Entrapment of multiple anti-Tb drugs in microemulsion system: Quantitative analysis, stability, and in vitro release studies

Article abstract of DOI:10.1002/jps.21971

The use of multiple drug regimes for the treatment of tuberculosis is restricted because of the degradation of Rifampicin in the presence of Isoniazid. In the present report, the microemulsion comprising Tween 80/ethanol/oleic acid/Buffer (pH 7.4) has been investigated for encapsulating Rifampicin (RIF), Isoniazid (INH), and Pyrazinamide (PZA) in different combinations. The main idea is to use the compartment of different domains of microemulsion for encapsulating drugs of different solubilties. The structural changes in the microstructure of microemulsion have been investigated in the light of the changes occurring in the presence of ATDs (anti-Tb drugs) in binary or ternary mixtures through conductivity and viscosity. Optical microscopy and ¹H NMR helped in estimating the shape and location of ATDs in microemulsion system. Electronic spectroscopy has been utilized qualitatively and quantitatively for understanding the stability and release of RIF from the composed formulation with various binary or ternary mixtures. Dissolution and release kinetics have been carried out to get an idea on the release of drugs from microemulsion formulation. ATDs in single and mixed drug formulations followed non-Fickian release behavior except for RIF in pH 7.4 release medium. The investigations employed have given us a fair success to predict not only the stability but also the release mechanism of ATDs in single and mixed drug formulations.

Full text of DOI:10.1002/jps.21971

Entrapment of Multiple Anti-Tb Drugs in Microemulsion
System: Quantitative Analysis, Stability, and In Vitro
Release Studies
S.K. MEHTA, GURPREET KAUR, K.K. BHASIN
Department of Chemistry and Center of Advanced Studies in Chemistry, Panjab University, Chandigarh 160 014, India
Received 7 July 2009; revised 12 August 2009; accepted 7 September 2009
Published online 5 November 2009 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.21971
ABSTRACT: The use of multiple drug regimes for the treatment of tuberculosis is
restricted because of the degradation of Rifampicin in the presence of Isoniazid. In
the present report, the microemulsion comprising Tween 80/ethanol/oleic acid/Buffer
(pH 7.4) has been investigated for encapsulating Rifampicin (RIF), Isoniazid (INH), and
Pyrazinamide (PZA) in different combinations. The main idea is to use the compartment
of different domains of microemulsion for encapsulating drugs of different solubilties.
The structural changes in the microstructure of microemulsion have been investigated
in the light of the changes occurring in the presence of ATDs (anti-Tb drugs) in binary or
1
ternary mixtures through conductivity and viscosity. Optical microscopy and H NMR
helped in estimating the shape and location of ATDs in microemulsion system. Electro-
nic spectroscopy has been utilized qualitatively and quantitatively for understanding
the stability and release of RIF from the composed formulation with various binary or
ternary mixtures. Dissolution and release kinetics have been carried out to get an idea
on the release of drugs from microemulsion formulation. ATDs in single and mixed drug
formulations followed non-Fickian release behavior except for RIF in pH 7.4 release
medium. The investigations employed have given us a fair success to predict not only the
stability but also the release mechanism of ATDs in single and mixed drug formulations.
ß 2009 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 99:1896–1911,
2010
Keywords: microemulsion; Rifampicin; Isoniazid; Pyrazinamide; stability; ratio
derivative method
INTRODUCTION
be treated using combination of three or four
drugs over a period of at least 6–9 months.1
Among the first line drugs RIF shows wide
variation in the solubility at physiologic pH. It is a
borderline class II drug of Biopharmaceutic
Classification System (BCS). It has been reported2
that RIF is well absorbed from stomach because of
its high solubility between pH 1 and 2. RIF is
available as oral formulation in the form of both
single and FDC. However, major hurdle in its use,
is its varied plasma profiles and unacceptable
bioavailability in comparison to the same dose
separate formulations.3 The major obstacle in
FDC use is degradation of RIF in acidic medium,
which is mainly triggered4 by INH. Also sometime
due to inappropriate prescribing or taking of
Tuberculosis is a systemic disease that can
potentially affect any organ system. The lung is
the major portal of entry for mycobacterium
tuberculosis. Rifampicin, Isoniazid, Pyrazina-
mide, and Ethambutol are the First line drugs
used for the treatment. Single drugs are effective
for prophylactic treatment but active disease must
Additional Supporting Information may be found in the
online version of this article.
Correspondence to: S.K. Mehta (Telephone: 91-172-
2534423; Fax: 91-172-2545074; E-mail: skmehta@pu.ac.in)
Journal of Pharmaceutical Sciences, Vol. 99, 1896–1911 (2010)
ß 2009 Wiley-Liss, Inc. and the American Pharmacists Association
1896
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 4, APRIL 2010

ENTRAPMENT OF MULTIPLE ANTI-Tb DRUGS IN MICROEMULSION SYSTEM
1897
medications, resistance to these first line drugs
occurs in monotherapy. In order to avoid this, it
has been recommended to use these drugs in
combinations. Fixed dose combination (FDC) of
two, three, or four drugs is a preferred dosage form
for better patient compliance, efficient reduction
in viable bacterial population and minimizing
development of resistance to antitubercular
drugs. Although separate formulations have been
prescribed for TB control, however, World health
organization (WHO) and International Union
against Tuberculosis and Lung Disease (IUATLD)
have emphasized the implementation of FDC
formulation.3
with lower release rates obtained from formula-
tions with high viscosity. A positively charged
microemulsion composed of lecithin and/or decyl
polyglycoside and a cationic charge-inducing
agent strarylamine, L-alanine benzyl ester or
CTAB, for the topical application of miconazole
nitrate (MVZ) has also been reported.11 The results
suggested that microemulsion formulation could be
used to optimize drug targeting without an increase
in systemic absorption. Like wise some successful
formulations for different drugs like, hydrocorti-
sone,12 lidocaine,13 and paclitaxel14 using wide
variety of surfactants in various microemulsion
formulations have also been reported.
Literature5 reveals that there are various
physical and chemical reasons involved in the
formation of a stable fixed dose combination.
Among various problems like adsorption of drug
by excipients and formulation factors, the most
concerned is the decomposition of RIF. It has been
recently reported6 that three or four drugs
combination containing RIF, INH along with
PZA and/or ethambutol showed more chemical
instability than two-drug FDCs containing RIF
and INH. Gohell and Sarvaiya7 have developed
the solid dosage form (tablet and capsule in
capsule) comprising a gastroretentive modified
release RIF tablet and an enteric-coated isoniazid
capsule.
Controlled release of therapeutic compounds is
of general interest, because of its ability to main-
tain the concentration of a drug in the desired
therapeutic range and to localize delivery of the
drug. Attempt8 to form successful formulations
has also been made in the form of suspensions,
capsule, tablet, microspheres, or nanoparticles, etc.
Carriers assembled using hydrophobic/hydrophilic
interactions include liposomes (vesicles), micelles,
and microemulsions. Mostly available drugs are
hydrophobic (such as paclitaxel, RIF, etc.) and
cannot be dissolved in aqueous buffers. For these
compounds, micelles and microemulsions are of
great interest as possible carriers. Formariz et al.9
worked on an anti tumor drug Doxorubicin (DOX)
and have found the relation between structural
features and in vitro release from biocompatible
anionic and neutral microemulsions containing
soya phosphatidylcholine. DOX release ratio was
directly dependent on the ratio of oil to surfactant.
For the ocular delivery of pilocarpine hydrochlor-
ide, Chan et al.10 proposed a mixed surfactant
system consisting of sorbitan mono laurate and
polyoxyethylene sorbitan mono-oleate. It was
found that drug release depended on the viscosity,
In the present work, the use of microemulsion
for the combination formulation of two drug and
three drug mixtures has been made. The main
idea is to explore the compartmentalization of
microemulsion system, which is due to the
presence of w/o or o/w droplets, as a reservoir
for drugs of diverse solubilties. Since, there are
separate domains of hydrophilic and hydrophobic
components separated by monolayer of surfactant
(ideally), these microemulsion systems have
proved beneficial for solubility of both labile and
poorly water soluble drugs. Microemulsions also
provide protection against enzymatic degrada-
tion, light, enhances solubility and stability, etc.15
In this article, Tween-based microemulsion has
been used to encapsulate rifampicin (RIF) þ
isoniazid (INH) and rifampicin (RIF) þ pyrazinamide
pyrazinamide (PZA) in binary mixture formula-
tions and (RIF þ INH þ PZA) in ternary mixture
formulations (in equal weight ratios) (Fig. 1).
Tween 80 is a PEG-based surfactant often used as
a food additive, our rationale for including this
surfactant is based on the well-established fact
that PEG polymers tend to improve biocompat-
ibility of materials and surfaces, and to retard
circulatory clearance.16
The study explores the structural and stability
aspects of the microemulsion system with dilution
and in the presence of other anti Tb drugs in
different combinations. The transformation of
structure was studied with different physico-
chemical methods (viz., conductivity, viscosity).
Spectroscopic view was explored to get the idea of
presence and stability of different drugs in different
regimes. Selected drug loaded microemulsions have
been studied for there in vitro release behavior by
the use of ratio derivative method. The objective of
this initiative is to encourage basic research aimed
at developing new modalities of treatment for
Tuberculosis.
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 4, APRIL 2010

1898
MEHTA, KAUR, AND BHASIN
Figure 1. Chemical structures of (a) Rifampicin (RIF), (b) Isoniazid (INH), and
(c) Pyrazinamide (PZA).
MATERIALS AND METHODS
Materials
Centrifugation
Physical stability of pure and drug loaded micro-
emulsions were tested by carrying out centrifuga-
tion (Tarson make) at 2000 rpm for 30 min.
Tween 80 (polyoxyethylene sorbitan monooleate),
RIF, INH, and PZA were purchased from
Sigma, Missouri, MO. Absolute ethanol and oleic
acid were obtained from Merck and CDH respec-
tively. Phosphate buffer (PB) (0.01M, pH 7.4) was
used as the hydrophilic phase.
Conductivity Measurements
The electric conductivity (s) was measured by
means of a PICO digital conductivity meter
(Labindia Instruments Pvt Ltd., Maharashtra,
India) operating at 50 Hz, maintained by a RE320
Ecoline thermostat at 30 ꢀ 0.018C. The error limit
of conductance measurements was ꢀ3%. The
conductivity of selected and drug loaded micro-
emulsions were also measured as a function of v.
Preparation of Microemulsion
The composition of microemulsion has been
selected from the isotropic region of the phase
diagram previously reported.17 The system is
composed of Tween 80/Ethanol (1:2)/oleic acid/PB
with v ¼ 0.74 (where v is molar ratio of water to
surfactant and cosurfactant).
Viscosity Measurements
Viscosities were measured with calibrated Ubbe-
lohde viscometer at 30 ꢀ 0.018C, with the experi-
mental error of 0.5%. For each measurement, the
viscometer was washed, rinsed and vacuum dried.
To follow the viscous behavior of the microemul-
sions, flow time was measured for all the selected
pure and drug loaded microemulsions.
Drug Incorporation in Microemulsion
For the preparation of drug loaded microemul-
sion, in binary (RIF þ INH and RIF þ PZA) or
ternary (RIF þ INH þ PZA) mixtures, RIF was
dissolved in lipophilic phase and other two drugs
were dissolved in hydrophilic component, followed
by the addition of remaining components of the
microemulsion. Each drug was 1 wt% of the total
weight of microemulsion.
Optical Microscopy
The optical observations were made in trans-
mitted light with a Leica microscope DM 3000
(Germany). The images were recorded with Leica
DFC420 camera. The sample was placed on slide
so as to minimize possible destruction.
Optical Transparency
The homogeneity and optical isotropy of pure and
drug loaded microemulsions were examined by a
cross polarizer and visual examination at room
temperature. Samples were analyzed for any
precipitation or phase separation in the presence
of drugs.
Electronic Spectroscopy
UV–visible spectra were recorded over the wave-
length range of 200–600 nm using JASCO-530
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 4, APRIL 2010
DOI 10.1002/jps

ENTRAPMENT OF MULTIPLE ANTI-Tb DRUGS IN MICROEMULSION SYSTEM
1899
UV–visible spectrophotometer. Three sets of
experiments were performed in which concentra-
tion RIF was varied in absence and presence of
INH and PZA. Spectra of RIF were recorded
against empty microemulsion to rule out the
interference of components.
spectrum of a standard solution of one of the
compounds to obtain the ratio spectrum. For
determining, the concentration of each component
the first derivative spectra of the ratio spectra was
calculated using Origin 6.0. The Savitzky-Golay
smoothing was employed taking 10 experimental
points. For binary mixture of RIF with PZA and
INH (for different concentrations and synthetic
mixture), the ratio spectra were obtained for
binary mixtures of RIF þ PZA and RIF þ INH by
dividing with PZA (5 mM) and INH (5 mM),
respectively. Similarly, for INH and PZA (for
different concentrations and synthetic mixture),
the ratio spectra were obtained by dividing RIF
(25 mM).
Sample preparation:
(i) Stability test of RIF in presence of other
drugs: Three sets of six standard solutions
were prepared, each containing 3.0 g micro-
emulsion loaded with RIF (varying concen-
tration 0.01–0.04 mM in microemulsion).
INH and PZA were coencapsulated with
RIF at a constant concentration of 0.23
and 0.25 mM, respectively.
(ii) Stability test of RIF with time: 6.0 g micro-
emulsion was prepared containing 1 wt% of
each drug (whether single or mixture for-
mulations). 0.1 g of each drug loaded micro-
emulsion were transferred into 10 mL
volumetric flask and diluted to volume with
ethanol. The absorbance of RIF (in single,
binary or ternary mixture formulations)
was analyzed using spectrophotometer
initially for 2 h and there after for 20 days.
The samples were kept at room tempera-
ture (25–308C).
Ternary Mixtures. The analysis of ternary
mixture of drugs was carried out using double
divisor ratio derivative method (DDRD). In this
method the coincident spectra of the derivative of
the ratio spectra is obtained by using a ‘‘double
divisor’’ (sum of two spectra) and measuring
absorbance at either the maximum or minimum
wavelengths. For determining the concentration
of each component, the absorbance spectra of the
mixture was divided by the double divisor of
standard spectrum to obtain ratio spectra. The
first derivative spectra were then calculated with
Origin 6.0. The Savitzky-Golay smoothing was
employed taking 10 experimental points. In case
of RIF (for different concentrations and synthetic
mixture), the ratio spectra were obtained by
dividing with double divisor PZA (5 mM) and
INH (5 mM). Similarly, for INH and PZA (for
different concentrations and synthetic mixture)
the ratio spectra were obtained by dividing with
double divisor RIF (5 mM) and PZA (5 mM) and RIF
(5 mM) and INH (5 mM), respectively.
1H NMR
1H NMR measurements were performed on a
Bruker 400 system. Chemical shifts of surfactant
in the presence of two or three drugs (1 wt%) were
determined using D2O as internal locking agent.
The concentrations of RIF, INH, and PZA were
calculated by measuring amplitude of either
maxima or minima at working l in the first
derivative ratio spectra in the range of 350–
550 nm for RIF and 200–335 nm for INH and PZA
in both the methods.
In Vitro Release
Analysis of a Multi-Component System by Divisor
Ratio Spectra Method
Preparation of Stock Solutions and Syn-
thetic Mixtures. Stock solutions of RIF
(250 mM), INH (100 mM), and PZA (100 mM) were
prepared in PB. The appropriate dilutions were
done with PB to obtain suitable concentrations.
Release Behavior and Kinetics
In vitro release was determined by the dialysis bag
method18 using cellulose tubing (avg. flat width
33 mm (1.3 in.)). To perform the study, 1.0 g
microemulsion containing 1 wt% of binary or
ternary drugs loaded microemulsion formulation
was introduced in 50 mL of PB (pH 7.4) main-
tained in a release cell at 37 ꢀ 0.18C. The release
Binary Systems. For binary mixtures of drug,
the ratio derivative (RD) method has been
employed. In RD method, the absorption spectrum
of the mixture is divided by the absorption
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 4, APRIL 2010

1900
MEHTA, KAUR, AND BHASIN
cell is made of a double wall jacket leading to an
efficient thermostating and allowing easy collec-
tion of aliquots. The system was stirred at
constant speed of 100 rpm. The aliquot of 2 mL
was withdrawn at regular intervals for 5 h and
was filtered through 0.45 mm membrane. Each
time the same amount of fresh PB (kept at the
same temperature) was added to maintain the
volume of the release medium. The drug concen-
trations were determined spectrophotometrically
at their characteristic wavelengths (i.e., RIF
475 nm, INH 262 nm, and PZA 26 8 nm) for single
loaded formulations. The RD and DDRD method
was used for determination of concentrations in
binary or ternary mixtures, respectively.
Drug release was determined by two para-
meters, first cumulative percent drug released
after 5 h and second the release rate of drug
expressed as slope of the linear regression line.
The order of the release kinetics was determined
by comparing linear regression line for different
rate equations. All the experiments were per-
formed in triplicate.
occurring in the microstructure of the microemul-
sion are manifested by the change in the slope of
plot of log s versus v (Fig. 2). For pure micro-
emulsion, the first change occurs at v ¼ 1.41,
owing to the change of w/o droplet into bicontin-
uous phase which consist of interwoven water and
oil channels separated by a surfactant monolayer.
The second change in the slope at v ¼ 2.94 depicts
the formation of o/w droplets from the bicontin-
uous phase.
In case of binary and ternary mixture formula-
tions of microemulsion, similar changes in the
slope at v ¼ 1.32 and 2.94 (Fig. 2b) reckon the
same structural changes in the microstructure.
The conductivity profiles reveal that no significant
change was observed in the w/o type region (for
pure or drug loaded formulations), except for the
early phase change in presence of drugs in
comparison to the pure system. However, for
RIF þ INH and RIF þ PZA, a slight deviation
appeared as compared to empty microemulsion
(Fig. 3) in the o/w type and bicontinuous region,
respectively. This speculation leads to the fact
that after the phase reversal the water soluble
drugs, that is, INH and PZA, are present in
dispersion medium and are contributing to the
enhancement of the conductance. For the ternary
mixture formulation, a significant increase in the
conductance is observed, once the aqueous phase
becomes the continuum phase. Also the presence
of drugs helps in early onset of process of
percolation (clustering of droplets).
RESULTS AND DISCUSSION
Conductivity
To understand the structural transition in the
microstructure of microemulsion with dilution
and in presence of all the three drugs in
combination, the conductivity experiments were
performed. The microemulsion was selected from
the single phase isotropic region of the phase
diagram, reported earlier.17 The results show that
upon dilution conductivity of the microemulsion
increases, that is, as water to surfactant ratio (v)
increases conductance increases. The changes
Viscosity Measurements
Generally, swelling microemulsion droplets tend
to self-interact more and more due to the
increasing volume and entanglement of the
Figure 2. Variation of log conductance in selected microemulsion (ME) formulation.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 4, APRIL 2010
DOI 10.1002/jps

ENTRAPMENT OF MULTIPLE ANTI-Tb DRUGS IN MICROEMULSION SYSTEM
1901
When the weight fraction of the aqueous phase
is low, it is well known that the interaction
between the globules is low and the microemul-
sion consists of isolated dispersed globules (hard
spheres) in the continuous oil medium.22 How-
ever, viscosity profiles mainly reflect structural
interactions and not necessarily the points at
which the transition occurs or is completed. The
viscosity is very low when the ‘‘water swollen
micelles’’ are dispersed in the oil, and is increasing
progressively.
With the addition of binary and ternary
mixtures of drugs in microemulsion formulations,
the viscosity values were found to be higher than
that of empty microemulsion systems. Since it is
believed that the viscosity measurements are
sensitive to interactions, the presence of both
hydrophilic and hydrophobic drugs affects the
viscosity throughout. Thus, indicating that the
presence of drug is affecting interdroplet interac-
tions.
Figure 3. Comparison of conductivity behavior of
empty, RIF þ PZA, RIF þ INH, and ternary microemul-
sion (ME) formulations.
aggregates.19 Such an increase in viscosity was
observed, for instance, in ethoxylated nonionic
microemulsions20 and lecithin-based (phosphati-
dylcholine) systems.21 The viscosity measure-
ments carried out for empty and drug loaded
microemulsion formulations are depicted in
Figure 4a. The dilutions were carried with PB
as aqueous phase. The increase in the viscosity
with the increase of the aqueous phase content is
either due to an increase in the dispersed globule
sizes, or due to the formation of clusters of
globules that are sufficiently close to each other,
and increases the interaction between them.
The plots of d2h/d2f versus v, reflects that the
transition from w/o to bicontinuous for at empty
and three drugs occur at v ¼ 1.41 and 1.32,
respectively (Fig. 4b). This transition point is
same as observed in the conductivity variations.
Optical Microscopy
The optical microscopic images of pure (Fig. 5a)
and drug loaded microemulsions (single and
combination Fig. 5b–e) have helped in under-
standing the behavior of microstructure of micro-
Figure 4. Plots of (a) kinematic viscosity (b) d2h/d2f versus v, showing the structural
change occurring in the microstructure of empty and drug loaded microemulsion (ME)
system.
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 4, APRIL 2010

1902
MEHTA, KAUR, AND BHASIN
Figure 5. Photomicrographs of (a) pure, (b) RIF, (c) RIF þ PZA, (d) RIF þ INH, and (e)
ternary mixture loaded microemulsion (ME) formulations in selected microemulsion
system.
emulsion in the presence of an external entity. As
mentioned earlier the microemulsion formulation
with v ¼ 0.74 was selected. Although the chosen
microemulsion system was polydisperse in nat-
ure, however, the spherical droplets (as expected
with critical packing parameter less than 1/3)23
could be easily seen through optical images. The
presence of drugs either in single, binary or
ternary mixtures in the microemulsion does not
rupture the microstructure as can be seen from
the presence of droplets in these drug loaded
microemulsion systems.
location of RIF in the microemulsion through
1H NMR spectroscopy. Since the microemulsion
system contains the components with long hydro-
1
carbon and oxyethylene groups, H NMR spectra
was expected to be complex. In order to avoid the
complexity, the experiments were performed with
surfactant in D2O having different combination of
drugs. The main results are summarized in
Table 1.
RIF is a bulkier molecule and therefore, most of
its peaks overlapped with the peaks of surfactant.
Presence of RIF (0.5%) does not show change in
the major peaks of surfactant (Fig. 6) except for
protons of oxyethylene peaks and small downfield
shifts for b1-CH2, CH3 and upfield shift for (CH2)n.
Whereas, the augmentation of RIF concentration
(1.0%) in the surfactant system showed the
enhancement in values of Dd for all the functional
1H NMR Spectroscopy
An attempt has been made to study the effect of
INH, PZA, and combination of both drugs on the
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 4, APRIL 2010
DOI 10.1002/jps

ENTRAPMENT OF MULTIPLE ANTI-Tb DRUGS IN MICROEMULSION SYSTEM
1903
moieties. Chemical shift of the order of 0.142 is
observed at low concentration (0.5%), however, it
is relatively negligible at 1% and in presence of
other drugs. The probable reason for decrease in
the chemical shift is due to lesser availability of
surfactant molecules. As the concentration of drug
increases, surfactant molecules are replaced by
drug molecules at the interface. In presence of the
two hydrophilic drugs (1.0%), the change in Dd is
same as that for pure RIF, only the upfield shift for
protons of oxyethylene peaks were observed along
a small change in Dd for a1-CH2, and a-CH2
functional moieties in PZA þ RIF system. The
three drug combination showed the cumulative
changes in Dd throughout. From the earlier
studies,24 it has been concluded that most of the
INH molecules showed insignificant shifts in
peaks of major moieties of surfactant ruling out
its probability near interface. Since the shifts are
most significant in presence of RIF that indicates
the probability of RIF being present at the
interface is larger whereas PZA is more or less
near the palisade layer. For mixed drug formula-
tions, it has been observed that INH or PZA does
not show any change in the Dd values of RIF
surfactant mixture, except for upfield shift of
protons of oxyethylene peaks in all the three
mixed drug combinations. This infers that the
presence of either of INH, PZA or both does not
affect the presence of RIF in the proximity of
surfactant.
Absorption Spectroscopy
UV–visible spectroscopy has been used to solve
the problem of the spectral overlap. It is easy and
sensitive method for direct estimation of drug
amount and gives information about the changes
in physical state of the drug.25 In the present
study, derivative spectrophotometry has been
utilized for extracting both qualitative and
quantitative information from spectra by using
the first derivatives of absorbance with respect to
wavelength.
Stability Studies
The stability of RIF in the presence of drugs (INH
and PZA) and with span of time was analyzed for
the selected drug loaded microemulsion formula-
tion using absorption spectroscopy. It was easy for
RIF since its characteristic peak at 475 nm was
used for the analysis (against blank microemul-
sion), however, it was difficulty for INH and PZA
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 4, APRIL 2010

1904
MEHTA, KAUR, AND BHASIN
Figure 6. Structure and NMR spectrum of Tween 80.
as their peaks were at 262 and 268 nm, respec-
tively, which were below the cut off range of the
microemulsion system. Therefore, their analysis
was carried out with dilution with ethanol (as
reference).
Figure 7 depicts the UV–visible spectra of RIF
in combination with the other drugs in formulated
drug loaded microemulsion system. No significant
difference was found between the slopes of the
working curves for the RIF either in single or in
binary mixture (RIF þ constant INH and
RIF þ constant PZA) loaded microemulsion sys-
tem (Tab. 2). RIF encapsulated in microemulsion
thus showed same behavior even in the presence
of other two hydrophilic drugs as that of pure RIF
system indicative of no signs of degradation in
presence of INH or PZA.
The stability studies of all four drug loaded
formulations were carried out immediately after
preparation for 2 h (Fig. 8). No change in the lmax
(480 nm) for RIF was seen in any of the drug
loaded microemulsion formulation, that is, single,
binary or ternary mixture loaded. Neither, the
change in lmax of RIF nor the emergence of any
other peak, illustrated the stability of RIF in the
presence of other two hydrophilic drugs, in all the
three mixed drug formulations with time.
The stability study results carried out for
20 days shows lmax for RIF (single) and RIF þ PZA
PZA (binary) at 480 nm (near the characteristic
Figure 7. UV–visible spectra of selected microemulsion (ME) containing (a) RIF,
(b) RIF co-encapsulated with INH (0.23 mM), and (c) RIF co-encapsulated with PZA
(0.25 mM). The inset shows corresponding regression line in each plot.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 4, APRIL 2010
DOI 10.1002/jps

ENTRAPMENT OF MULTIPLE ANTI-Tb DRUGS IN MICROEMULSION SYSTEM
1905
Table 2. Statistical Analysis for Determination of RIF Alone (RIF þ constant INH) and (RIF þ constant PZA)
Encapsulated in Microemulsion by UV Spectroscopy
Drug
RIF Alone
RIF þ INH
RIF þ PZA
Slope
Intercept
Correlation coefficient
11.62
0.02
0.9917
12.26
ꢁ0.25
0.9919
10.52
ꢁ0.03
0.9989
Varying amount (0.01–0.04 mM) of RIF.
Varying amount (0.01–0.04 mM) of RIF and constant amount (0.23 mM) of INH.
Varying amount (0.01–0.04 mM) of RIF and constant amount (0.25 mM) of PZA.
peak of RIF 475 nm). Whereas the lmax for
RIF þ PZA þ INH (ternary) and RIF þ INH (bin-
ary) showed a red shift after day 1 (487 nm), that
indicated the formation of 3-Formyl rifampicin SV
(characteristic peak 492 nm) due to the hydrolysis
of RIF in presence of isoniazid.4 Figure 9 elabo-
rates the time study of all the formulations. UV
studies show that no significant change in
absorbance patterns or l was observed with time
except for RIF þ INH and combo formulations (on
the 20th day). In addition, no precipitation of drug
and their mixtures was observed. Thus, the
studies employed facilitate us to predict to a fair
success that in the formulated drug loaded
microemulsion RIF is stable for a good spam of
time even in presence of INH and protects the
drug from degradation.
component system.26 This technique offers an
alternative approach to the enhancement of
sensitivity and specificity in mixture analysis.
Salinas et al.27 have proposed the ratio-derivative
(RD) spectrophotometry, for the simultaneous
determination of two compounds in binary mix-
tures and Dinc28 has proposed the double divisor
ratio spectra derivative (DDRD) method for the
simultaneous determination of ternary mixtures.
In this work, the RD and DDRD methods have
been employed for the simultaneous determina-
tion of RIF, INH, and PZA in binary mixtures
(RIF þ INH, RIF þ PZA) and in ternary mixture
(RIF þ INH þ PZA). The method was further
applied for the assay of formed drug loaded
microemulsion formulations. Such studies for
optimizing different methods have recently been
In Vitro Release
Quantitative Assay of the Multi-Component System
Derivative spectrophotometry has been widely
used for the simultaneous analysis of a multi-
Figure 9. UV–visible spectra of time study of selected
microemulsion (ME) containing (a) RIF, (b) RIF þ INH,
(c) RIF þ PZA, and (d) ternary mixture loaded micro-
emulsion formulations.
Figure 8. UV–visible spectra of selected microemul-
sion (ME) containing RIF, RIF þ PZA, RIF þ INH and
ternary formulations for 2 h from the preparation.
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 4, APRIL 2010

1906
MEHTA, KAUR, AND BHASIN
reported on the same drugs by Youssef and
Maher.29 Details have been followed from there.
INH and PZA in ternary mixture were obtained by
dividing absorption of solutions of different
concentrations with RIF (5 mM) and PZA (5 mM)
and RIF (5 mM) and INH (5 mM), respectively
(Fig. 11c–f). Statistical analysis of the calibration
data for determination of RIF, PZA, and INH in
binary and ternary mixtures was obtained by
plotting the graphs obtained by the derivative
method (i.e., RD and DDRD) and concentrations of
RIF, PZA and INH. As observed in Table S1 the
plots give small intercepts and good correlation
coefficients. The plots showed good linear relation-
ships at different working l for the different
combinations.
To access the accuracy of both the methods,
three synthetic mixtures were prepared within
the linearity range and percentage recoveries
obtained for each component are given in Table 3.
Reasonable values of percentage recoveries show
the validity of the procedure for both binary and
ternary mixtures. The methods were further
utilized to evaluate the release from the formu-
lated microemulsion loaded with different combi-
nations of ATDs.
Binary Mixtures
Figure 10a and c shows the ratio spectra of RIF in
RIF þ INH, RIF þ PZA mixtures which were
obtained by dividing absorption of solutions of
different concentrations of RIF with INH (5 mM)
and PZA (5 mM), respectively. The first derivative
spectra are depicted in Figure 10b and d.
Similarly, the ratio spectra of INH in RIF þ INH
and PZA in RIF þ PZA mixtures were obtained by
dividing absorption of solutions of different
concentrations with RIF (25 mM). The ratio
spectra and first derivative plots are given in
Figure 10e–h.
Ternary Mixtures
Figure 11a and b shows the ratio spectra and first
derivative spectrum of RIF in ternary mixture
obtained by dividing absorption of solutions of
different concentrations of RIF with INH (5 mM)
and PZA (5 mM). Similarly, the ratio spectra of
Figure 10. Ratio spectra and First derivative spectra of different concentrations of
RIF (a and b), RIF (c and d), INH (e and f), PZA (g and h) with INH, PZA, RIF, RIF as
divisors respectively, in RD method.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 4, APRIL 2010
DOI 10.1002/jps

ENTRAPMENT OF MULTIPLE ANTI-Tb DRUGS IN MICROEMULSION SYSTEM
1907
Table 3. Percentage Recoveries of Different
Components in Various Synthetic Mixtures Determined
by RD and DDRD Methods
Mixtures
Percentage Recoveries
RIF in RIF þ INH
98.15
98.00
98.35
100.00
97.50
105.50
105.00
RIF in RIF þ PZA
INH in RIF þ INH
PZA in RIF þ PZA
RIF in ternary mixture
INH in ternary mixture
PZA in ternary mixture
The release of RIF shows an initial bust and
then becomes constant (Fig. 12a). Similar beha-
vior was observed for all the studied formulations.
Among the three first line anti-TB drugs, that is,
RIF, PZA, and INH, the extent of release were
higher for PZA and INH in comparison to RIF for
the single drug microemulsion formulation
(Fig. 12a). This stands to reason that the release
rate seems to depend on the partitioning of these
drugs from the hydrophobic phase into water,
since PZA and INH have high solubility compared
to RIF as indicated by partition coefficient values
studied earlier.24
The release of RIF loaded microemulsion
system was studied further in different pH
mediums as RIF shows variation of solubility
with pH.2 Different pH medium, that is, pH 1.2
(Simulated Gastric Fluid), 6.8 (Simulated intest-
inal fluid), and 7.4 were selected owing to different
physiologic pH of the GI tract. The effect of pH of
the medium on the different fixed dose formula-
tions (marketed products) has been reported by
Agrawal and Panchagnula.30 They stated that in
different pH medium the dissolution rate depends
on the interaction with the acidic excipients. It
was observed in the present study that decrease
in pH delayed the release of RIF (Fig. 12b). Since,
the composed (empty) microemulsion system
has pH 5.48,31 therefore, most likely the formation
of a complex is occurring, which readily disso-
ciates as pH increases. So the increase in pH
increases the extent of release but affects the rate
of release. The release profile was studied in the
presence of other ATDs (Fig. 12c), the extent of
release of RIF remained same in presence of PZA
but show slight increase after 1 h in case of binary
(RIF þ INH) and ternary mixtures.
Figure 11. Ratio spectra (a) and First derivative
spectra (b) of different concentrations of RIF (a and
b), INH (c and d), PZA (e and f) with PZA þ INH,
RIF þ PZA, RIF þ INH, as double divisors respectively,
in DDRD method.
Release Behavior
Dissolution is a dynamic process strongly depen-
dent on both the composition of medium and
hydrodynamics. The release was carried out in
binary and ternary mixture loaded drugs in the
microemulsion formulation systems and was
further compared with that of pure system.
On the other hand, the release profile of INH
shows (Fig. 12d) decrease in the extent of release
in the presence of RIF as compared to single drug
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 4, APRIL 2010

1908
MEHTA, KAUR, AND BHASIN
Kinetics and Release Mechanism
In order to investigate the mode of release from
the microemulsion system, the release data were
analyzed with different mathematical models:
Zero order kinetic (1), First order kinetic (2),
Higuchi kinetics (3), Korsmeyer and Peppas
equation (4), and first order exponential (5).
Q ¼ kot
(1)
(2)
ln Q ¼ ln Q0 ꢁ k1t
Q ¼ kHt1=2
(3)
(4)
(5)
Mt
¼ kptn
M
Q ¼ Q0 ꢁ b eꢁkt
where Q is the cumulative percent of drug
released at time t, Mt/M is the fraction of drug
released at time t, and k0, k1, kH, kp, and k are the
rate constants for respective models. Q0 and b (in
Eqs. 2 and 5) are constants.
The data have been compiled in Table 4
representing the rate constants and correlation
coefficient values. The values of correlation
coefficients depict that neither of the formulation
followed zero order kinetics. The best agreement
of the fitting was achieved with the equation of
first order exponential decay. The closeness of
correlation coefficient (R2) depicts that the release
of RIF, PZA, and INH in all the formulations
showed a typical profile of first order exponential
curve (Eq. 5) for cumulative percent of drug
released versus time t (Fig. 12) and depicting the
controlled release of drugs.32,33
The rate constant values show that as the pH
increases, rate decreases. Whereas in the pre-
sence of other drugs in mixed drugs formulations
(binary and ternary), PZA and INH show
enhanced release rates in comparison to single
drug formulations. In case of RIF, single, binary
(RIF þ PZA) and ternary drug formulations, the
release showed comparable rates except for
RIF þ INH where decrease in rate was observed.
By using Korsmeyer and Peppas equation (4),
attempt has been made to study the release
mechanism where kp is a kinetic constant which
incorporate structural and geometric character-
istics of the release device and n exponent is
related to the release mechanism termed as the
diffusional exponent. The value of n should be
equal to 0.5 for diffusional (Fickian) release, 1 for
zero-order kinetics and 0.5 < n < 1 for anomalous
Figure 12. Release profiles of various single and mul-
tiple combinations of drugs from selected microemul-
sion (ME) formulations (n ¼ 3).
formulation. Whereas, PZA release profile does
not show any considerable effect in the presence of
binary or ternary mixtures in the formulation
(Fig. 12e).
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 4, APRIL 2010
DOI 10.1002/jps

ENTRAPMENT OF MULTIPLE ANTI-Tb DRUGS IN MICROEMULSION SYSTEM
1909
Table 4. Rate Constants and Regression Correlations Using Rate Equations (1–5) for the Release of Drugs from
Selected Microemulsions
Zero Order
Kinetics
First Order
Kinetics
Higuchi
Kinetics
Exponential
First Order
KP Model
Drugs
R2
k0
R2
k1
R2
kH
R2
n
kp
R2
k
RIF (pH 1.2)
RIF (pH6.8)
RIF (pH7.4)
RIF þ PZA
0.79747 0.0209 0.90306 0.0027 0.89938 0.3433 0.93963 0.6392 0.0029 0.99690 0.0133
0.89688 0.0424 0.98250 0.0326 0.96357 0.6641 0.98706 0.6688 0.0028 0.99531 0.0116
0.95437 0.0519 0.98123 0.0400 0.98241 0.7602 0.98734 0.5007 0.0069 0.99690 0.0090
0.92810 0.0527 0.97949 0.0420 0.99142 0.7615 0.98337 0.7404 0.0023 0.99386 0.0101
0.96670 0.0737 0.93726 0.0574 0.99706 1.0655 0.99502 0.5682 0.0075 0.98496 0.0067
RIF þ INH
RIF þ INH þ PZA 0.94281 0.0705 0.97636 0.0552 0.99532 0.9539 0.96847 0.7803 0.0024 0.99323 0.0105
PZA
PZA þ RIF
0.58010 0.3998 0.93612 0.6372 0.76130 6.1768 0.91994 0.6059 0.0508 0.97734 0.0339
0.40987 0.3923 0.84830 0.3615 0.66785 6.8148 0.90687 0.6053 0.0602 0.98803 0.0585
PZA þ RIF þ INH 0.48105 0.4276 0.91775 0.4135 0.66891 6.3453 0.87910 0.7615 0.0329 0.95248 0.0519
INH
INH þ RIF
0.71623 0.3987 0.94748 0.4657 0.87052 5.7047 0.91930 0.6323 0.0366 0.98824 0.0230
0.66603 0.2971 0.97736 0.2065 0.83906 4.6067 0.92533 0.6062 0.0473 0.97980 0.0314
INH þ PZA þ RIF 0.71623 0.3987 0.90357 0.1242 0.87052 5.7047 0.95689 0.6330 0.0378 0.98824 0.2306
(non-Fickian) release. The values of n for INH and
PZA in all the respective drug loaded microemul-
sion formulations indicated a non-Fickian release
mechanism. Moreover, the presence of binary or
ternary mixtures did not affect the mechanism of
release. On the contrary RIF showed diffusional
release from the single drug formulation at pH 7.4
and also the change of medium and presence of
other drugs in the formulation changed the
behavior of release to anomalous as the value of
n was found to be greater than 0.5.
spectroscopy helped in determining the stability
of RIF in microemulsion system in the presence of
both INH and PZA and with the span of time.
Hence, these microemulsion formulations have
proved to be efficient in controlling the drug-drug
interactions to certain extend by showing no
change in physical or chemical stability. However,
Bhutani et al.34 have emphasized on the strong
interaction of these ATDs with each other in a
multiple and complex manner. The stability of the
chosen combinations in the present case has also
been supported by their studies on different
marketed products. It was shown that the release
of these drugs could be predicted to a certain
extent. Most of the composed formulations showed
anomalous behavior of drug release. Thus, pre-
sent study not only helps us to understand the
stability of different ATDs in the presence of each
other and with time but also predicts the probable
location of these drugs in the formulated Tween
microemulsion (Scheme 1).
CONCLUSION
It can be concluded that the selected Tween 80
based microemulsion has successfully encapsu-
lated the anti TB drugs in different combinations,
without any precipitation or phase separation.
The Tween based microemulsion has been able to
utilize its domains of diverse polarities to encap-
sulate these drugs, in order to minimize the
contact between RIF, INH, and PZA. The con-
ductivity and viscosity studies depicted the
continuous structural changes from w/o to o/w
occurring in the microemulsion system, both in
the absence and presence of drugs. The presence
of drugs only fastens the process of percolation.
Optical microscopy showed the same shape of
droplets in absence and presence of binary or
ternary mixtures of drugs. Spectroscopic techni-
ques offered an alternative approach to depict the
location of drug molecule in the proximity of
surfactant and nullifying the effect of INH and
PZA on the position of RIF. The electronic
Scheme 1. Showing the probable location of all the
three drugs in Tween 80 microemulsion system.
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 4, APRIL 2010

1910
MEHTA, KAUR, AND BHASIN
ACKNOWLEDGMENTS
11. Peira E, Carlotti ME, Trotta C, Cavalli R, Trotta M.
2008. Positively charged microemulsions for topical
application. Int J Pharm 346:119–123.
S.K.M. is thankful to DST India for financial
assistance. G. Kaur is thankful to CSIR for senior
research fellowship. Authors are thankful to
RSIC, Panjab University, Chandigarh for provid-
ing optical images using Leica microscope DM
3000 (Germany).
12. Maghraby GME. 2008. Transdermal delivery of
hydrocortisone from eucalyptus oil microemulsion:
Effects of cosurfactants. Int J Pharm 355:285–292.
13. Yuan JS, Ansari M, Samaan M, Acosta EJ. 2007.
Linker-based lecithin microemulsions for transder-
mal delivery of lidocaine. Int J Pharm 349:130–143.
14. Nornoo AO, Osborne DW, Chow DSL. 2008. Cre-
mophor-free intravenous microemulsions for pacli-
taxel I: Formulation, cytotoxicity and hemolysis.
Int J Pharm 349:108–116.
REFERENCES
15. Park KM, Kim CK. 1999. Preparation and evalua-
tion of flurbiprofen-loaded microemulsion for par-
enteral delivery. Int J Pharm 181:173–179.
16. Allen TM, Chonn A. 1987. Large unilamellar lipo-
somes with low uptake into the reticuloendothelial
system. FEBS Lett 223:42–46.
17. Mehta SK, Kaur G, Bhasin KK. 2008. Incorporation
of antitubercular drug isoniazid in pharmaceuti-
cally accepted microemulsion: Effect on microstruc-
ture and physical parameters. Pharm Res 25:226–
237.
18. Omotoshjo A, Whatelety L, Florenac ET. 1989.
Methotrexate transport from the internal phase
of multiple w/o/w emulsions. J Microencapsulation
6:183–192.
19. Garti N, Clement V, Fanun M, Leser ME. 2000.
Some characteristics of sugar ester nonionic micro-
emulsions in view of possible food applications.
J Agric Food Chem 48:3945–3956.
20. Garti N, Aserin A, Ezrahi S, Tiunova I, Bercovic G.
1996. Water behavior in nonionic surfactant sys-
tems I: Subzero temperature behaviour of water in
nonionic microemulsions studied by DSC. J Colloid
Interface Sci 178:60–68.
21. Ezrahi S, Aserin A, Garti N. 1998. Structural evo-
lution along water dilution lines in nonionic sys-
tems. J Colloid Interface Sci 203:222–224.
1. Blomberg B, Spinaci S, Fourie B, Laing R. 2001. The
rationale for recommending fixed dose combination
tablets for treatment of tuberculosis. Bull World
Health Organ 79:1.
2. Mariappan TT, Singh S. 2003. Regional gastroin-
testinal permeability of rifampicin and isoniazid
(alone and their combination) in the rat. Int J
Tuberc Lung Dis 7:797–803.
3. Panchagnula R, Agrawal S. 2004. Biopharmaceutic
and pharmacokinetic aspects of variable bioavail-
ability of rifampicin. Int J Pharm 271:1–4.
4. Shisoo CJ, Shah SA, Rahthod IS, Savale SS, Kote-
cha JS, Shah PB. 1999. Stability of rifampicin in
dissolution medium in presence of isoniazid. Int J
Pharm 190:109–123.
5. Agrawal S, Singh I, Kaur KJ, Bhade SR, Kaul CL,
Panchagnula R. 2004. Comparative bioavailability
of rifampicin, isoniazid and pyrazinamide from a
four drug fixed dose combination with separate
formulations at the same dose levels. Int J Pharm
276:41–49.
6. Bhutani H, Singh S, Jindal KC, Chakraborti AK.
2005. Mechanistic explanation to the catalysis by
pyrazinamide and ethambutol of reaction between
rifampicin and isoniazid in anti-TB FDCs. J Pharm
Bio Anal 39:892–899.
22. Garti N, Yaghmur A, Leser ME, Clement V, Watzke
HJ. 2001. Improved oil solubilization in oil/water
food grade microemulsions in the presence of poly-
ols and ethanol. J Agric Food Chem 49:2552–2562.
23. Israelachvili JN, Mitchell DJ, Ninham BW. 1976.
Theory of self-assembly of hydrocarbon amphi-
philes into micelles and bilayers. J Chem Soc Fara-
day Trans II 72:1525–1567.
24. Mehta SK, Kaur G, Bhasin KK. 2009. AAPS
PharmScitech (AAPSPT-D-08-00343R2).
25. Gursoy A, Senyucel B. 1997. Characterization of
ciprofloxacin liposomes: Derivative ultraviolet spec-
trophotometric determinations. J Microencapsulation
14:769–776.
7. Gohell MC, Sarvaiya KG. 2007. A novel solid dosage
form of rifampicin and isoniazid with improved
functionality. AAPS PharmSciTech 8: article 68.
8. Toit LC, Pillay V, Danckwerts MP. 2006. Tubercu-
losis chemotherapy: Current drug delivery
approaches. Resp Res 7:118–136.
9. Formariz TP, Chiavacci LA, Sarmento VHV, Fran-
zini CM, Silva-Jr AA, Scarpa MV, Santilli CV, Egito
EST, Oliveira AG. 2008. Structural changes of bio-
compatible neutral microemulsions stabilized by
mixed surfactant containing soya phosphatidylcho-
line and their relationship with doxorubicin release.
Colloids Surf B 63:287–295.
10. Chan J, Maghraby GME, Craig JP, Alany RG. 2007.
Phase transition water-in-oil microemulsions as
ocular drug delivery systems: In vitro and in vivo
evaluation. Int J Pharm 328:65–71.
26. Ojeda BC, Rojas FS. 2004. Recent developments in
derivative ultraviolet/visible absorption spectro-
photometry. Anal Chim Acta 518:1–24.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 4, APRIL 2010
DOI 10.1002/jps

ENTRAPMENT OF MULTIPLE ANTI-Tb DRUGS IN MICROEMULSION SYSTEM
1911
27. Salinas F, Berzas Nevado JJ, Epsinosa MA. 1990.
A new spectrophotometric method for quantitative
multicomponent analysis resolution of mixtures of
Salicylic and Salicyluric acids. Talanta 37:347–351.
28. Dinc E. 1999. The spectrophotometric multicompo-
nent analysis of a ternary mixture of ascorbic acid,
acetylsalicylic acid and paracetamol by the double
divisor-ratio spectra derivative and ratio spectra-
zero crossing methods. Talanta 48:1145–1157.
containing fixed dose combination formulations. Int
J Pharm 287:97–112.
31. Mehta SK, Kaur G, Bhasin KK. 2007. Analysis of
Tween based microemulsion in the presence of TB
drug rifampicin. Colloids Surf B 60:95–104.
32. Dai H, Chen Q, Qin H, Guan Y, Shen D, Hua Y,
Tang Y, Xu J. 2006. A temperature-responsive
copolymer hydrogel in controlled drug delivery.
Macromolecules 39:6584–6589.
29. Youssef RM, Maher HM. 2002. A new hybrid double
divisor ratio spectra method for the analysis of
33. Zhang L, Sun X, Zhang Z-R. 2005. An investigation
on liver-targeting microemulsions of norcanthari-
din. Drug Delivery 12:289–295.
ternary mixtures. Spectrochim Acta Part
A
70:1152–1166.
30. Agrawal S, Panchagnula R. 2004. Dissolution test
as a surrogate for quality evaluation of rifampicin
34. Bhutani H, Singh S, Jindal KC. 2005. Drug-drug
interaction studies on first-line anti-tuberculosis
drugs. Pharm Dev Tech 10:517–524.
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 4, APRIL 2010