J. Sep. Sci.

2011, 34, 2223–2231 2223

Nafiz O. Can Research Article

Department of Analytical
Chemistry, Faculty of Pharmacy, HPLC determination of cefprozil in tablets
Anadolu University, Eskisehir,
Turkey using monolithic and C18 silica columns
Cefprozil (CPZ) is a second-generation semi-synthetic cephalosporin antibiotic that
Received February 28, 2011
commonly exists as the mixture of Z and E diastereoisomers, at the ratio of approximately
Revised March 20, 2011
Accepted March 28, 2011 9:1. A novel reversed-phase HPLC method for the determination of CPZ in tablets was
described. The separation of CPZ diastereoisomers and caffeine (internal standard) was
carried out by applying the same analytical and instrumental conditions on two stationary
phases, which have different surface chemistries. The columns used in the study were
monolithic silica Merck Chromolith Performance RP-18e and conventional C18 silica
Phenomenex Synergi Hydro RP columns. In total, 10 mL aliquots of samples were injected
into the system and eluted using water–acetonitrile (90:10, v/v) solution, which was
pumped through the column at a flow rate of 1.0 mL/min. The analyte peaks were
detected at 200 nm using diode array detector with high specificity. CPZ diastereoisomers
and caffeine were measured within 13 min using the C18 column, whereas o5 min was
required for the monolithic one. Validation studies were performed according to official
recommendations. Value of a monolithic column for the assay of diastereoisomers in
pharmaceutical tablets was evaluated for the first time and found as a powerful alternative
to highly efficient C18 columns.

Keywords: Cefprozil / Diastereoisomer separation / Monolithic silica column /

Pharmaceutical application
DOI 10.1002/jssc.201100170

1 Introduction and isosilybin [12] are reported as the applications that

monolithic materials were used as stationary phases in the
Diastereoisomers are defined as the stereoisomers not analysis of these compounds. In addition, Schulte and
related as mirror images [1]. Separation of diastereo- Dingenen [13] have demonstrated the use of monolithic
isomers is not a frequent application in analytical field materials on preparative chromatography of diastereo-
due to limited number of compounds that have multiple isomers. On the other hand, assay of therapeutic
stereocenters. Diastereoisomers naturally show similar compounds or excipients in pharmaceutical preparations
physicochemical profiles and from a point of view, their were successfully carried out using columns with mono-
chromatographic analysis is informative especially for the lithic packings [14–18]; however, no study exists on the
assessment of separation efficiencies of novel stationary determination of pharmaceutical diastereoisomers. Thus, to
phases. There are only a few studies in the literature on the investigate the suitability of monolithic sorbents in the assay
applicability of monolithic sorbents for the separation of of mentioned compounds and to increase the number of
diastereoisomers although several successful examples were examples, new studies on this issue are required.
demonstrated utilizing conventional silica particle-based Cefprozil (CPZ) ((6R,7R)-7-[[(2R)-2-amino-2-(4-hydro-
stationary phases [2–6]. Separation of diasteroisomeric xyphenyl)acetyl]amino]-8-oxo-3-[(E)-prop-1-enyl]-5-thia-1-
oligonucleotide adducts [7], oligonucleotide phospheorothio- azabicyclo[4.2.0]oct-2-ene-2-carboxylic acid) is a second-
ate diastereoisomers [8, 9], chiral phosphinic acid pseudo- generation semi-synthetic cephalosporin that exists as the
dipeptides [10], allethrin [11], and diastereomers of silybin mixture of Z and E diastereoisomers, at the ratio of
approximately 9:1. The chemical structure of CPZ is shown
in Fig. 1; its molecular formula is C18H19N3O5S and it has a
Correspondence: Dr. Nafiz O. Can, Department of Analytical molecular weight of 389.43 g/mol. The exact mechanism by
Chemistry, Faculty of Pharmacy, Anadolu University, Yunus-
which CPZ exerts its bactericidal action is the inhibition of
emre Campus, Eskisehir 26470, Turkey
E-mail: nafizoc@anadolu.edu.tr the synthesis of bacterial cell wall.
Fax: 190-222-3350750 LC methods were preferred in a number of analytical
and bioanalytical studies for the determination of CPZ
Abbreviations: CPZ, cefprozil; ICH, International Conference diastereoisomers and related compounds. Being mostly
on Harmonization; USP, the United States Pharmacopeia focused on bioavailability and bioequivalency, the reported

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
2224 N. O. Can J. Sep. Sci. 2011, 34, 2223–2231

Figure 1. Chemical structure of CPZ.

assays describe quantitative determination of CPZ in was carried out. To improve robustness of the proposed
biological fluids such as breast milk [19], plasma [20], and method, caffeine was utilized as internal standard. Analytical
middle ear fluid [21, 22], and in some pharmaceutical procedure, on the whole, is advantageous for method transfer
preparations such as tablets [23, 24] and oral suspensions between laboratories due to its simplicity. At the end of this
[25]. Within the methods used to determine CPZ, LC with study, separation and quantification of CPZ diastereoisomers
MS [26–28] or UV [19, 20, 22, 24–25, 28–31] detection was in a pharmaceutical tablet formulation were achieved using a
popular. Among these, MS detection shows its superiority monolithic silica column for the first time.
over UV detection in terms of specificity and detection
limits because of unique capabilities such as multiple and
selected reaction monitoring. 2 Materials and methods
Utilization of state-of-the-art instrumentation or equip-
ment is a key point in today’s LC applications to improve 2.1 Chemicals
separations and lower total analysis costs. In this sense, one
of the basic ways to improve separation efficiency is to use CPZ certified reference material (at 99.9% purity and USP
novel stationary phases, such as sub-2 mm particles, mono- compliant) was the product of Ranbaxy Fine Chemicals,
lithic or hybrid materials. The columns packed with sub- (Mumbai, India) and kindly supplied as a gift from Fargem
2 mm particles have some advantages, e.g., low solvent A.S. (Duzce, Turkey). Reference standard of anhydrous
consumption and unequally high efficiency in shorter caffeine (>99% purity) and HPLC-grade acetonitrile was
analysis time; however, specially designed instrumentation, purchased from Sigma-Aldrich Lab GmbH (Seelze,
known as ultra performance or ultra fast, etc., liquid chro- Germany). HPLC-grade water (with total organic carbon
matographs, is required for acceptable separations. On the o5 ppb and resistivity of at least 18.2 MOhm/cm) was
other hand, pressure problems are easily overcome by using produced in our laboratory using Milli-Q Synthesis A10
the columns packed with monolithic sorbents without system from Millipore SAS (Molsheim, France). Tablet
comprising efficiency, thanks to unique bimodal structure. formulation of CPZ (labeled as Prefixs 500 mg Film Tablet)
Besides, versatile synthesis mechanisms of monolithic mate- was provided by Nobel Ilac- A.S. (Duzce, Turkey) with label
rials are open to modifications and even chiral compounds claim of 500 mg CPZ.
can be easily separated via chemical derivatization with
remarkable pressure drops [32–34]. Briefly, 3 mL/min or
faster mobile phase transfer through monolithic columns is a 2.2 Instrumentation
common application in pharmaceutical analyses with ordin-
ary LC systems [35–38]. Hybrid or polymeric materials are Analyses were performed using a Prominence series of
another choice, especially for wide-range pH or specialized HPLC system from Shimadzu (Japan), which was
separations. The columns utilized in this study, Synergi composed of DGU-20A5 online degasser, LC-20AT tandem
Hydro RP (Phenomenex) and Chromolith Performance dual plunger pump equipped with low-pressure gradient
RP-18e (Merck), were selected according to the basis described unit, SIL-20A auto-sampler, CTO-10ASVP column oven,
above. Literature search revealed that there is no previous CBM-20A communications bus module, and SPD-M20A
study in which these columns were used in CPZ assays. diode array detector. The system was controlled and
A single LC method which was carried out on two chromatograms were integrated by using LCSolutions 1.11
different types of stationary phases was developed, validated SP1 software (Japan). An Explorer E12140 model analytical
according to International Conference on Harmonization balance from Ohaus (Switzerland), an RK 100 H model
(ICH) [39], and the United States Pharmacopeia (USP) [40] ultrasonic bath from Bandelin (Germany), an RO 15
recommendations, and its application to CPZ tablet assay multipoint stirrer from IKA (Germany), and a 1–6 model

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
J. Sep. Sci. 2011, 34, 2223–2231 Liquid Chromatography 2225

centrifuge from Sigma (Germany) were used in the 2.6 Method validation
preparation of samples and solutions.
2.6.1 System suitability testing

2.3 Chromatographic parameters As a crucial part of HPLC method development, system

suitability parameters were investigated to prove the
Water–acetonitrile solution (90:10, v/v) was used as mobile performance of HPLC instrumentation. Column efficiency
phase. Chromatographic analyses were performed on two (N), capacity factor (k), resolution between CPZ-Z, CPZ-E,
different columns: A conventional C18 column (Synergi and caffeine peaks (Rs), asymmetry factor (As), tailing factor
Hydro RP, 150 mm  4.6 mm id, 4 mm particle with 80 pore (T), and separation factor (a) were calculated according to
size from Phenomenex) and a monolithic silica column USP methods.
(Chromolith Performance RP-18e, 100 mm  4.6 mm id,
2 mm macropore and 13 nm micropore size from Merck 2.6.2 Specificity
KGaA). The column oven temperature was set at 251C, flow
rate was 1.0 mL/min, and injection volume was 10 mL. The In ICH Q2(R1) guideline, use of a second well-characterized
diode array detector was set at 200 nm wavelength and real- procedure to compare the test results of samples possible to
time spectra were recorded between 190 and 380 nm at contain impurities or degradation products is recom-
40 Hz data-sampling frequency. All solutions were filtered mended. Accordingly, possible effects of CPZ impurities
through nonsterile polyvinylidene fluoride syringe filters and pharmaceutical excipients to the assay could be
(13 mm id, 0.2 mm pore size, from Orange Scientific, assessed by comparing two chromatograms acquired by
Belgium) prior to injection. using both HPLC columns. The results obtained by using a
column were used as the other one’s comparative to
determine specificity. The chromatograms and peaks of
2.4 Preparation of reference standards interest were examined for susceptible signals to confirm
that a positive or negative response was not obtained.
An accurately weighed 20.0 mg CPZ was transferred to a Besides, peak purities were checked using diode array
100-mL volumetric flask, acetonitrile/water solution (10:90, detector to examine the analyte and internal standard peaks
v/v) was added to dissolve the substance, and the content were not attributable to more than one compound.
was diluted to volume with the same solvent. Resulting
solution was diluted to 10 mg/mL with the same solvent. For 2.6.3 Linearity and range
caffeine solution, an accurately weighed 10.0 mg anhydrous
caffeine standard was transferred to a 100-mL volumetric The linearity plot was prepared covering six concentration
flask, acetonitrile/water solution (10:90, v/v) was added to levels of CPZ (0.1, 0.5, 1.0, 1.5, 2.0, and 2.5 mg/mL). The
dissolve the substance, and the content of the flask was range of the method corresponded to 10–250% of estimated
diluted to volume with the same solvent to obtain a concentration of the test solutions. Each solution was
concentration of 100 mg/mL caffeine. The stock solutions analyzed using both of the columns, injected triplicate,
were found to be stable at least 7 days when kept at 51C away and average values were accepted as representative.
from daylight. Linearity was evaluated by linear regression analysis with
intra- and interday replicates.

2.5 Tablet assay preparation 2.6.4 Accuracy

Thirty tablets were used for the assay. Each tablet was Recovery experiments were performed in order to determine
transferred to a separate 100-mL volumetric flask and 50 mL assay accuracy. Preanalyzed tablet samples were spiked with
of acetonitrile/water solution (10:90, v/v) was added. The known amount of CPZ solutions at three different levels
content of the flask was stirred using a magnetic stirrer for corresponding to 50, 100, and 150% of assay concentration.
20 min until dissolution. Subsequently, the flask was spiked Three sets were prepared for each concentration level.
with internal standard, filled up to volume with the same Spiked samples were reanalyzed and mean recovery was
solvent and content was sonicated for 10 min. Ten milliliter calculated.
portion of the resulting mixture was transferred into a
polypropylene centrifuge tube and centrifuged at 6000 rpm 2.6.5 Precision
for 10 min to precipitate insoluble excipients. Three
milliliter of the supernatant was taken and filtered through Precision studies included intraday and interday (inter-
a 0.20-mm syringe filter. The estimated CPZ concentration mediate) precision experiments. Precision was evaluated by
of final solution was 5 mg/mL and this solution was diluted analyzing the standard solution for eight times using
by 1:5000 v/v to the working range using initial solvent proposed methods on the same and three consecutive days.
mixture. The standard solution was chosen to be at 100% of the

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
2226 N. O. Can J. Sep. Sci. 2011, 34, 2223–2231

concentration of test solution. Results were analyzed specified wavelength, which allows lower noise and back
statistically including mean, standard deviation, RSD%, pressure, and better mass transfer. Signal acquisition
and confidence interval at 95% confidence level. frequency of diode array detector was set at 40 Hz to be
able to record more signals that are representative. After
2.6.6 Limits of detection and quantitation initial studies, optimum flow rate was found as 1.0 mL/min.
As an advantage of highly porous structure of monolithic
The limit of detection (LOD) and limit of quantitation (LOQ) stationary phases, the back pressure did not exceed 30 bar in
were determined according to ICH recommendations, all tests, whereas it was about 120 bar for Synergi Hydro RP.
based on the standard deviation of the response and the Lower system pressure was advantageous for longevity of
slope. For LOD, the ratio of the standard deviation of y- pump seals, which are mechanically destructed because of
intercepts of regression lines (s) to the slope was multiplied high pressure. The effect of column temperature on the
by 3.3, whereas it was 10 for LOQ. s and S were calculated retention of analytes was also investigated by changing
from pooled data obtained from linearity experiments. In column temperature between 25 and 401C and 251C was
addition to calculation method, evaluation of LOD and LOQ selected as a conventional temperature.
based on signal-to-noise ratio was also used to confirm To enhance robustness of the developed method, an
values. Signal-to-noise ratio was taken as 3.3 for LOD and 10 internal standard was utilized. Absorptivity, solubility, and
for LOQ. hydrophobicity properties of some compounds were exam-
ined; caffeine was considered as a suitable candidate
2.6.7 Robustness because it is easy to find and has chromatographic proper-
ties similar to CPZ (caffeine has log p value of 0.07, water
To determine robustness, some instrumental conditions solubility of 21.7 g/mL, and UV absorbance maximum
which have major effect on the retention were purposely about 210 nm). Initial studies revealed that CPZ and
altered by 2% one at a time. Robustness of the method was caffeine could be detected simultaneously, due to similar
evaluated by analyzing test solutions under slight but retention and signal intensity. Normalization procedure was
adequate variations, such as 70.02 mL/min change in flow used in order to minimize the external or internal factors
rate, 70.2% change in acetonitrile ratio in the mobile that affect the analysis, such as variation of mobile phase
phase, and 71.81C change in the column temperature. composition, pH, etc.
Relative change in peak intensity, retention time, and Finally, in the light of above-mentioned findings, assay
number of theoretical plates was assessed, and test results of conditions were concluded as follows: Mobile phase was
both columns were compared with each other to provide an composed of water–acetonitrile (90:10, v/v) and pumped at a
additional opinion on robustness. flow rate of 1.0 mL/min; column temperature was set to
251C and CPZ diastereoisomers and caffeine were detected
at 200 nm. Analysis of CPZ and caffeine standards at the
3 Results and discussion specified conditions using Chromolith Performance RP-18e
and Synergi Hydro RP columns produced the chromato-
In accordance with the aim underlined in Section 1, grams shown in Fig. 2. As shown in Fig. 2, peaks of CPZ
selection of appropriate analytical conditions was the major diastereoisomer and caffeine were well separated with high
step of the method development. Acetonitrile and methanol resolution. Elution of caffeine just after CPZ was preferred,
were initially tested as organic modifiers in the mobile because when the internal standard shows the correct
phase, as they are mostly found in quality-control studies. retention time and area response, it means that all peaks
Low hydrophobic (log p 5 0.69) and high water-soluble eluted earlier also were eluted under the proper chromato-
(3.47 g/L) character of CPZ resulted in occurrence of graphic conditions. The peaks eluted after the internal
detection under 20 min when organic solvent ratio was standard may be depending on equipment errors that would
kept between 5 and 15%. On the other hand, highest signal not be apparent from an examination of the internal stan-
intensity of the analyte was found to be about 20074 nm in dard peak [41]. When using Chromolith Performance RP-
a supplementary UV-spectrophotometric study (relevant 18e column, CPZ Z and E isomers and caffeine peaks were
data were not given here). Neither pH buffers nor peak eluted at 2.270.1, 2.870.1, and 4.270.1 min, whereas
enhancers such as triethylamine or octylamine were 4.170.1, 6.270.1, and 12.270.1 min were necessary for
intended to use due to acceptable peak shape for the Synergi Hydro RP. It is clear that despite its relatively lower
determination and closeness of the detection wavelength to hydrophobicity, the monolithic column adequately sepa-
far UV region (it is well known that many reagents used for rated the diastereoisomers like C18 one. Synergi Hydro RP is
these purposes have UV absorbance and thus increase noise packed with high-density (475 m2/g) C18-bonded phase,
resulting significant increase in LODs and LOQs). Minimal which was endcapped with a proprietary polar group to
band broadening was observed for both of the columns that provide high retention of polar compounds even under
allowed detection of the compounds with adequate sharp- 100% aqueous conditions, whereas bimodal highly porous
ness and peak symmetry. Acetonitrile was used instead of monolithic silica rod is placed in Chromolith Performance
methanol because of lower viscosity and absorbance at RP-18e. When evaluated according to the hydrophobic

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
J. Sep. Sci. 2011, 34, 2223–2231 Liquid Chromatography 2227

Figure 2. Typical chromatograms of CPZ (1.0 mg/mL) and caffeine (1.0 mg/mL) standards using Chromolith Performance RP-18e and
Synergi Hydro RP columns.

Table 1. System suitability values calculated for CPZ-Z, CPZ-E, and IS peaks at the specified conditions

Parameter Chromolith Performance RP-18e Synergi Hydro RP Recommended value

CPZ-Z CPZ-E IS CPZ-Z CPZ-E IS

Retention time (min) 2.21 2.85 4.21 4.12 6.21 12.2 N/A
Asymmetry (As) 1.26 1.42 1.39 1.05 0.95 1.23 0.95oxo1.5
Tailing (T) 1.19 1.28 1.34 1.01 0.91 1.28 o2
Selectivity (a) 1.32 1.82 1.59 2.05 41
Capacity factor (k) 9.34 12.3 22.48 5.65 9.01 18.5 42
Resolution (Rs) 3.78 9.95 6.85 12.92 42
Theoretical plates (N)/m 27 970 37 454 64 012 26 380 34 773 47 786 >2000

subtraction model of reversed-phase column selectivity [42], etc. Results of the system suitability tests were in agreement
Synergi Hydro RP was shown to have one of the highest with the specified values. Relevant data are summarized in
ligand concentrations among conventional C18 phases with a Table 1.
value of 4.05 mmol/m2 and pore diameter of 80 Å [43]; on the Following the system suitability tests, proposed method
other hand, Chromolith Performance RP-18e has ligand was validated for both of the columns according to official
concentration of 3.6 mmol/m2 and pore diameter of 130 Å recommendations. The specificity of the method was
[44]. In fact, these findings underline that the C18 column determined by checking signals of possible interferences in
used in this study is more hydrophobic than the monolithic the assay solutions. Peaks of the claimed pharmaceutical
one. However, monolithic column showed slightly better excipients or possible impurities were not observed close to
separation efficiency when compared with C18 column by CPZ and caffeine peaks. Besides, peak purities of CPZ and
means of number of theoretical plates per meter (Table 1). caffeine were checked and results confirmed that analyte
As an integral part of the analytical procedure, system peaks were not attributable to more than one compound in
suitability was checked prior to injection of standards to all of the analyzed quality control samples. Peak purity
ensure that the system is performing adequately for the values were found as 0.999999, 1.000000, and 1.000000 for
assay. These parameters were evaluated according to the CPZ-Z, CPZ-E, and IS, respectively.
recommendations of ICH [39] and USP [40], including The linearity was evaluated by linear regression analysis
number of theoretical plates, resolution, selectivity factor, using six different concentrations of CPZ. The ratio of the

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
2228 N. O. Can J. Sep. Sci. 2011, 34, 2223–2231

Table 2. Results of linearity and precision studies

Validation parameter Chromolith Performance RP-18e Synergi Hydro RP

Linearity range (mg/mL) 0.1–2.5 0.1–2.5

Slope7SD (intraday, n 5 6) 0.714570.001375 0.685670.002568
Intercept7SD (intraday, n 5 6) 0.0135270.002082 0.0853970.003889
Regression coefficient (intraday, n 5 6) 0.9999 0.9999
Slope7SD (interday, n 5 18) 0.712470.004332 0.678170.004959
Intercept7SD (interday, n 5 18) 0.0184270.006560 0.100670.007510
Regression coefficient (interday, n 5 18) 0.9999 0.9998
Repeatability (intraday, mean, n 5 8) 0.7492 0.7790
Repeatability (intraday, RSD%, n 5 8) 1.65 1.79
Repeatability (intraday, standard error, n 5 8) 0.004368 0.004654
Repeatability (intraday, CIa), n 5 8) 0.0104 0.0107
Repeatability (interday, mean, n 5 24) 0.7484 0.7766
Repeatability (interday, RSD%, n 5 24) 1.80 1.85
Repeatability (interday, standard error, n 5 24) 0.002745 0.002934
Repeatability (interday, CIa), n 5 24) 0.0057 0.0061

a) CI, confidence interval at 95% confidence level.

sum peak area of CPZ peaks to caffeine peak versus response (s) and the slope. The ratio of the standard devia-
concentration was used for the quantitation. Signals were tion of y-intercepts of regression lines (s) to the slope was
found to be directly proportional to the concentrations of multiplied by 3.3 for LOD, whereas it was 10 for LOQ. As an
calibration solutions and obeying Beer’s law in the range of additional confirmation, these values were recalculated based
0.1–2.5 mg/mL. The regression equations were y 5 0.7124x1 on signal-to-noise ratio, which was equaled to 3.3 for LOD
0.01842 (r2 5 0.9999) for Chromolith Performance RP-18e and 10 for LOQ. As a result, LOD and LOQ of the method for
and y 5 0.6781x10.1006 (r2 5 0.9998) for Synergi Hydro RP Chromolith Performance RP-18e column were calculated to
columns, where x is the concentration in mg/mL and y is the be 0.030 and 0.092 mg/mL, whereas they were 0.032 and
peak area ratio. Statistical data related to linearity studies are 0.096 mg/mL for Synergi Hydro RP column, respectively.
summarized in Table 2. These values were found satisfying when compared with the
Precision of the methods was investigated at two steps: previous studies on CPZ. The explanation of this improve-
intraday (repeatability) and interday (intermediate) preci- ment was associated to the very low noise of the assays (about
sion. Method transfer from Synergi Hydro RP to Chromo- 0.25–0.50 mAU), which was interpreted as a total result of
lith Performance RP-18e column did not cause any change acetonitrile usage in mobile phase, high absorptivity of CPZ
in precision results; thus, reproducibility issues were not at 200 nm and avoiding use of buffers to maintain pH. In
investigated separately since method was found to be highly addition, literature search shows that 245 nm [31], 254 nm
precise for both columns. With a general statistical [24], and 280 nm [19, 20, 22, 25, 28–30] wavelengths were
approach, mean, standard deviation, RSD%, and confidence mainly preferred in UV-based detections of CPZ. Method
interval were calculated. The overall RSD% of the methods development studies exhibited that CPZ absorbs UV light
was between 1.65 and 1.79 for repeatability and 1.80 and about 79% less at 245 nm, 81% less at 254 nm, and 70% less
1.85 for intermediate precision studies. Precision data are at 280 nm when compared with 200 nm (Fig. 3). High signal
summarized in Table 2. intensity and low determination limits were possibly asso-
Accuracy was interpreted as the closeness of assay ciated to high absorptivity of CPZ at 200 nm. Although use of
results to the values that are accepted as conventional true this wavelength makes the method vulnerable to systematical
values [39]. The proposed methods were tested for accuracy errors, effect of any impurity and degradation product or
in the working range. Standard addition method was degradation of CPZ can be monitored more accurately as a
employed for recovery experiments, making nine indepen- result of increased sensitivity.
dent determinations over three different concentrations Typical sources of variation in HPLC analyses includes
covering the range. Preanalyzed tablet samples were flow rate, mobile phase composition, and ambient
reanalyzed after spiking with known amounts of CPZ temperature, which frequently affect assay results in
solutions. The mean recovery of CPZ was found between method transfer. As a measure of the developed method’s
99.4271.56 and 100.4071.58% for both columns. Data performance to remain unaffected by small but deliberate
obtained from accuracy studies are summarized in Table 3. variations in method parameters, robustness studies were
The determination limits of the proposed method for performed. For this purpose, these parameters were altered
both of the columns were presented by LOD and LOQ. As by 2% to assess the deviation from initial values: Flow rate
described before, these were determined according to ICH was changed to 0.98 and 1.02 mL/min; mobile phase was
recommendations, based on the standard deviation of the prepared to include water and acetonitrile at the ratios of

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
J. Sep. Sci. 2011, 34, 2223–2231 Liquid Chromatography 2229

Table 3. Statistical evaluation of accuracy studies performed using both columns

Added (mg/mL) Found (mg/mL)7SD Recovery (%)7SD Mean recovery (%)7SD RSD (%)

Chromolith Performance RP-18e

0.50 0.4870.02 96.6773.05 97.6772.11 2.16
1.50 1.4670.02 97.1371.40
2.50 2.4870.03 99.2071.08
Synergi Hydro RP
0.50 0.4870.01 96.6771.16 98.3672.01 2.03
1.50 1.4870.04 98.6772.65
2.50 2.4970.02 99.7370.62

Figure 3. CPZ-Z, CPZ-E, and caffeine signals

recorded at 200, 245, 254, and 280 nm
wavelengths under specified chromato-
graphic conditions.

90.2:9.8 v/v and 89.8:10.2 v/v, and column temperatures was filtered through a 0.2 mm filter (Section 2.5). In the
were changed to 25.5 and 24.51C. It was found that the assay, there were no interferences to CPZ and caffeine
results obtained using altered conditions were so close to peaks observed in the assay of Prefixs 500 mg Film Tablet
the results of initial conditions with a deviation from initial (Nobel Ilac A.S.). Typical chromatograms for both columns
values about 1% or less by means of assay, retention time, are shown in Fig. 4. For instance, separate analysis of each
and number of theoretical plate. Hence, the variations were tablet (n 5 30) and pool of whole data were considered as
accepted as not causing distortion on the methodology. more representative for statistical evaluation of assay results.
At the final stage of the experimental studies, the Each sample solution was injected into the system triplicate
proposed methods were applied for the assay of CPZ in film and average values were accepted as representative. Prism 5
tablet formulation. As a part of previously published proto- (free trial) from GraphPad Software, USA, was used for
cols, different sample preparation methods were also paired t-test, variance ratio test (F-test), and common
reported; for biological samples, protein precipitation was statistical calculations. The level of significance was 5% for
followed by centrifugation, filtration, and introduction of t-test and F-test. The p-value, which was calculated as 0.1350
sample to chromatograph, whereas solid-phase extraction for paired data (two-tailed), indicates that difference of assay
was preferred for aquatic environmental samples [28]. A results obtained by using both columns is not statistically
simple dissolution was applied in our study, which was significant. Related data are summarized in Table 4.
followed by addition of IS and sonication. Ten microliter of On the whole, the developed method was successfully
the resulting mixture was centrifuged and final supernatant applied to the real samples and represented relatively

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
2230 N. O. Can J. Sep. Sci. 2011, 34, 2223–2231

Figure 4. Assay chromatograms of CPZ tablet formulation carried out by using Chromolith Performance RP-18e and Synergi Hydro
RP columns.

Table 4. Assay results of Prefixs 500 mg Film Tablet (n 5 30) (even internal standard), chromatography instrumentation,
and other apparatus used in sample preparation, which
Chromolith Performance Synergi Hydro basically exist in all quality-control laboratories. Besides,
RP-18e RP
sophisticated procedures such as solid-phase extraction and
Mean 499.46 499.01 additional sample cleanup are not required like LC-MS/MS
Minimum 478.70 479.62 methods [28]. HPLC runs do not need buffer use and extra
Median 500.02 500.10 column washing. All of these reduced the cost per analysis
Maximum 517.18 516.58 while making the total procedure so simple and fast.
SD 10.33 9.89 Method range is better than some of the previous studies
RSD % 2.07 1.98 [20, 23, 28] and it offers practical, robust, and fast results
Standard error of mean 1.89 1.81 which can be acquired within a couple of minutes. Total
Confidence intervala) 73.86 73.69 analysis time was under 5 min for monolithic column that
p-Valueb) 0.1350
can be accepted as fast if compared with the
t0.05 valuec) 1.5377
previous studies. Analysis time was found to be better than
F0.05 valued) 0.8147
some previous methods [20, 28] and comparable to
a) Confidence interval at 95% confidence level. LC-MS/MS method reported by Manna and Valvo [24].
b) p40.05: non-significant difference between groups. Moreover, monolithic material appeared as a valuable
c) t0.05o1.699: nonsignificant difference between means of the alternative to C18.
groups at 5% level of significance. Although LOD and LOQ of described method cannot be
d) F0.05o2.66: nonsignificant difference between variances of the
evaluated as much better than the LC-MS method reported
groups at 5% level of significance.
by Wang et al. [27], our results were better than almost all of
the previously published studies including some LC-MS
good performance for both columns from analytical point of methods [19, 20, 24, 28]. LOD and LOQ were reported
view. between 0.05 and 20 mg/mL in the previous studies, whereas
proposed method allowed about 0.03 and 0.09 mg/mL,
respectively.
4 Concluding remarks Briefly, three main points can be underlined: (i) The
proposed method was adequately validated, tested for its
The method proposed herein performed better than the intended purpose, and its applicability was proven on
methods described in the previous studies at some critical different column chemistries. (ii) Results of the study
points. Main advantage was the low cost of the chemicals indicate that monolithic materials can be used as effectively

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com
J. Sep. Sci. 2011, 34, 2223–2231 Liquid Chromatography 2231

as C18 phases for the analysis of diastereoisomers. (iii) The [21] Nicolau, D. P., Sutherland, C. A., Arguedas, A.,
analytical method can be accepted as fast and robust for the Dagan, R., Pichichero, M. E., Paediatr. Drugs 2007, 9,
determination of CPZ levels in film tablets for quality- 119–123.
control laboratories. [22] Jang, C. H., Park, S. Y., Int. J. Pediatr. Otorhinolaryngol.
2003, 67, 965–968.
The author has declared no conflict of interest. [23] Rao, C. K., Tang, S. X., Zhen, L., Ji, Y. P., Yang, W. Y.,
Wang, F. C., Pharm. Care Res. 2006, 6, 348–350.
[24] Manna, L., Valvo, L., Chromatographia 2004, 60,
645–649.
5 References [25] Jelinska, A., Medenecka, B., Zajac, M., Knajsiak, M.,
Acta Pol. Pharm. 2008, 65, 261–265.
[1] Moss, G. P., Pure Appl. Chem. 1996, 68, 2193–2222. [26] Chong, X. M., Hu, C. Q., Chromatographia 2008, 68,
[2] Can, N. O., Altiokka, G., Aboul-Enein, H. Y., Anal. Chim. 759–766.
Acta 2006, 576, 246–252. [27] Wang, M., Shi, A. M., Hua, W. Y., Zhang, Q. Y., Chin. J.
[3] Garcia-Valcarcel, A. I., Tadeo, J. L., J. Sep. Sci. 2009, 32, Antibiot. 2010, 35, 214–218.
3890–3897. [28] Nageswara Rao, R., Venkateswarlu, N., Narsimha, R.,
[4] Han, C., Chen, X., Xie, W., Zhu, Z., Liu, C., Chen, F., J. Chromatogr. A 2008, 1187, 151–164.
Shen, Y., J. Sep. Sci. 2010, 33, 3319–3325. [29] Shyu, W. C., Shukla, U. A., Shah, V. R., Papp, E. A.,
[5] Zhao, L., Tian, X., Fan, P. C., Zhan, Y. J., Shen, D. W., Barbhaiya, R. H., Pharm. Res. 1991, 8, 992–996.
Jin, Y., J. Chromatogr. A 2008, 1210, 168–177. [30] Shyu, W. C., Gleason, C. R., Barbhaiya, R. H., Clin.
[6] Gao, X., Chen, W., Zhu, G., Yi, R., Wu, Z., Xu, P., Zhao, Pharmacokinet. 1993, 25, 237–242.
Y., J. Chromatogr. A 2011, 1218, 1416–1422. [31] Wu, L. P., Ye, L. M., Chen, C., Wu, J. Q., Chen, Y.,
[7] Xiong, W., Glick, J., Lin, Y., Vouros, P., Anal. Chem. Biomed. Chromatogr. 2008, 22, 606–615.
2007, 79, 5312–5321. [32] Guiochon, G., J. Chromatogr. A 2007, 1168, 101–168.
[8] Thayer, J. R., Flook, K. J., Woodruff, A., Rao, S., Pohl, C. [33] Xu, R. N., Fan, L., Rieser, M. J., El-Shourbagy, T. A.,
A., J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. J. Pharm. Biomed. Anal. 2007, 44, 342–355.
2010, 878, 933–941.
[34] Ikegami, T., Tanaka, N., Curr. Opin. Chem. Biol. 2004, 8,
[9] Thayer, J. R., Wu, Y., Hansen, E., Angelino, M. D., Rao, 527–533.
S., J. Chromatogr. A 2011, 1218, 802–808.
[35] Aboul-Enein, H. Y., Ali, I., Talanta 2005, 65, 276–280.
[10] Preinerstorfer, B., Lubda, D., Mucha, A., Kafarski, P.,
[36] Aboul-Enein, H. Y., Hefnawy, M. M., Talanta 2005, 65,
Lindner, W., Lammerhofer, M., Electrophoresis 2006,
67–73.
27, 4312–4320.
[37] Wu, J. T., Zeng, H., Deng, Y., Unger, S. E., Rapid
[11] Mancini, F., Fiori, J., Bertucci, C., Cavrini, V., Bragieri,
Commun. Mass Spectrom. 2001, 15, 1113–1119.
M., Zanotti, M. C., Liverani, A., Borzatta, V., Andrisano,
V., J. Chromatogr. A 2004, 1046, 67–73. [38] Gerber, F., Krummen, M., Potgeter, H., Roth, A.,
Siffrin, C., Spoendlin, C., J. Chromatogr. A 2004, 1036,
[12] Hadad, G. M., Emara, S., Abdel-Salam, R. A., Chroma-
127–133.
tographia 2009, 70, 217–221.
[39] ICH Expert Working Group, Validation of Analytical
[13] Schulte, M., Dingenen, J., J. Chromatogr. A 2001, 923,
Procedures: Text and Methodology Q2(R1), Interna-
17–25.
tional Conference on Harmonisation of Technical
[14] Can, N. O., Arli, G., J. AOAC Int. 2010, 93, 1077–1085. Requirements for Registration of Pharmaceuticals for
[15] Borges, K. B., Sanchez, A. J., Pupo, M. T., Bonato, P. S., Human Use, Geneva 2005.
Collado, I. G., J. AOAC Int. 2010, 93, 1811–1820. [40] US Pharmacopeial Convention, US Pharmacopeia 27th
[16] Gotta, J., Grisales, J. O., Reta, M. R., Castells, C. B., Ed., Rockville, 2004.
J. Sep. Sci. 2010, 33, 2645–2653. [41] Snyder, L. R., Kirkland, J. J., Dolan, J. W., Swartz, M.,
[17] Liu, J., Sun, J., Zhang, W., Gao, K., He, Z., J. Pharm. in: Snyder, L. R., Kirkland, J. J., Dolan, J. W. (Eds.),
Biomed. Anal. 2008, 46, 405–409. Introduction to Modern Liquid Chromatography, 3rd
Edn., Wiley, New Jersey 2010, pp. 523–526.
[18] Satinsky, D., Chocholous, P., Salabova, M., Solich, P.,
J. Sep. Sci. 2006, 29, 2494–2499. [42] Snyder, L. R., Dolan, J. W., Carr, P. W., Anal. Chem.
2007, 79, 3254–3262.
[19] Shyu, W. C., Shah, V. R., Campbell, D. A., Venitz, J.,
Jaganathan, V., Pittman, K. A., Wilber, R. B., Barbhaiya, [43] Wilson, N. S., Gilroy, J., Dolan, J. W., Snyder, L. R.,
R. H., Antimicrob. Agents Chemother. 1992, 36, J. Chromatogr. A 2004, 1026, 91–100.
938–941. [44] Gilroy, J. J., Dolan, J. W., Snyder, L. R., J. Chromatogr.
[20] Park, T. H., Kim, J. K., Jee, J. P., Park, J. S., Kim, C. K., A 2003, 1000, 757–778.
J. Pharm. Biomed. Anal. 2004, 36, 243–248.

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com