Spectrophotometric and liquid chromatographic determination of fenofibrate and vinpocetine and their hydrolysis products

Article abstract of DOI:10.1016/j.farmac.2005.01.013

Several spectrophotometric and HPLC methods are presented for the determination of fenofibrate, vinpocetine and their hydrolysis products. The resolution of either fenofibrate or vinpocetine and their hydrolysis products has been accomplished by using numerical spectrophotometric methods as partial least squares (PLS-1) and principal component regression (PCR) applied to UV spectra; and graphical spectrophotometric methods as first derivative of ratio spectra (¹DD) or first (¹D) and second (²D) derivative spectrophotometry for vinpocetine and fenofibrate, respectively. In addition HPLC methods were developed using ODS column with mobile phase consisting of acetonitrile-water (80:20, v/v, pH 4) with UV detection at 287:nm for fenofibrate and a mobile phase consisting of acetonitrile-10:mM KH ₂PO₄, containing 0.1% diethylamine (60:40, v/v, pH:4.6) with UV detection at 270:nm for vinpocetine. The proposed methods were successfully applied for the determination of each drug and its hydrolysis product in laboratory-prepared mixture and pharmaceutical preparation. The proposed HPLC and derivative spectrophotometric methods were used to investigate the kinetics of acidic and alkaline hydrolytic processes of each drug. The pH-rate profile of hydrolysis of each drug in Britton-Robinson buffer solutions was studied.

Full text of DOI:10.1016/j.farmac.2005.01.013

Il Farmaco 60 (2005) 425–438
http://france.elsevier.com/direct/FARMAC/
Spectrophotometric and liquid chromatographic determination
of fenofibrate and vinpocetine and their hydrolysis products
Alaa El-Gindy a,*, Samy Emara a, Mostafa K. Mesbah b, Ghada M. Hadad a
a Pharmaceutical Analytical Chemistry Department, Faculty of Pharmacy, Suez Canal University, Ismailia 41522, Egypt
b Pharmacognsy Department, Faculty of Pharmacy, Suez Canal University, Ismailia 41522, Egypt
Received 5 July 2004; received in revised form 11 January 2005; accepted 26 January 2005
Available online 27 April 2005
Abstract
Several spectrophotometric and HPLC methods are presented for the determination of fenofibrate, vinpocetine and their hydrolysis prod-
ucts. The resolution of either fenofibrate or vinpocetine and their hydrolysis products has been accomplished by using numerical spectropho-
tometric methods as partial least squares (PLS-1) and principal component regression (PCR) applied to UV spectra; and graphical spectro-
photometric methods as first derivative of ratio spectra (1DD) or first (1D) and second (2D) derivative spectrophotometry for vinpocetine and
fenofibrate, respectively. In addition HPLC methods were developed using ODS column with mobile phase consisting of acetonitrile–water
(80:20, v/v, pH 4) with UV detection at 287 nm for fenofibrate and a mobile phase consisting of acetonitrile–10 mM KH2PO4, containing
0.1% diethylamine (60:40, v/v, pH 4.6) with UV detection at 270 nm for vinpocetine.
The proposed methods were successfully applied for the determination of each drug and its hydrolysis product in laboratory-prepared
mixture and pharmaceutical preparation. The proposed HPLC and derivative spectrophotometric methods were used to investigate the kinetics
of acidic and alkaline hydrolytic processes of each drug. The pH-rate profile of hydrolysis of each drug in Britton–Robinson buffer solutions
was studied.
© 2005 Elsevier SAS. All rights reserved.
Keywords: Fenofibrate; vinpocetine; HPLC; derivative spectrophotometry; derivative ratio spectra; chemometry; kinetics of hydrolysis; pH-rate profile
1. Introduction
Vinpocetine (VP), Ethyl (3a, 16a)-eburnamenine-14-
carboxylate, is a derivative of vincamine. It has been used in
cerebrovascular and cognitive disorders [1]. Determination
ofVP was carried out using HPLC [15–18], sensitive enzyme
immunoassay [19] and GC-MS [20–24].
Fenofibrate (FB), Isopropyl 2-[4-(4-chlorobenzoyl)
phenoxy]-2-methylpropionate, is fibric acid derivative, used
for regulating plasma lipids and treatment of hyperlipopro-
teinaemias [1]. The literature survey reveals that FB was ana-
lyzed in environmental water samples using solid phase
extraction followed by GC-MS [2,3] or HPLC electrospray
ionization MS-MS [3]. HPLC was used for determination of
FB [4–6], its metabolites [7–11] and related impurities
[12,13]. Other HPLC methods for assay and purity of FB and
an NMR method for related compounds in FB raw materials
were reported [14]. The reported HPLC methods resolved
11 known and six unknown impurities from FB using Waters
symmetry ODS column. HPLC is the technique of choice for
the determination of the impurities [12].
Both FB and VP are ester type drugs. Improper storage
hydrolyzed them producing 2-[4(4-chlorobenzoyl) phenoxy]-
2-methylpropanoic acid (fenofibric acid) (FA) and (3a,16a)-
Eburnamenine-14-carboxylic acid (apovincaminic acid) (VA),
respectively. Both FA and FB are active, however the latter is
preferred for medicinal use as FA is absorbed more slowly
and to a smaller extent than FB while VA is inactive. There-
fore, it was necessary, to study the stability of both drugs
towards acidic and alkaline hydrolysis. The aim of this work
was to develop analytical methods for determination of FB,
VP and their hydrolysis products using first or second deriva-
tive (1D or 2D) spectrophotometry, first derivative of the ratio
spectrophotometry (1DD); principal component regression
(PCR) and partial least square (PLS-1) applied to the UV
* Corresponding author. Tel.: + 20 33934945; fax: + 20 64 3561877.
E-mail address: ghhadad@hotmail.com (A. El-Gindy).
0014-827X/$ - see front matter © 2005 Elsevier SAS. All rights reserved.
doi:10.1016/j.farmac.2005.01.013

426
A. El-Gindy et al. / Il Farmaco 60 (2005) 425–438
absorption spectra. The spectrophotometric methods can be
used as useful alternative approaches to HPLC for the analy-
sis of the FB and VP in the presence of their hydrolysis prod-
ucts. The developed HPLC and 2D for FB or 1D for VP were
used to investigate the kinetics of the acidic and alkaline
hydrolytic processes and to calculate the activation energy
for both drug hydrolysis. The proposed HPLC method was
used for pH-rate profile study of hydrolysis of both FB and
VP in Britton–Robinson buffer solutions.
Pharmaceutical, Abu-Sultan, Ismailia, Egypt, labeled to con-
tain 5 mg VP per tablet.
2.3. HPLC conditions
The mobile phases were prepared by mixing acetonitrile
and distilled de-ionized water in a ratio of 80:20 v/v and
adjusted to the apparent pH 4.0 using phosphoric acid for
FB, or acetonitrile and 10 mM KH2PO4 containing 0.1%
diethylamine in a ratio of 60:40 v/v, and adjusted to the appar-
ent pH 4.6 using phosphoric acid for VP. The flow rate was
1.5 or 2 ml min–1 for FB or VP, respectively. All determina-
tions were performed at room temperature. The injection vol-
ume was 20 µl.
2. Experimental
2.1. Instrumentation
A double-beam Shimadzu (Japan) UV-Visible spectropho-
tometer, model UV-1601 PC equipped with 1 cm quartz cells
and connected to an IBM compatible computer and a HP
600 inkjet printer was used. The bundled software was UVPC
personal spectroscopy software version 3.7 (Shimadzu). The
spectral bandwidth was 2 nm and the wavelength scanning
speed was 2800 nm min–1 PLS and PCR analysis was carried
out by using PLS-toolbox software version 2.1-PC [25] for
use with MATLAB 5.
The HPLC (Shimadzu, Kyoto, Japan) instrument was
equipped with a model series LC-10 ADVP pump, SCL-
10 AVP system controller, DGU-12 A Degasser, Rheodyne
7725i injector with a 20 µl loop and a SPD-10AVP UV-VIS
detector, separation and quantitation were made on a 150 ×
4.6 mm (I.d) TSK-Gel 5 µm ODS-80 TM column (Tosoh,
Japan). The detector was set at k 287 nm for FB and its
hydrolysis product and k 270 nm for VP and its hydrolysis
product. Data acquisition was performed on class-VP soft-
ware.
2.4. Preparation of the alkali-induced hydrolysis product
One gram of FB or VP was dissolved in 15 ml methanol
and refluxed separately with 100 ml 0.1 M sodium hydroxide
at 100 °C for 45 min or 1 h for FB or VP, respectively. Sub-
sequently, the pH of the solution was adjusted to 1.5 or 2.3,
using 1 M hydrochloric acid to precipitate the hydrolysis prod-
ucts of FB or VP, respectively. The precipitate was filtered,
dried under vacuum. The dried precipitate was analyzed by
IR and NMR and found to be 2-[4(4-chlorobenzoyl)
phenoxy]-2-methylpropanoic acid (FA) as a hydrolyzed of
FB and (3a, 16a)-Eburnamenine-14-carboxylic acid (VA) as
a hydrolyzed of VP.
The PMR spectrum of FA in deutrated chloroform showed
protons signals of propianoate chain protons at d 1.701 ppm
(singlet, 6H, CH3–C(CH3)–C–OO) and appearance of aro-
matic protons at d 6.897–6.968 ppm (doublet, 2H, aromatic
C3-H, C5-H); d 7.423–7.466 ppm (doublet, 2H, aromatic
C2-H, C6-H); d 7.686–7.722 ppm (doublet, 2H, aromatic
C3-H, C5-H) and d 7.757–7.766 ppm (doublet, 2H, aromatic
C2-H, C6-H).
The IR spectrum (KBr) of FA revealed the OH associa-
tion, C=O stretching and in-plane C–O bending of the COOH
at 2905–3430, 1708 and 1306 cm–1, respectively.
The PMR spectrum of VA in dimethylsulphoxide showed
aromatic protons at d 7.070–7.486 ppm.
The IR spectrum (KBr) of VA revealed the OH associa-
tion, C=O stretching and in-plane C–O bending of the COOH
at 2366–3287, 1662 and 1328 cm–1, respectively.
The IR spectrophotometer used was a Bruker Vector 22,
Germany.
PMR spectra were recorded on a Varian Gemini 200 PMR
spectrometer (200 mHz), USA.
2.2. Materials and reagents
Pharmaceutical grade of FB was kindly supplied by Minap-
harm Egypt company for pharmaceuticals industries and cer-
tified to contain 99.99% and pharmaceutical grade ofVP was
kindly supplied by Medical Union Pharmaceutical, Abu-
Sultan, Ismailia, Egypt and certified to contain 99.98%.Aceto-
nitrile used was HPLC grade. Methanol, sodium hydroxide,
potassium dihydrogen phosphate, hydrochloric, phosphoric,
boric, acetic acids and diethylamine were analytical grade.
Ten millimolar KH2PO4 was prepared by dissolving 1.36 g
of KH2PO4 in one liter of distilled water.
The commercial lipanthyl capsules used was manufac-
tured by Minapharm Egypt company for pharmaceuticals
industries under licence of laboratories FOURNIER, France,
labeled to contain 300 mg FB per capsule and commercial
Angiovan tablets used was manufactured by Medical Union
2.5. Preparation of the acid-induced hydrolysis product
One gram of FB (firstly dissolved in 15 ml methanol) or
VP was refluxed separately with 100 ml 1 M hydrochloric
acid at 100 °C for 10 or 14 h for FB or VP, respectively. Sub-
sequently, the pH of the solution was adjusted to 1.5 or 2.3,
using 2 M sodium hydroxide to precipitate the hydrolysis
products of FB or VP, respectively. The same procedure for
separation of FA and VA previously described under prepa-
ration of the alkali-induced hydrolysis product was followed.

A. El-Gindy et al. / Il Farmaco 60 (2005) 425–438
427
2.6. Standard solutions and calibration graphs
2.6.1.4. HPLC method. Triplicate 20 µl injections were made
for each concentration of FB and FA and chromatographed
under the conditions described above. The peak area of each
concentration was plotted against the corresponding concen-
tration to obtain the calibration graph of each compound.
2.6.1. For FB and FA
Stock solutions were prepared by separately dissolving FB
and FA in methanol to obtain a concentration of 50 µg ml–1.
The standard solutions were prepared by diluting the stock
solutions with methanol (for spectrophotometric methods) or
mobile phase (for HPLC method) to reach the concentration
range of 4–20 µg ml–1 for FB and 1–20 µg ml–1 for FA.
2.6.2. For VP and VA
Stock solutions were prepared by separately dissolvingVP
and VA in methanol to obtain a concentration of 100 µg ml–1.
The standard solutions were prepared by diluting the stock
solutions with 0.1 M hydrochloric acid (for spectrophotomet-
ric methods) or mobile phase (for HPLC method) to reach
the concentration range of 5–40 µg ml–1 for VP and
1–40 µg ml–1 for VA.
2.6.1.1. Second derivative method. The second derivative
spectra (2D) of the standard solutions were recorded in the
range of 200–400 nm against solvent blank using Dk = 8 nm
2
and scaling factor of 20. The values of D amplitudes at
313 nm (zero-crossing of FA) were measured for the deter-
mination of FB in presence of FA and the values of 2D ampli-
tudes at 305 nm (zero-crossing of FB) were measured for the
determination of FA in presence of FB. The second deriva-
tive amplitudes of FB and FA at 313 and 305 nm, respec-
tively, versus their concentrations were plotted in order to
obtain the calibration graphs of each compound.
2.6.2.1. First derivative method. The first derivative spectra
(1D) of the standard solutions were recorded in the range of
200–400 nm against solvent blank using Dk = 8 nm and scal-
1
ing factor of 10. The values of D amplitudes at 302.6 nm
(zero-crossing of VA) were measured for the determination
of VP in presence of VA and the values of 1D amplitudes at
292.6 nm (zero-crossing of VP) were measured for the deter-
mination of VA in presence of VP. The first derivative ampli-
tudes of VP and VA at 302.6 and 292.6 nm, respectively, ver-
sus their concentrations were plotted in order to obtain the
calibration graphs of each compound.
2.6.1.2. First derivative ratio spectra method. For determi-
nation of FB: The UV absorption spectra of standard solu-
tions of FB were recorded in the wavelength range of 230–
400 nm and divided by a normalized spectrum of FA [a
spectrum of unit concentration], where the hydrolysis prod-
uct (isopropanol) has neglected absorbance in this wave-
length range. The first derivative was calculated for the
obtained ratio spectra with Dk = 8 nm. The first derivative of
the ratio spectra obtained was smoothed with four experimen-
tal points and scaling factor of 10. The amplitude at 300.2 nm
was measured and found to be proportional to the concentra-
tion of FB.
For determination of FA: The UV absorption spectra of
standard solutions of FA were recorded in the wavelength
range 230–400 nm and divided by a normalized spectrum of
FB [a spectrum of unit concentration]. The first derivative
was calculated for the obtained ratio spectra with Dk = 8 nm.
The first derivative of the ratio spectra obtained was smoothed
with four experimental points and scaling factor of 10. The
amplitude at 311 nm was measured and found to be propor-
tional to the concentration of FA.
2.6.2.2. First derivative ratio spectra method. For determi-
nation of VP: The UV absorption spectra of standard solu-
tions of VP were recorded in the wavelength range of 230–
400 nm and divided by a normalized spectrum of VA [a
spectrum of unit concentration] where the hydrolysis prod-
uct (ethanol) has neglected absorbance in this wavelength
range. The first derivative was calculated for the obtained ratio
spectra with Dk = 8 nm. The first derivative of the ratio spec-
tra obtained was smoothed with four experimental points and
scaling factor of 10. The amplitude at 296.2 nm was mea-
sured and found to be proportional to the concentration of
VP.
For determination of VA: The UV absorption spectra of
standard solutions of VA were recorded in the wavelength
range of 230–400 nm and divided by a normalized spectrum
of VP [a spectrum of unit concentration]. The first derivative
was calculated for the obtained ratio spectra with Dk = 8 nm.
The first derivative of the ratio spectra obtained was smoothed
with four experimental points and scaling factor of 10. The
amplitude at 294.0 nm was measured and found to be propor-
tional to the concentration of VA.
2.6.1.3. For PLS and PCR methods. A training set of 30 labo-
ratory-prepared mixtures with different concentrations of FB
and FA in the range of 4–20 µg ml–1 for FB and 1–20 µg ml–1
for FA within concentration ratio ranged from 1:0.05 to 1:5 for
FB:FA were prepared in methanol.
The UV absorption was recorded over the range of 250–
320 nm. The data points of the spectra were collected at every
0.4 nm. The computations were made using PLS-Toolbox
software version 2.1.
PLS-1 and PCR models were applied to the UV absorp-
tion spectra of these mixtures using two latent variables (or
two principal component) for determination of FB and FA.
2.6.2.3. For PLS and PCR methods. A training set of 30 labo-
ratory-prepared mixtures with different concentrations ofVP
and VA in the range of 5–40 µg ml–1 for VP and 1–40 µg ml–1
for VA within concentration ratio ranged from 1:0.025 to
1:8 for VP:VA were prepared in 0.1 M hydrochloric acid.
The UV absorption was recorded over the range of 250–
340 nm. The data points of the spectra were collected at every

428
A. El-Gindy et al. / Il Farmaco 60 (2005) 425–438
0.6 nm. The computations were made using PLS-Toolbox
software version 2.1.
PLS-1 and PCR models were applied to the UV absorp-
tion spectra of these mixtures using two latent variables (or
two principal component) for determination of VP and VA.
diluted to volume with methanol. Separate 1 ml aliquot of FB
or VP solution and 2 ml of the Britton–Robinson buffer
solutions were transferred into stoppered conical flasks. The
pH-values of Britton–Robinson buffer solutions [26] used for
the measurement of the pH-rate profile of the hydrolysis of
FB and VP were as follows; pH 2, 2.9, 3.6, 4.5, 5.8, 6.7, 8, 9,
10.5 and 12. The flasks were placed in a thermostatic oven at
80 °C for different time intervals. At the specified time
interval, the contents of flasks were neutralized, using 1 M
sodium hydroxide or 1 M hydrochloric acid solutions. The
contents of flasks were transferred into 50 ml volumetric flasks
and diluted to volume with the mobile phase.Aliquot of 20 µl
of each solution was chromatographed under the conditions
described above and the concentration of the remaining FB
and VP were calculated at each pH value and time interval.
2.6.2.4. HPLC method. Triplicate 20 µl injections were made
for each concentration of VP and VA and chromatographed
under the conditions described above. The peak area of each
concentration was plotted against the corresponding concen-
tration to obtain the calibration graph of each compound.
2.7. Sample preparation
The contents of 20 Lipanthyl® capsules or 20 Angiovan®
tablets were weighed and finely powdered. Accurately
weighed portion of the powder equivalent to about 50 mg of
FB orVP was separately extracted and diluted to 100 ml with
methanol. The sample solution was filtered.
3. Results and discussion
Further dilutions of the filtrated capsule and tablet samples
were carried out with the specified solvent (for spectropho-
tometric methods) or mobile phase (for HPLC methods) to
reach concentration range of 4–20 or 5–40 µg ml–1 for FB or
VP, respectively. The general procedures for spectrophoto-
metric and HPLC methods of FB or VP described under cali-
bration were followed.
When FB orVP were boiled with 0.1 M sodium hydroxide
for 45 min or 1 h or 1 M hydrochloric acid for 10 or 12 h, FA
orVA could be isolated from the reaction mixture as hydroly-
sis products of FB or VP, respectively. The suggested path-
way for the hydrolysis of FB andVP in 0.1 M sodium hydrox-
ide and 1 M hydrochloric acid is presented in Scheme 1.
3.1. Assay parameters
2.8. Kinetic investigation
3.1.1. 1D or 2D method
Accurately weighed 50 or 100 mg of FB or VP, respec-
tively, were dissolved separately in 100 ml methanol. Sepa-
rate 2 ml aliquots of these solutions were transferred into sepa-
rate stoppered conical flasks and mixed with 2 ml of 0.2 M
sodium hydroxide or 2 M hydrochloric acid. The flasks were
placed in a thermostatic oven at different temperatures (90,
85, 80, 70, 60 °C for acidic hydrolysis and 80, 70, 60, 50,
40 °C for alkaline hydrolysis of FB andVP) for different time
intervals.At the specified time, the contents of the flasks were
neutralized using predetermined volumes of 1 M sodium
hydroxide and 0.1 M hydrochloric acid solutions. The con-
tents of the flasks were transferred into 50 ml volumetric flasks
and diluted to volume with methanol or 0.1 M hydrochloric
acid for 2D (for FB) or 1D (for VP), respectively and mobile
phase for the HPLC methods.
The UV absorption spectra of FB with its hydrolysis prod-
uct FA in methanol or VP with its hydrolysis product VA in
0.1 M hydrochloric acid are overlapped (Figs. 1a, 2a), while
their second or first derivative spectra, respectively, (Figs. 1b,
2b) showed significant differences in some areas that permit
For the 2D (for FB) or 1D (for VP) method: The 2D or 1D
amplitudes were measured for the solution directly as
described under the calibration graph and the concentration
of the remaining FB or VP were calculated at each tempera-
ture and time interval.
For HPLC methods, aliquots of 20 µl of each solution were
chromatographed under the conditions described above and
the concentration of the remaining FB or VP was calculated
at each temperature and time interval.
2.9. pH-rate profile
Accurately weighed 100 or 200 mg of FB or VP, respec-
tively were transferred into 100 ml volumetric flask and
Scheme 1
. Suggested pathway for the hydrolysis of FB and VP in 0.1 M
sodium hydroxide and 1 M hydrochloric acid.

A. El-Gindy et al. / Il Farmaco 60 (2005) 425–438
429
Fig. 1
(------) in methanol.
.
UV absorption spectra (a) and second derivative spectra (b) of 14 µg ml–1 of FB (––––), 14 µg ml–1 of FA (– – – –) and 14 µg ml–1 of isopropyl alcohol
Fig. 2.
UV absorption spectra (a) and first derivative spectra (b) of 10 µg ml–1 of VP (––––), 10 µgml–1 of VA (– – – –) and 10 µg ml–1 of ethyl alcohol (------)
in 0.1 M hydrochloric acid.
their determination. FB was determined by measurement of
its second derivative amplitude at the zero-crossing point of
its hydrolysis product FA at 313 nm and its hydrolysis prod-
uct FA was determined by measurement of its second deriva-
tive amplitude at the zero-crossing point of FB at 305 nm.
While VP was determined by measurement of its first deriva-
tive amplitude at the zero-crossing point of its hydrolysis prod-
uct VA at 302.6 nm and its hydrolysis product VA was deter-
mined by measurement of its first derivative amplitude at the
zero-crossing point ofVP at 292.6 nm. Isopropanol and etha-
nol do not interfere with such determinations.
3.1.2. 1DD method
To optimize the 1DD method for the determination of each
drug and its hydrolysis product, it is necessary to test the influ-
ence of the variables: divisor concentration, Dk and smooth-
ing function. All these variables were studied and Dk = 8 nm
was selected as optimum value. The best results in terms of
signal to noise ratio, sensitivity and repeatability followed
using normalized spectrum as divisor. Due to the extent of
the noise levels on the ratio spectra, a smoothing function
was used and four experimental points were considered as
suitable. Using the procedures of first derivative ratio spectra

430
A. El-Gindy et al. / Il Farmaco 60 (2005) 425–438
Fig. 3
.
First derivative ratio spectra for (a) different concentrations of FB (4, 6, 8, 10, 14, 18, 20 µg ml–1) using normalized spectrum of FA as divisor and (b)
different concentrations of FA (1, 4, 8, 10, 14, 18, 20 µg ml–1) using normalized spectrum of FB as divisor.
method described under calibration, the amplitudes at 300.2,
311, 296.2 and 294 nm were found to be proportional to the
concentrations of FB, FA,VP andVA, respectively (Figs. 3,4).
The characteristics parameters for the regression equation
of 1DD method for determination of FB, FA,VP andVA were
given in Tables 1 and 2.
FB and FA where there are absorption characteristics of FB
and FA. Wavelengths less than 250 nm were rejected, as there
is little contribution from isopropanol while wavelength more
than 320 nm were not used because they are less informative
wavelength for FB and FA.
While forVP andVA, the spectra were measured at 0.6 nm
intervals over the range 250–340 nm. Wavelengths less than
250 nm were rejected, as there is little contribution from etha-
nol, whereas wavelengths more than 340 nm were not used
because they are less informative wavelength for VP and VA.
To select the number of factors in the PLS-1 and PCR algo-
rithms, a cross-validation method leaving out one sample at a
time [28] was employed using a training (calibration) set of
3.1.3. PCR and PLS-1
The quality of multicomponent analysis results is depen-
dent on the wavelength range and spectral mode used [27]. In
this work, spectral resolution was assayed with absorbance
spectra for PLS-1 and PCR methods. The spectra were mea-
sured at 0.4 nm intervals over the range of 250–320 nm for
Fig. 4
different concentrations of VA (1, 10, 15, 25, 30, 35, 40 µg ml–1) using normalized spectrum of VP as divisor.
.
First derivative ratio spectra for (a) different concentrations of VP (5, 10, 20, 25, 30, 35, 40 µg ml–1) using normalized spectrum of VA as divisor and (b)

Table 1
Characteristic parameters for the regression equations of the proposed HPLC, second derivative (2D) and first derivative–ratio spectra (1DD) methods for determination of FB and FA
Parameters
HPLC
FA
2D
1DD
FB
FB
FA
FB
FA
Calibration range (µg ml–1
)
4–20
1–20
4–20
1–20
4–20
1–20
Detection limit (µg ml–1
Quantitation limit (µg ml–1
)
2.00 × 10–2
6.66 × 10–2
3.87 × 104
3.29 × 102
0.85
3.84 × 104–3.90 × 104
–1.91 × 103
4.20 × 103
–5.99 × 103–2.16 × 103
0.9999
2.15 × 10–2
7.16 × 10–2
4.06 × 104
0.025
0.028
1.69 × 10–3
5.63 × 10–3
3.239 × 10–2
2.33 × 10–5
0.072
3.237 × 10–2–3.241 × 10–2
–4.74 × 10–5
2.97 × 10–4
–3.35 × 10–4–2.41 × 10–4
0.9999
1.53 × 10–3
5.11 × 10–3
1.0643 × 10–1
6.96 × 10–5
)
0.085
0.093
Regression Eq. (Y)a: Slope (b)
Standard deviation of the slope (Sb)
Relative standard deviation of the slope (%)
Confidence limit of the slope b
Intercept (a)
2.88 × 10–3
3.12 × 10–5
1.08
2.82 × 10–3–2.94 × 10–3
2.16 × 10–4
7.83 × 10–4
–9.8 × 10–4–5.4 × 10–4
0.9998
3.64 × 10–3
4.31 × 10–5
1.18
3.59 × 10–3–3.68 × 10–3
3.24 × 10–4
5.40 × 10–4
–2.01 × 10–4–8.49 × 10–4
0.9998
3.71 × 102
0.91
0.065
4.02 × 104–4.09 × 104
6.04 × 102
4.65 × 103
1.0636 × 10–1–1.0649 × 10–1
–5 × 10–4
8.73 × 10–4
Standard deviation of the intercept (Sa)
Confidence limit of the intercept b
Correlation coefficient (r)
–3.92 × 103–5.13 × 103
–1.35 × 10–3–3.49 × 10–4
0.9999
0.9999
Standard error of estimation
1.85 × 103
2.42 × 103
3.46 × 10–4
2.81 × 10–4
1.31 × 10–4
4.54 × 10–4
a Y = a + bC, where C is the concentration of FB and FA in µg ml–1 and Y is the peak area in the HPLC method and the amplitude at the specified wavelength in the 2D and 1DD methods.
b 95% confidence limit.

Table 2
Characteristic parameters for the regression equations of the proposed HPLC, first derivative (1D) and first derivative–ratio spectra (1DD) methods for determination of VP and VA
Parameters
HPLC
VA
1D
1DD
VA
VP
VP
VA
VP
Calibration range (µg ml–1
)
5–40
1–40
5–40
1–40
5–40
1–40
Detection limit (µg ml–1
Quantitation limit (µg ml–1
)
2.02 × 10–2
6.72 × 10–2
3.40 × 104
1.10 × 102
0.86
3.39 × 104–3.45 × 104
–1.78 × 103
1.39 × 104
–1.53 × 104–1.17 × 104
0.9998
2.22 × 10–2
7.40 × 10–2
2.65 × 104
1.09 × 10–2
3.62 × 10–2
6.28 × 10–3
2.91 × 10–5
0.46
6.25 × 10–3–6.31 × 10–3
–2.2 × 10–5
7.67 × 10–4
–7.70 × 10–4–7.23 × 10–4
0.9998
5.28 × 10–3
1.76 × 10–2
9.51 × 10–3
2.14 × 10–5
0.23
9.49 ×10–3–9.53 × 10–3
–4.1 × 10–4
5.53 × 10–4
–9.50 × 10–4–1.30 × 10–4
0.9998
8.79 × 10–3
2.93 × 10–2
21.247 × 10–2
7.96 × 10–4
0.37
6.52 × 10–4
2.17 × 10–3
7.615 × 10–2
2.12 × 10–5
0.028
)
Regression Eq. (Y)a: Slope (b)
Standard deviation of the slope (Sb)
Relative standard deviation of the slope (%)
Confidence limit of the slope b
Intercept (a)
2.50 × 102
0.95
2.63 × 104–2.68 × 104
–1.60 × 103
6.48 × 103
21.244 × 10–2–21.250 × 10–2 7.613 × 10–2–7.617 × 10–2
–1.20 × 10–4
7.96 × 10–4
–3.80 × 10–4
5.47 × 10–4
–9.10 × 10–4–1.50 × 10–4
Standard deviation of the intercept (Sa)
Confidence limit of the intercept b
Correlation coefficient (r)
–7.90 × 103–4.70 × 103
–8.90 × 10–4–6.54 × 10–4
0.9999
3.28 × 103
0.9999
3.58 × 10–4
0.9999
2.77 × 10–4
Standard error of estimation
6.26 × 103
3.45 × 10–4
2.80 × 10–4
a Y = a + bC, where C is the concentration of VP and VA in µg ml–1 and Y is the peak area in the HPLC method and the amplitude at the specified wavelength in the 2D and 1DD methods.
b 95% confidence limit.

A. El-Gindy et al. / Il Farmaco 60 (2005) 425–438
433
30 calibration spectra. The predicted concentrations of the
components in each sample were compared with the actual
concentrations in the training samples and the root mean
square error of cross validation (RMSECV) was calculated
for each method as follows:
of the models. The RMSECV obtained by optimizing the cali-
bration matrix of the absorption spectra for the PLS-1 and
PCR methods are shown in Table 3 indicating good accuracy
and precision.
Another diagnostic test was carried out by plotting the con-
centration residuals against the predicted concentrations. The
residuals appear randomly distributed around zero, indicat-
ing adequate models.
PRESS
RMSECV =
where n is the number of training
ͱ
n
samples
2
3.1.4. HPLC method
PRESS = R Ypred − Ytrue
͑
͒
The effect of acetonitrile composition and the apparent pH
of the mobile phase on the capacity factor (K/) of each com-
pound were studied and optimized. The reversed-phase liq-
uid chromatographic separation of basic drugs such as VP
often involves the interaction with residual silanol groups,
resulting in poor chromatographic performance [30]. The
addition of an organic amine such as diethylamine to the
aqueous-organic eluent greatly improves the sharpness of the
VP peak.
Using the HPLC conditions described under section 2.3.,
the average retention time S.D. for FB, FA, VP and VA
were found to be 4.9 0.007, 2.3 0.005, 4.1 0.008 and
1.9 0.005 min, respectively, for 10 replicates. The peaks
obtained were sharp and have clear base line separation
(Figs. 5,6).
where Ypred and Ytrue are predicted and true concentrations in
µg ml–1, respectively.
The RMSECV was used as a diagnostic test for examin-
ing the errors in the predicted concentrations. It indicates both
of the precision and accuracy of predictions.
Appropriate selection of the number of factors to be used
to construct the model is a key to achieving correct quantation
in PLS-1 and PCR calibrations. The most usual procedure
for this purpose involves choosing the number of factors that
results in the minimum RMSECV. However, this criterion is
subjected to some constraints since, occasionally, the
RMSECV does not reach a sharp minimum, but decreases
gradually above a given number of factors. On another hand,
it is calculated from a finite number of samples, so it is
error-prone. For these reasons, the method developed by
Haaland and Thomas [29] was used for selecting the optimum
number of factors, which involves selecting that model
including the smallest number of factors that results in an
insignificant difference between the corresponding RMSECV
and the minimum RMSECV. A number of factor 2 was found
to be optimum for each component by PLS-1 and PCR
methods.
3.2. Pharmaceutical products analysis
The five proposed methods were applied to the determina-
tion of FB or VP in their commercial dosage forms. Seven
replicates determinations were made. Satisfactory results were
obtained in a good agreement with the label clams (Table 4).
The results for determination of FB and VP were compared
with the official and published HPLC methods [4,17], respec-
tively. Statistical comparison of the results was performed
with regards to accuracy and precision using Student’s t-test
and the F-ratio at 95% confidence level (Table 4). There is no
significant difference between the proposed and the pub-
lished methods with regard to accuracy and precision.
The evaluation of the predictive abilities of the models was
performed by plotting the actual known concentrations against
the predicted concentrations. The results are obtained in
Table 3. Satisfactory correlation coefficient (r) value was
obtained for each compound in the training set by PLS-1 and
PCR optimized models indicating good predictive abilities
Table 3
Cross validation results for simultaneous determination of FB with FA and VP with VA by PLS-1 and PCR methods
Item
Method
Compound
VP
FB
FA
VA
RMSECV
PLS-1
PCR
3.45 × 10–4
3.69 × 10–4
2.26 × 10–4
–2.30 × 10–4
1.0000
4.45 × 10–4
4.15 × 10–4
–4.50 × 10–4
4.52 × 10–4
1.0000
2.63 × 10–4
4.13 × 10–4
4.46 × 10–4
–1.60 × 10–4
1.95 × 10–4
0.9999
2.84 × 10–4
–3.13 × 10–3
2.70 × 10–5
1.0000
Intercept
PLS-1
PCR
Slope
PLS-1
PCR
0.9999
0.9997
0.9999
1.0000
Correlation coefficient (r)
S.E. of intercept
S.E. of slope
PLS-1
PCR
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
0.9999
PLS-1
PCR
2.36 × 10–4
2.36 × 10–4
1.62 × 10–4
1.62 × 10–4
2.76 × 10–4
2.76 × 10–4
1.84 × 10–4
1.84 × 10–4
3.62 × 10–3
4.32 × 10–3
1.39 × 10–4
1.66 × 10–4
1.31 × 10–3
1.54 × 10–3
5.61 × 10–5
6.61 × 10–5
PLS-1
PCR

434
A. El-Gindy et al. / Il Farmaco 60 (2005) 425–438
Table 4
Determination of FB with FA and VP with VA in laboratory-prepared mixtures and commercial pharmaceutical products using the proposed methods
Mean found SDa
HPLC
2D or 1D
1DD
PCR
PLS-1
Published
method [4,17]
For FB with FA laboratory-prepared mixtures
FB
99.9 0.36
99.9 0.40
99.8 0.75
99.9 0.60
100.0 0.48
99.8 0.57
100.0 0.03
99.9 0.04
100.0 0.03
100.0 0.04
FA
Lipanthyl capsuled
FB
99.9 0.65
0.54
99.9 0.92
0.45
99.9 0.68
0.53
100.0 0.50
0.89
100.0 0.53
0.88
99.7 0.73
(2.18)b
(4.28)b
t
F
1.26
1.59
1.15
2.13
1.90
Recoveryc
FB
99.8 0.57
99.9 0.60
99.9 1.06
99.8 0.91
99.8 0.62
99.9 0.58
99.9 0.66
99.9 0.57
100.0 0.62
99.8 0.41
FA
For VP with VA laboratory-prepared mixtures
VP
100.1 0.41
100.1 0.46
100.0 0.79
99.8 0.90
99.9 0.51
99.9 0.55
99.9 0.05
99.9 0.03
VA
100.0 0.05
100.0 0.03
Angiovan tablete
VP
99.3 0.57
0.59
99.4 0.97
0.67
99.3 0.62
0.57
99.3 0.46
0.64
99.3 0.39
0.67
99.1 0.69
(2.18)b
(4.28)b
t
F
1.47
1.98
1.24
2.25
3.13
Recoveryc
VP
99.9 0.48
99.8 0.65
100.0 1.04
99.8 0.98
100.0 0.59
99.9 0.72
100.1 0.49
99.9 0.51
100.0 0.50
100.0 0.43
VA
a Mean and S.D. for seven determinations; percentage recovery from the label claim amount.
b Theoretical values for t and F at P = 0.05.
c For standard addition of different concentrations of FB and FA or VP and VA.
d Batch number: 010038.
e Batch number: 032146.
Fig. 5. HPLC chromatogram of 20 µl injection of laboratory-prepared mix-
ture of 14 µg ml–1 of FB and 14 µg ml–1 of FA.
Fig. 6. HPLC chromatogram of 20 µl injection of laboratory-prepared mix-
ture of 20 µg ml–1 of VP and 20 µg ml–1 of VA.
Expired commercial dosage forms stored at ambient tem-
perature under normal conditions were analyzed by the five
proposed methods. The percentage of declared content of FB
were found to be 88.6, 87.5, 88.4, 87.9, 88.3% determined by
to be 86.6, 85.5, 86.4, 85.2, 85.1%, determined by 1D, 1DD,
HPLC, PCR and PLS-1 methods, respectively.
The HPLC chromatogram of expired commercial dosage
forms for FB or VP showed the peak of the hydrolysis prod-
uct of FA or VA, respectively (Figs. 7,8).
1
2D, DD, HPLC, PCR and PLS-1 methods, respectively.
While, the percentage of declared content of VP were found

A. El-Gindy et al. / Il Farmaco 60 (2005) 425–438
435
Fig. 7
. HPLC chromatogram of expired FB capsule solution containing FB
and its hydrolysis product FA.
Fig. 9. Pseudo first-order plots for the hydrolysis of FB in 1 M hydrochloric
acid (a) and in 0.1 M sodium hydroxide (b) at various temperatures using
HPLC method.
Key: 40 (n); 50(m); 60(x); 70(D); 80(M); 85 (C) and 90 (*); Ct, concentra-
tion at time t, and C0, concentration at zero time.
sible to calculate the apparent first order hydrolysis rate con-
stant and the half-life at each temperature for both acidic and
2
alkaline hydrolytic processes determined by D and HPLC
1
methods for FB and by D and HPLC methods for VP
(Table 5). Plotting log Kobs values versus 1/T, the Arrhenius
plots (Fig. 11) were obtained, which were found to be linear
in the temperature range 60–90 °C for acidic hydrolysis and
40–80 °C for alkaline hydrolysis of FB and VP. The activa-
tion energy was calculated for FB and found to be 10.82 and
10.88 kcal mol–1 for acidic hydrolytic process and 6.58 and
6.74 kcal mol–1 for alkaline hydrolytic process, determined
by the proposed HPLC and 2D methods, respectively. While,
it was calculated for VP and found to be 17.09 and
16.93 kcal mol–1 for acidic hydrolytic process and 8.68 and
8.79 kcal mol–1 for alkaline hydrolytic process, determined
by the proposed HPLC and 1D methods, respectively.
Fig. 8. HPLC chromatogram of expired VP tablet solution containing VP
and its hydrolysis product VA.
3.3. Kinetic investigation
The kinetics of hydrolysis of FB andVP were investigated
in 0.1 M sodium hydroxide and 1 M hydrochloric acid, since
the hydrolysis rates of FB andVP at lower strengths of hydro-
chloric acid were too slow to obtain reliable kinetic data. A
regular decrease in the concentration of intact FB and VP
with increasing time intervals was observed. At the selected
temperatures (60, 70, 80, 85 and 90 °C for the acidic hydroly-
sis; and 40, 50, 60, 70 and 80 °C for alkaline hydrolysis) the
hydrolytic process followed pseudo first order kinetics
(Figs. 9,10). From the slopes of the straight lines it was pos-
The pH-rate profiles of hydrolysis of FB andVP in Britton–
Robinson buffer solutions were studied at 80 °C using HPLC
method (Fig. 12). Britton–Robinson buffer solutions were
used throughout the entire pH range in order to avoid pos-
sible effects of different buffer species. The apparent first order
hydrolysis rate constant and the half-life were calculated for
each pH value (Table 6). FB and VP were found to be most
stable at a pH of 3.6.

436
A. El-Gindy et al. / Il Farmaco 60 (2005) 425–438
Fig. 10. Pseudo first-order plots for the hydrolysis of VP in 1 M hydrochlo-
ric acid (a) and in 0.1 M sodium hydroxide (b) at various temperatures using
HPLC method.
Fig. 11. Arrhenius plots for the hydrolysis of FB (a) andVP (b) in 1 M hydro-
chloric acid (n) and 0.1 M sodium hydroxide (m) using HPLC method.
Key: 40 (n); 50(m); 60(x); 70(D); 80(M); 85 (C) and 90 (*);Ct, concentra-
tion at time t, and C0, concentration at zero time.
adherence of the system to Beer’s law were validated by the
high value of the correlation coefficient and the intercept
value, which was not statistically (p < 0.05) different from
zero (Tables 1 and 2).
3.4. Validation
3.4.1. Linearity
The linearity of the proposed methods was evaluated by
analyzing seven concentrations of FB, FA, VP and VA rang-
ing between 4–20, 1–20, 5–40 and 1–40 µg ml–1, respec-
tively. Each concentration was repeated three times. The assay
was performed according to experimental conditions previ-
ously established. The linearity of the calibration graphs and
3.4.2. Precision
For evaluation of the precision estimates, repeatability and
intermediate precision were performed at three concentra-
tion levels for FB, FA, VP and VA. The data for each concen-
tration level were evaluated by one-way ANOVA. A 8 days ×
2 replicates design was performed. Statistical comparison of
Table 5
Hydrolysis rate constant (Kobs) and half-life (t1/2) for FB and VP in 1 M hydrochloric acid and 0.1 M sodium hydroxide (determined by HPLC and 2D or 1D for
FB or VP, respectively)
Temperature (°C)
Kobs(h–1
)
t1/2(h)
FB
2D
VP
1D
FB
2D
VP
1D
HPLC
HPLC
HPLC
HPLC
In 1 M hydrochloric acid:
60
0.083
0.136
0.212
0.258
0.322
0.083
0.133
0.211
0.258
0.324
0.048
0.100
0.211
0.288
0.408
0.050
0.100
0.210
0.302
0.402
8.380
5.100
3.270
2.690
2.150
8.360
5.210
3.280
2.690
2.140
14.318
6.930
3.284
2.406
1.699
13.832
6.930
3.300
2.294
1.724
70
80
85
90
In 0.1 M sodium hydroxide:
40
50
60
70
80
1.128
1.580
2.150
2.879
3.731
1.125
1.574
2.190
2.881
3.850
0.674
1.059
1.573
2.303
3.293
0.671
1.060
1.592
2.272
3.321
0.614
0.439
0.322
0.241
0.186
0.616
0.440
0.316
0.241
0.180
1.028
0.654
0.440
0.301
0.210
1.033
0.654
0.435
0.305
0.209

A. El-Gindy et al. / Il Farmaco 60 (2005) 425–438
437
3.4.4. Detection and quantitation limits
According to ICH recommendations [31], the approach
based on the S.D. of the response and the slope was used for
determining the detection and quantitation limits. The theo-
retical values were assessed practically and given in Tables 1
and 2.
3.4.5. Selectivity
Methods selectivity was achieved by preparing different
laboratory-prepared mixtures of FB with FA and VP with VA
within the linearity range concentration within ratio 1:0.05 to
1:5 for FB:FA and 1:0.025 to 1:8 for VP:VA, respectively.
The laboratory-prepared mixtures were analyzed according
to the previous procedures described under the five proposed
methods. Satisfactory results were obtained (Table 4), indi-
cating the high selectivity of the proposed methods for deter-
mination of each drug and its hydrolysis product.
3.4.6. Accuracy
This study was performed by addition of known amounts
of each drug and its hydrolysis product to a known concen-
tration of the commercial pharmaceutical products (standard
addition method). The resulting mixtures were assayed and
the results obtained for each drug and its hydrolysis product
were compared with the expected results. The excellent recov-
eries of standard addition method (Table 4) suggested the good
accuracy of the proposed methods.
The influence of the commonly used tablet or capsule
excipients (lactose, pregelatinized starch, magnesium stear-
tae, sodium lauryl sulfate and crospovidone) was investi-
gated before the determination of each drug in its pharmaceu-
tical dosage form. No interference could be observed with
the proposed methods.
Fig. 12. pH-rate profile for the hydrolysis of FB (a) and VP (b) in Britton–
Robinson buffer at 80 °C.
the results was performed using the P-value of the F-test.
Three univariate analyses of variance for each concentration
level were made. Since the P-value of the F-test is always
greater than 0.05, there is no statistically significant differ-
ence between the mean results obtained from one level of
day to another at the 95% confidence level.
3.4.7. Robustness
Variation of the pH of the mobile phase by 0.1 and its
organic strength by 2% did not have a significant effect on
HPLC chromatographic resolution. Variation of strength of
0.1 M hydrochloric acid by 0.02 M did not have a signifi-
cant effect on spectrophotometric methods for determination
of VP and VA.
3.4.3. Range
The calibration range was established through consider-
ation of the practical range necessary, according to FB and
VP concentration present in pharmaceutical product, to give
accurate, precise and linear results. The calibration ranges of
the proposed methods are given in Tables 1 and 2.
3.4.8. Stability
Table 6
The FB and FA solutions in mobile phase or methanol
exhibited no chromatographic or spectrophotometric changes
for 6 h when kept at room temperature and for 3 days when
stored refrigerated at 5 °C.
Hydrolysis rate constant (kobs) and half-life (t1/2) for FB and VP in Britton–
Robinson buffer at different pH values and 80 °C
PH
kobs (h–1
)
t1/2(h)
VP
FB
VP
FB
The VP and VA solutions in mobile phase or 0.1 M hydro-
chloric acid exhibited no chromatographic or spectrophoto-
metric changes for 4 h when kept at room temperature, and
for 2 days when stored refrigerated at 5 °C.
2.0
2.9
3.6
4.5
5.8
6.7
8.0
9.0
10.5
12.0
0.196
0.116
0.023
0.072
0.157
0.368
0.668
1.373
1.616
1.938
0.166
0.085
0.017
0.051
0.117
0.215
0.457
0.937
1.216
1.534
3.540
6.018
29.501
9.676
4.425
1.881
1.038
0.505
0.429
0.358
4.181
8.141
39.620
13.424
5.943
3.230
1.517
0.739
0.570
0.452
4. Conclusion
2
1
1
The proposed HPLC, D or D, DD, PCR and PLS-
1 methods provide simple, accurate and reproducible quanti-

438
A. El-Gindy et al. / Il Farmaco 60 (2005) 425–438
[14] P.M. Lacroix, B.A. Dawson, R.W. Sears, D.B. Black, T.D. Cyr,
J.C. Ethier, Fenofibrate raw materials: HPLC methods for assay and
purity and an NMR method for purity, J. Pharm. Biomed. Anal. 18
(1998) 383–402.
[15] M. Gazdag, G. Szepesi, K. Csomor, High-performance liquid chro-
matography of eburnane alkaloids. III. Application of different phase
systems, J. Chromatogr. 243 (1981) 315–322.
[16] X. Wang, Determination of vinpocetine and related substances in
pharmaceutical tablets by HPLC, Yaowu Fenxi Zazhi 15 (1995)
23–26.
[17] A. Abd Elbary, N. Foda, O. El-Gazayerly, M. El Khatib, Reversed-
phase liquid chromatographic determination of vinpocetine in human
plasma and its pharmacokinetic applications, Anal. Lett. 35 (2002)
1041–1054.
[18] B. Herenyi, S. Goeroeg, Chiral high-performance liquid-
chromatographic separations on an 1-acid glycoprotein column. II.
Separation of the diastereomeric and enantiomeric analogues of vin-
pocetine (Cavinton), J. Chromatogr. 592 (1992) 297–299.
[19] B. Reck, E. Dingler, A. Lohmann, Development of a sensitive enzyme
immunoassay for the determination of vinpocetine in human plasma
Arzneim.-, Forsch. 42 (1992) 1171–1174.
[20] M. Vatsova, S. Tzvetanov, A. Drenska, J. Goranscheva,
N. Tyutyulkova, Improved gas chromatographic–mass-spectrometric
method for the quantitative determination of vinpocetine in human
plasma, J.Chromatogr. B, Biomed. Appl. 702 (1997) 221–226.
[21] A. Lohmann, E. Dingler, New assay method for the determination of
vinpocetine in human plasma by gas chromatography–mass spec-
trometry without trans-esterification caused by solvents. Reply to
Hammes and Weyhenmeyer, J. Chromatogr. Biomed. Appl. 105
(1991) 506–507 (2 (J. Chromatogr. 567)).
[22] W. Hammes, R. Weyhenmeyer, New assay method for the determina-
tion of vinpocetine in human plasma by gas chromatography–mass
spectrometry without trans-esterification caused by solvents: a reply,
J. Chromatogr. Biomed. Appl. 105 (1991) 504–505 (2 (J. Chromatogr.
567)).
[23] A. Lohmann, E. Dingler, New assay method for the determination of
vinpocetine in human plasma with gas chromatography–mass spec-
trometry without transesterification caused by solvents, J. Chro-
matogr. Biomed. Appl. 94 (1990) 442–448 (2 (J. Chromatogr. 529)).
[24] W. Hammes, R. Weyhenmeyer, Quantitative determination of vinpo-
cetine in human plasma by capillary gas chromatography–mass spec-
trometry, J. Chromatogr. Biomed. Appl. 57 (1987) 264–269 (J. Chro-
matogr. 413).
tative analysis for the determination of FB, VP and their
hydrolysis products in their pharmaceutical dosage form,
without any interference from dosage form excipients. The
HPLC method was found to be more specific than the spec-
trophotometric methods. While the spectrophotometric meth-
ods are less expensive methods and they do not require sophis-
ticated instrumentation and any prior separation step. The
PLS-1 and PCR approaches used in this work are simple to
perform, with adequate software support and provides a clear
example of the high resolving power of this technique. It was
found that FB and VP are rapidly hydrolyzed in alkaline
medium, while they are more stable in acid medium. The most
hydrolytic stability of FB and VP was found to be at pH 3.6.
References
[1] S.C. Sweetman, Martindale—The Complete Drug Reference, Thirty-
third ed, Pharmaceutical Press, 2002.
[2] K. Reddersen, T. Heberer, Multi-compound methods for the detection
of pharmaceutical residues in various waters applying solid phase
extraction (SPE) and gas chromatography with mass spectrometric
(GC-MS) detection, J. Sep. Sci. 26 (2003) 1443–1450.
[3] F. Sacher, F.T. Lange, H.J. Brauch, I. Blankenhorn, Pharmaceuticals
in groundwaters. Analytical methods and results of a monitoring
program in Baden-Wurttemberg, Germany, J. Chromatogr. A. 938
(2001) 199–210.
[4] D.H. Calan, et al., British Pharmacopeia, Her Majesty Stationery
Office, London, 2001.
[5] D. Schnädelbach, et al., European Pharmacopoeia, Fourth ed, Council
of Europe, Strasbourg, France, 2002.
[6] L.A. Romanyshyn, P.R. Tiller, Ultra-short columns and ballistic
gradients: considerations for ultra-fast chromatographic liquid
chromatographic-tandem mass spectrometric analysis, J. Chromatogr.
A. 928 (2001) 41–51.
[7] G.R. Fan, M. Lin, Z.X. Zhang, D.K. An, Determination of the fenofi-
brate bioactive metabolite, fenofibric acid, in human plasma by
HPLC, Yaowu Fenxi Zazhi 20 (2000) 231–234.
[8] B. Streel, P. Hubert, A. Ceccato, Determination of fenofibric acid in
human plasma using automated solid-phase extraction coupled to
liquid chromatography, J. Chromatogr. B Biomed. Appl. 742 (2000)
391–400.
[9] S. Abe, K. Ono, M. Mogi, T. Hayashi, High-performance liquid-
chromatographic method for the determination of fenofibric acid and
reduced fenofibric acid in human blood, plasma and urine, Yakugaku
Zasshi 118 (1998) 447–455.
[25] B.M. Wise, N.B. Gallagher, PLS-Toolbox Version 2.1, Eigenvector
Research, Inc., 830 Wapato Lake road, Manson, W A 98831, 2000.
[26] M. Brezina, P. Zuman, Polarography in Medicine, Biochemistry, and
Pharmacy, Interscience, New York, 1958 p.731.
[27] M. Blanco, J. Coello, F. Gonzalez, H. Iturriaga, S. Maspoch, Spectro-
photometric analysis of a pharmaceutical preparation by principal
component regression, J. Pharm. Sci. 82 (1993) 834–837.
[28] A. Espinosa-Mansilla, F. salinas, I. De Orbe Paya, Simultaneous
determination of sulfadiazine, doxycycline, furaltadone and trimetho-
prim by partial least squares multivariate calibration, Anal. Chim.
Acta 313 (1995) 103–112.
[10] L.D. Masnatta, L.A. Cuniberti, R.H. Rey, J.P. Werba, Determination
of bezafibrate, ciprofibrate and fenofibric acid in human plasma by
high-performance liquid chromatography, J. Chromatogr. B Biomed.
Appl. 687 (1996) 437–442.
[11] A. Cotta Ramusino, A. Carozzi, Simple and rapid method for deter-
mining procetofenic acid, an active metabolite of procetofen [fenofi-
brate], in biological fluids by solid-phase extraction and high-
performance liquid chromatography, J. Chromatogr. Biomed. Appl.
56 (1986) 419–424 (2 (J. Chromatogr. 383)).
[12] R.N. Rao, V. Nagaraju, An overview of the recent trends in develop-
ment of HPLC methods for determination of impurities in drugs, J.
Pharm. Biomed. Anal. 33 (2003) 335–377.
[29] D.M. Haaland, E.V. Thomas, Partial least-squares methods for spec-
tral analyses. 1. Relation to other quantitative calibration methods and
the extraction of qualitative information, Anal. Chem. 60 (1988)
1193–1202.
[30] J.A. De Schutter, P. De Moerloose, Determination of oxybutynin
hydrochloride in pharmaceuticals by reversed-phase ion-pair liquid
chromatography with two counter-ions in the eluent, J. Chromatogr.
450 (1988) 337–342.
[13] X. Ji, C.Ye, Determination of the impurities and content of fenofibrate
by reversed-phase HPLC, ZhongguoYaoke Daxue Xuebao 18 (1987)
284–286.
[31] The European Agency for The Evaluation of Medical Products, ICH
Topic Q2B Note for Guidance onValidation ofAnalytical Procedures:
Methodology GPMP/ICH/281/95,1996.