khalighi-Sigaroodi F.
Determination of azithromycin in biological samples by LLLME
reported studies ammonium acetate was used in the mobile combined with LC. Chromatographia 2010; 72: 731–735.
phase to supply the ionic strength as in the buffers with weaker Farmer S, Li Z and Hancock REW. Influence of outer membrane mutations on
strength the peak shapes were not satisfactory. susceptibility of Escherichia coli to the dibasic marcrolide azithromycin.
The Journal of Antimicrobial Chemotherapy 1992; 29: 27–33.
Fouda HB and Schneider RP. Quantitative determination of antibiotic
azithromycin in human serum by high-performance liquid chroma-
Discussion tography (HPLC)–atmospheric pressure chemical ionization mass
Azithromycin is one of the best-selling marcolide antibiotics in spectrometry: correlation with standard HPLC-electrochemical
method. Therapeutic Drug Monitoring 1995; 17: 179–183.
world. Its unique pharmacokinetic and improved spectrum of Foulds G, Shepard RM and Johnson RB. The Journal of Antimicrobial
activity have attracted many researchers to explore highly sensi- Chemotherapy 1990; 25: 73–82.
tive bioanalytical methods in biological methods using HPLC Lalak NJ and Morris DL. Azithromycin clinical pharmacokinetics. Clinical
and LC-MS/MS. As azithromycin does not have a specific UV Pharmacokinetics 1993; 25: 370–374.
Liu F, Xu Y, Huang J, Gao S and Guo Q. Sensitive liquid chromatography/
chromophore, all the reported HPLC methods were based on
mass spectrometry assay for the quantification of azithromycin in
fluorometric detection following derivatization of azithromycin, human plasma. Biomedical Chromatography 2007; 21: 1272–1278.
and the attained LLOQs (10–100 ng/mL) were comparable to Nirogi RVS, Kandikere VN, Shukla M, Mudigonda K, Maurya S, Boosi R and
mass spectrometric methods (1–10 ng/mL). However, as the Yerramilli A. Sensitive and selective liquid-chromatography–tandem
derivatization process is labour-intensive and associated with re- mass spectrometry method for the quantification of azithromycin in
human plasma. Analytica Chimica Acta 2005; 553: 1–8.
producibility problems, LC-MS/MS is becoming the method of Peters DH, Friedel HA and McTavish D. Azithromycin. A review of its
choice for quantitation of azithromycin in various biological antimicrobial activity, pharmacokinetic properties and clinical efficacy.
matrices to determine its pharmacokinetic profile for a better Drugs 1992; 576: 358–362.
understanding of its mechanism of action and ensuring a more Schentag JJ and Ballow CH. Tissue directed pharmacokinetics. American
Journal of Medicine 1991; 91: 5S–11S.
efficient therapeutic application. Reported LC-MS or LC-MS/MS Shen Y, Yin C, Su M and Tu J. Rapid, sensitive and selective liquid chroma-
methods allowed the quantitation of azithromycin up to ≥120 tography–tandem mass spectrometry (LC-MS/MS) method for the
h post-dosing. The LC-MS/MS method has advantages in provid- quantitation of topically applied azithromycin in rabbit conjunctiva tis-
ing a shorter analytical run time, high specificity, excellent sensi- sue. Journal of Pharmaceutical and Biomedical Analysis 2010; 52: 99–104.
tivity, a much lower limit of quantification and robustness for Shepard RM and Falkner FC. Pharmacokinetics of azithromycin in rats
and dogs. The Journal of Antimicrobial Chemotherapy 1990; 25: 49–60.
drug measurement; however, the choice of IS, the organic Torano JS and Guchelaar HJ. Quantitative determination of the macorlide
solvents (for sample processing) and the matrix effect should antibiotics erythromycin, roxithromycin, azithromycin and clarithromycin
be evaluated carefully during method validation. in human serum by high-performance liquid chromatography using pre-
column derivatization with 9-fluorenylmethyloxycarobnylchloride and
fluorescence detection. Journal of Chromatography B 1998; 720: 89–98.
References Wilms E, Trumpie H, Veenendal W and Touw D. Quantitative determina-
tion of azithromycin in plasma, blood and isolated neutrophils by
Bahrami G and Mohammadi B. A new on-line, in-tube pre-column liquid chromatography using pre-column derivatization with 9-
derivatization technique for high performance liquid chromato- fluorenylmethyloxycarbonylchloride and fluorescence detection.
graphic determination of azithromycin in human serum. Journal of Journal of Chromatography B 2005; 814: 37–42.
Chromatography B 2006; 830: 355–358. Yuzuak N, Ozden Y, Eren S and Toptan S. Analysis of azithromycin in
Bahrami G, Mirzaeei S and Kiani A. High performance liquid chromato- human plasma by LC-MS/MS. Chromatographia 2007; 66: S115–S118.
graphic determination of azithromycin in serum using fluorescence Xu F, Zhang Z, Bian Z, Tian Y, Jiao H and Liu Y. Azithromycin quantitation
detection and its application in human pharmacokinetic studies. in human plasma by high-performance liquid chromatography
Journal of Chromatography B 2005; 820: 277–281. coupled to electrospray mass spectrometry: application to bioequiv-
Barrett B, Borek-Dohalsky V, Fejt P, Vaingatova S, Huclova J, Nemec B and alence study. Journal of Chromatographic Science 2008; 46: 479–484.
Jelinek I. Validated HPLC-MS-MS method for determination of Xue-Min Z, Jie L, Juan G, Quan-Shing Y and Wen-Yan W. Determination of
azithromycin in human plasma. Analytical and Bioanalytical Chemistry azithromycin in human plasma by LC-MS-MS and its pharmacokinet-
2005; 383: 210–217. ics. Pharmazie 2007; 62: 255–257.

wileyonlinelibrary.com/journal/bmc Copyright © 2013 John Wiley & Sons, Ltd. Biomed. Chromatogr. 2013