Investigation on factors affecting drug delivery using polymers and phospholipids 5

104
CHAPTER 7
Effect of Polymer Transition Temperature on the Topical Delivery of
Levothyroxine
7.1 Introduction
Polymeric microspheres are being widely used in drug delivery systems.
Biodegradable polymers can be designed to control and prolong drug release by
adjusting the degradation rate of the polymer. Ethyl cellulose (EC), a non-toxic,
inexpensive and biodegradable polymer, has been used in a variety of applications in
pharmaceutical dosage forms such as sustained release and controlled delivery of
drugs (Kang et al., 2006; Duarte et al., 2006; Crowley et al., 2004). Due to its
hydrophilic structure it can help increase the bioavailability of poorly water-soluble
compounds by forming an inclusion (Duarte et al., 2006). Smart polymers can
release the drug by external stimuli such as change in temperature, pH or ionic
composition (Singh et al., 2007; Schmaljohann 2006). Thermosensitive polymers,
such as Poly-N-isopropylacrylamide (PNIPAM), exhibit a temperature-dependent
shrinking at temperatures below the low critical solution temperature (LCST). These
changes corresponding to swelling or de-swelling of the polymer can be controlled to
release the encapsulated drugs in response to external temperature changes (Dimitrov
et al., 2007; Makino et al., 2001; Gutowska et al., 1992; Zhang et al., 2002; Guo et al.,
2008).
105
PNIPAM microgels have been studied as transdermal carriers, however results did
not show penetration enhancement across human epidermis (Lopez et al., 2005).
PLGA microparticles were able to increase drug retention in the epidermis and
decrease the drug permeation through the skin (Ga de Jalón et al., 2001a and b;
Rolland et al., 1993; Tsujimoto et al., 2007). Several studies have shown sustained
and controlled release of drugs from transdermal patches which contained EC
(Mutalik and Udupa 2005; Mukherjee et al., 2005; Rama Rao et al., 2006; Rama Rao
2003; Mayorga et al., 1997 and 1996; Rama Rao and Diwan 1998; Amnuaikit et al.,
2005).

These polymers have been studied individually as transdermal carriers in previous
works. Variation on the effect of partition coefficient and skin permeability was
studied using PNIPAM microgels (Lopez et al., 2005). Formulation variation of
PLGA microparticles was also investigated (Santoyo et al., 2002; Tsujimoto et al.,
2007). Formulation strategies have also been studied using EC polymer as a
transdermal vehicle ((Mutalik and Udupa 2005; Mukherjee et al. 2005; Rama Rao et
al. 2006). Here we wanted to compare and see the effect of polymer hydrophobicity,
polymer transition temperature on the skin permeation.
The aim of this work is to determine if topical application of T
4
can produce systemic
effects. Four types of polymers of different molecular weights and different
hydrophobicities were used to encapsulate T
4
. The microspheres were characterized
for drug entrapment efficiency, storage stability, in vitro drug release and skin
penetration.
106
7.2 Materials and Methods
7.2.1 Materials
L-levothyroxine, poly vinyl alcohol (MW 31,000), poly (N-isopropylacrylamide)
(MW 20,000-25,000), ethyl cellulose and phosphate buffer saline tablets were
purchased from Sigma, Singapore. PLA (R 203H) and PLGA 48/52 (RG 503H) were
gifts from Boehringer Ingelheim (Germany). The density of PLA and PLGA were
0.34 dl/g and 0.52 dl/g respectively.
7.2.2 Preparation of Microparticles
Levothyroxine-loaded microparticles were prepared by emulsification-solvent
evaporation technique. The organic phase consisted of 50 mg of polymer dissolved in
1 ml of dicloromethane which was then emulsified in a PVA aqueous solution (5 ml,
5% w/v PVA, 2 mg/ml T

4
). The system was stirred continuously at 700 rpm for 5
hours to allow the evaporation of the organic solvent.
7.2.3 Encapsulation Efficacy and Stability Studies
The drug-loaded microparticles were centrifuged at 17 000 rpm for 45 min at 20
o
C.
The free levothyroxine in the supernatant was determined by HPLC method and the
encapsulation efficacy (EE %) was calculated using Eq. 2-3.
Encapsulation efficacy was investigated after 14-week storage at room temperature
(20
o
C) and in the fridge (4
o
C).
107
7.2.4 Microparticle Characterization
The surface morphology and appearance of microparticles were examined using
FESEM mentioned in section 4.2.3.
7.2.5 HPLC Analysis
T
4
concentration was analyzed using the method mentioned in section 6.2.6.
7.2.6 In vitro Drug Release Studies
In vitro drug release from the drug-loaded beads was studied in phosphate buffer
saline (PBS; pH 7.4) at 37
o
C in a horizontal shaker. At specific intervals, 1-ml
samples were taken and the microparticulate dispersions were centrifuged to remove
impurities before being assayed for drug content by HPLC method. An equal volume

of fresh PBS was immediately added to the receptor cell after each sampling.
7.2.7 In vitro Skin Permeation Studies
Permeation studies of drug-loaded microparticles were performed using a flow-
through diffusion cell apparatus (described in section 2.2.8.). The donor compartment
was filled with 1 ml of aqueous polymeric microparticle solution and the receptor
compartment was phosphate buffer saline pH 7.4. Samples from the receptor
compartment were collected at predetermined time points over a 24-h period, and the
amount of T
4
permeated was analyzed by HPLC.
108
7.2.8 FTIR of Skin Sample
FTIR spectra of the skin samples treated with polymeric particles were obtained with
Perkin Elmer Spectrum 100 (USA). After treating the epidermis with each
formulation for 24 h, the samples were washed 3 times with PBS and vacuum-dried at
room temperature. Samples were then subjected to FTIR measurements. Details are
mentioned in section 4.2.4.
7.2.9 Confocal Studies of the Treated Skin
To study the effect of polymeric microparticles on the extent of skin penetration,
confocal study was carried out. Skin samples were treated with polymeric
microparticles and then aquous solution of 0.03% w/v fluorescein dye was applied
and its skin penetration was viewed using a CLSM described in section 2.2.9.
7.3 Results and Discussion
7.3.1 FESEM Characterization of Microparticles
FESEM images of the microparticles are shown in Fig. 7.1. It can be seen that the
appearance of the microparticles clearly varied with the polymer type. Ethyl cellulose
microspheres had a uniform microporous and sponge-like structure. No considerable
difference was observed between the microstructures of PLA and PLGA
microparticles. Cracks in the surface of PLGA microparticles were probably artifacts
due to the high energy of the electron beam at high magnifications. PLGA has a low

glass transition temperature (T
g
), therefore the polymer transforms from a glassy to a
109
rubbery state which is more susceptible to the vacuum pressure of FESEM (Wischke
et al., 2006). PNIPAM microcapsules were fragmented but not deformed.
The rate of solvent evaporation, polymer precipitation and stability of the inner
aqueous phase play a major role in microcapsule morphology (Crotts and Park 1995).
Surface tension of the solution greatly affects the microparticle structure. Reduction
in surface tension of the solution will lead to fast and rapid solvent evaporation which
will result in fewer pores on the particle surface (Niwa et al., 1993). In our study,
parameters such as compositions of solvent system and aqueous phase were kept
constant, therefore any difference in the morphology or structure of the particles is
likely to be related to the intrinsic properties of the polymers.
110
(a) (b)
(c) (d)
Fig. 7.1 FESEM images of T
4
-loaded (a) PLA, (b) PLGA, (c) EC and (d) PNIPAM
microparticles.
7.3.2 Determination of Encapsulation Efficacy
The encapsulation efficiencies of T
4
-loaded microparticles consisting of different
polymers and their physical stability over a 14-week period are displayed in Fig. 7.2.
It was found that ethyl cellulose microparticles exhibited the highest drug
encapsulation of 85.98 ± 8.84%. PLA and PLGA resulted in similar encapsulation
efficacy of 75.54 ± 12.11% and 76.47 ± 17.88%, respectively. PNIPAM had the
lowest T

4
encapsulation of 67.59 ± 1.81%.
111
0
20
40
60
80
100
EC PLA PLGA PNIM
Encapsulation Efficiency %
Day 0
Week 2 (20ºC)
Week 2 (4ºC)
Week 4 (20ºC)
Week 4 (4ºC)
Week 14 (20ºC)
Week 14 (4ºC)
Fig. 7.2 Stability of lipid suspensions: Encapsulation efficacy of the vesicles over time at
room temperature (20
o
C) and fridge temperature (4
o
C), n=3.
The in vitro degradation behavior of polymeric microparticles was investigated at
20
o
C and 4
o
C (Fig. 7.2). It was found that irrespective of the storage temperature,

ethyl cellulose microparticles remained stable during the 14-week storage period
without significant drug leakage (p > 0.05). The degradation rate of PNIPAM
microparticles was faster than PLA and PLGA microparticles. PLGA microparticles
stored at 4
o
C did not show any significant drug loss over the study period (p > 0.05)
however, storage at 20
o
C resulted in significant drug leakage (p < 0.05). After 14
weeks at 20
o
C and 4
o
C, the T
4
contents of PLA and PNIPAM microparticles were
significantly lower than the original (p < 0.05).
7.3.3 In vitro Release of Levothyroxine
The release rate of T
4
from microspheres of EC, PLA, PLGA and PNIPAM are
shown in Fig. 7.3. The profiles show the influence of polymer type on the in vitro
release of T
4
. It was found that T
4
exhibited a burst release from all formulations
112
irrespective of polymer type. The drug release from microparticles seems to occur in
two phases: an initial rapid release followed by a slow release. The initial burst effect

is probably due to the adsorption of the drug onto the wall of the microparticles which
would be immediately released. After which, the drug release profile displayed a
delayed release that may be attributed to diffusion of the drug entrapped within the
core of the microparticles.
0
20
40
60
80
100
0 60 120 180 240 300 360 420 480 540
Time (min)
T
4
released (%)
EC
PLA
PLGA
PNIPAM
Fig. 7.3 In vitro release profile of T
4
from microparticles in phosphate buffer (pH 7.4) at the
body temperature (37
o
C), n = 3.
7.3.4 In vitro Skin Permeation Studies
In vitro skin permeation studies were performed to evaluate the skin absorption of T
4
from these preparations. Fig. 7.4 depicts the permeation profile of T
4

from the
polymeric particles. The systems with PLA, PLGA and PNIPAM did not provide any
T
4
penetration, however EC microparticles showed some drug penetration across the
113
skin. This work showed that the use of the T
4
-loaded PLA, PLGA and PNIPAM
microparticles increased drug retention in the epidermis and decreased drug
permeation through the skin. Consequently, these polymeric microparticles represent
a good delivery system to retard the release rate of drugs into the skin and improve
topical therapy.
0
30
60
90
120
150
0 4 8 12 16 20 24
Time (h)
Cumulative T
4
(μg/cm
2
)
Control
EC
PLGA, PLA, PNIPAM
Fig. 7.4 Permeation profile of T

4
across human epidermis (n=6-8).
Amorphous polymers exhibit glass transition temperature (T
g
). Below this
temperature, polymer is in a glass-like state. Above this temperature, the polymer
passes from a glassy to a rubber-like state which may cause coalescence and
precipitation of the polymer network on the surface (Wischke et al., 2006; Kangarlou
et al., 2008; Middleton and Tipton 2000). T
g
may be lowered by loading polymers
with other molecules. This could be critical with respect to storage stability and drug
release profile (Wischke et al., 2006). Water has a plasticizing effect on polyesters
and lowers the T
g
from > 40
o
C to around 37
o
C (Wischke et al., 2006; Middleton and
114
Tipton 2000; Packhaeuser and Kissel 2007; Passerini and Craig 2001; Henry et al.,
2005; Royall et al., 2001). As compared to polyesters, EC has a high T
g
of >100
o
C
(Tarvainen et al., 2003; Frohoff-Hülsmann et al., 1999; Hyppölä et al., 1996; Rowe et
al., 1984). PNIPAM is a synthetic thermosensitive polymer that responds to internal
or external stimuli. Aqueous PNIPAM solutions exhibit a LCST of 32

o
C. At
temperatures below the LCST, PNIPAM is hydrophilic and exists in a random coil
form; however, above the LCST, it becomes insoluble and precipitates out from the
aqueous solution. Furthermore, this precipitation could delay the drug release by
acting as an additional diffusion barrier (Choi et al., 2006; Geever et al., 2006).
It is possible to hypothesize that the low T
g
of PLA and PLGA, and low LCST of
PNIPAM may cause precipitation of the rubber-like, insoluble polymer on the skin
surface. This could create an impermeable barrier and prevent drug penetration
across the epidermis. T
4
skin penetration observed for EC microparticles may be due
to its high T
g
therefore polymer remains in the glassy state which is soluble and does
not precipitate.
Luengo and coworkers studied the effect of PLGA nanoparticles on the skin
permeation of flufenamic acid. At shorter incubation times there was no significant
differences in the permeated amount of drug, however after long incubation time due
to the degradation of the polymer to lactic and glycolic acid and the reduction of the
pH of the donor compartment, skin permeation was enhanced (Luengo et al., 2006).
The authors could not explain the non-enhancing effect of nanoparticles at shorter
115
incubation time, however this phenomenon might be due to the change in the physical
state of the polymer with regards to its low transition temperature.
An alternative view is that the presence of oxygen atoms in the polymer molecule
could facilitate the formation of hydrogen bonds with the skin lipids. This could have
stabilized the rigidity of the solid-lipid state of the skin structure by increasing the

skin T
g
and therefore retarding the skin penetration (Hadgraft et al., 1996; El
Maghraby et al., 2005; Asbill and Michnaik et al., 2000).
Confocal images of the skin samples treated with the polymeric microparticles are
shown in Fig. 7.5. It can be seen that the fluorescein dye easily penetrated through the
control skin samples. Skin samples treated with EC microparticles show some skin
penetration of flourescein. However it can be seen that the fluorescein dye could not
penetrate the skin samples treated with polyesters and PNIPAM microparticles and
the dye was mainly focused on the outer layer of the skin. This may be due to the
impermeable barrier of the polymer on the skin surface which prevents the
penetration of the fluorescein.
116
Control EC PNIPAM
PLA PLGA
Fig 7.5 Binary image of the epidermis and localization of green fluorescence on the skin after treatment with the polymeric
particles. For better visualization cell nuclei were counter stained with DAPI.
117
7.3.5 FTIR of Human Skin Samples
Fig. 7.6 presents polymer-induced changes in the skin structure monitored through
FTIR. Spectra of the skin samples were recorded at the end of the in vitro permeation
study. The CH
2
asymmetric and symmetric stretching vibrations were observed at
2920 and 2851 cm
-1
respectively. These peaks indicate that the majority of the
stratum corneum lipids are in the solid-gel state. The shift of these peaks to a higher
wavenumber after skin treatment would suggest increased lipid fluidity of the skin
structure. Peaks typical of SC proteins were those occurring in the region of 1500-

1700 cm
-1
. The carbonyl stretching observed at 1743 cm
-1
is due to C-O stretching.
Amide I band at 1650 cm
-1
is associated with α-helix conformation of the protein
backbone (Bernard et al., 2007; Tanojo et al., 1997; Goates and Knutson 1993, 1994).
After the exposure of the skin samples to microparticle solutions, no peak shift in the
CH stretching area and the protein domain was observed. These polymers did not
alter the lipid fluidity of the SC, and did not interact with the SC proteins.
118
Fig. 7.6 FTIR spectrum of (a) untreated human epidermis, and skin treated with (b) 10% w/v
PVA mat, (c) 10% w/v PNIPAM mat and (d) PBS.
7.4 Conclusion
Formulations containing polymeric microparticles suitable for topical and transdermal
delivery systems were studied using four different polymers, poly D,L lactide (PLA),
poly D,L lactide co glycoside (PLGA), poly (N-isopropylacrylamide) (PNIPAM) and
ethyl cellulose (EC). It was found that ethyl cellulose microparticles had the highest
drug encapsulation and minimal drug leakage during the 14-week storage period.
PNIPAM microparticles had the lowest drug encapsulation efficiency and a fast
degradation rate. PLGA microparticles exhibited a temperature dependent drug
leakage. It was observed that transition temperature (T
g
) may influence the skin
permeation rate of the drug from these microparticles. Polyesters (PLA and PLGA)
and PNIPAM acted as skin penetration retardant. These microparticles have potential
119
use in skin formulations containing sunscreens and other active ingredients that are

meant to be concentrated on the skin surface. However skin permeation was observed
from EC microparticles, therefore such polymers may be used as carriers in
transdermal formulations and can help achieve therapeutic concentrations of the drug
in the plasma
120
CHAPTER 8
Effect of Skin Lipid Fluidization and Drug Encapsulation on the
Transdermal Delivery of Diclofenac
8.1 Introduction
Diclofenac sodium is a widely used non-steroid-type anti-inflammatory agent. Its
administration is associated with adverse gastro-intestinal effects. It is extensively
metabolized in the liver and has a short biological half-life. These challenges have
been overcome via topical administration (Boinpally et al., 2003; Sintove and Botner
2006; Escribanoa et al., 2003).
Vesicular lipid systems are used as carriers for topical and transdermal drug delivery.
However, it is generally agreed that conventional liposomes have little or no effect on
the penetration of drugs through the skin and are chemically and physically unstable
(Desai and Finlay 2002; Hashizume et al., 2003). Niosome vesicles from nonionic
surfactants, were thought to be an improvement over the conventional liposomes
(Hofland et al., 1991; Hao et al., 2002). Alternatively, polyethyleneglycol (PEG)
containing niosomes (Liu et al., 2007; Hua and Liu et al., 2007) and other
formulations such as proniosomes, containing cholesterol and non-ionic surfactants,
were developed (Alsarra et al., 2005; Fang et al., 2001; Hu and Rhodes 1999).
Ethanol, a skin permeation enhancer, was incorporated into liposomes and termed
ethosomes. These vesicles have the ability to permeate through the human skin and
effect intracellular delivery (Touitou et al., 2000 and 2001; Paolinoa et al., 2005).
121
Transferosomes, another form of lipid carriers, are regarded as deformable liposomes.
These ultra-deformable carriers contain an edge activator and have high elasticity
which enables them to squeeze through intercellular regions of stratum corneum (SC)

(Cevc et al., 1998; Mishra et al., 2007; Elsayed et al., 2006). Recently, liposomes
with similar composition to that of the SC, cerosomes, have been formulated and used
to enhance the skin delivery of drugs (Hatziantoniou et al., 2007; Contreras et al.,
2005).
The aim of this work was to identify the most effective formulation for delivering a
hydrophilic drug in terms of drug encapsulation efficiency, stability and skin
permeation properties. The vesicles of conventional liposomes, ethosomes,
transferosomes, proniosomes, niosomes and polyethyleneglycol-block-
polypropyleneglycol-block-polyethyleneglycol (PEG-PPG-PEG) niosomes were
formulated and compared to SC liposomes (cerosomes) for their ability to increase
skin permeation of diclofenac.
Although the lipid vesicles have different names but actually they only vary in some
of the ingredients. The effect of various compositions of the lipid vesicles on the skin
permeation of diclofenac was studied. Alterations of the biophysical structure of the
SC in the presence of these vesicles were determined using fourier transform infrared
spectroscopy.
122
8.2 Materials and Methods
8.2.1 Materials
Lipoid E 80 (Phosphatidylcholine from egg yolk lecithin) and Cerosome 9005 were
gifts from Lipoid GmbH (Ludwigshafen, Germany). Diclofenac sodium, cholesterol,
sodium phosphate monobasic monohydrate, PEG-PPG-PEG, Span 85 and Tween 20
were purchased from Sigma, Singapore. Tween 80 was purchased from Bio-Rad
Laboratories (Singapore).
8.2.2 Preparation of Diclofenac Sodium-Loaded Vesicles
The compositions of different vesicle formulations are listed in Table 8.1.
Diclofenac-loaded conventional liposomes were prepared by cast film method as
reported previously with slight modification (Dubey et al., 2007). Briefly, Lipoid E 80
was dissolved in ethanol in a clean and dry round bottom flask followed by removal
of the organic solvents using rotary vacuum evaporator above the lipid transition

temperature to form a thin film on the wall of the flask. After removal of solvent
traces, thin lipid film was hydrated with phosphate buffer saline (PBS) pH 7.4
containing diclofenac by magnetic stirring (1000 rpm, 10 min) at the corresponding
temperature.
Ethosome colloidal suspensions, PEG-PPG-PEG niosomes, transferosomes, niosomes
were prepared as reported elsewhere (Paolinoa et al., 2005; Liu et al., 2007; Elsayed
et al., 2006; Devaraj et al., 2002). Ingredients were solubilized in absolute ethanol.
123
PBS containing diclofenac sodium was added gradually while mixing at 1000 rpm
with a magnetic stirrer.
Proniosomes were prepared as described previously with slight modification (Alsarra
et al., 2005). Briefly, Tween 20, Lipoid E 80, and cholesterol (9:9:2) were mixed
with absolute ethanol and warmed in a water-bath sonicator at 65
o
C for 5 min. Then
PBS containing diclofenac sodium was added and the mixture was further warmed in
the water bath for about 2 min until a clear solution was obtained. The mixture was
allowed to cool at room temperature to form the proniosomal suspension.
All formulations were finely homogenized for 1 min at amplitude 30, and at pulser of
2 by means of an ultrasonic processor (ITS Science).
Cerosomes (containing 6.6% v/v SC lipids) were used as obtained without further
modification. Diclofenac sodium was physically blended with this readily made
cream and stirred to make a homogenous formulation, therefore vesicle size and
encapsulation efficacy was not calculated for this formulation. A PBS solution of
diclofenac was used as control. All formulations contained a total of 5 mg/ml
diclofenac sodium.
124
Table 8.1 Composition of lipid suspensions.
.
Composition (% w/v)

Formulation
PBS Ethanol Lecithin Cholesterol PEG-PPG-PEG Tween 20 Tween 80 Span 85
Conventional liposome 65 30 5 - - - - -
Ethosome 65 30 5 - - - - -
Proniosome 20.3 31.6 22.8 2.5 - 22.8 - -
Niosome 63 30 5 1 - 1 - -
Transferosome 64 30 5 - - - 1 -
PEG-PPG-PEG niosome 63 30 - - 5 1 - 1
125
8.2.3 Determination of Encapsulation Efficiency
The unentrapped diclofenac was removed following centrifugation at 17 000 rpm for
45 min at 20
o
C and analysed by HPLC method. The encapsulation efficiency (EE %)
was calculated using Eq. 2.3.
8.2.4 Storage Stability of Vesicles
Storage stability studies are important in the development of pharmaceutically
acceptable product. The ability of vesicles to retain the drug was assessed by keeping
the formulation suspensions at 4 ± 2
o
C (fridge) and 20 ± 2
o
C (room temperature) for a
period of 60 days. Drug leakage was observed by measuring encapsulation efficacy
of the vesicular formulations. The diameters of the particles were measured on days
30 and 60 at two different temperatures. The vesicular suspensions were kept in
sealed vials. Mean vesicle sizes of drug-loaded liposomes were determined using
Zetasizer 300HSA (Malvern Instruments, Malvern, UK). Analysis (n = 3) was
carried out at room temperature and an angle of detection of 90
o

.
8.2.5 In vitro Skin Permeation Studies
The permeation profiles of drug from drug-loaded ethosomes, niosomes,
proniosomes, PEG-PPG-PEG niosomes, conventional liposomes, cerosomes and
control samples through the skin were determined by using a flow-through diffusion
cell as described previously in section 2.2.8. A 1-ml formulation containing 5 mg/ml
diclofenac sodium was applied on the skin in the donor compartment. The receptor
126
medium was PBS. Samples were withdrawn at specific time intervals for a period of
48 h. Diclofenac was quantified using HPLC method.
8.2.6 FTIR Studies of the Human Skin
After treating the epidermis with each formulation for 48 h, the samples were washed
3 times with PBS and vacuum-dried at room temperature. The samples were then
subjected to FTIR spectroscopy mentioned in section 4.2.4.
8.2.7 HPLC Assay of Diclofenac Sodium
The quantitative determination of diclofenac was performed by HPLC using
acetonitrile/pH 3 buffer solutions (35:65 v/v) delivered at a flow rate of 1 ml/ min. A
sample of 20 μl was eluted from the Agilent hypersil column, C18, 5 µm, 4.6 × 250
mm. Drug peaks, detected at wavelength 290 nm, were separated at 5.2 min.
8.3 Results and Discussion
8.3.1 Vesicle Size Measurment
Fig. 8.1 presents the sizes of the vesicles over an 8-week storage period. The smallest
vesicles were transferosomes with a mean vesicle size of 118.9 ± 9.8 nm.
Proniosomes had the largest vesicles of 710.8 ± 135.3 nm. This may be due to the
high concentrations of cholesterol and lecithin incorporated in the vesicles.
The size of the PEG-PPG-PEG niosomes increased upon 8-week storage compared to
1-h after preparation. The highest flocculation rate was observed for formulations
127
stored at room temperature. Aggregation was not found to be temperature-dependent
for the rest of the formulations as it was similarly observed during storage at room

temperature and in the fridge. Lack of net electrical charge of conventional
liposomes could have caused the aggregation of the vesicles. This effect might have
resulted from storage of the liposomal formulation in PBS or other aqueous phase
containing polyvalent ion (Grit and Crommelin 1993; Fransen et al., 1986; Lau et al.,
2005). However, vesicles of ethosomes, transferosomes, niosomes and proniosomes
decreases in size after 8 weeks as compared to 1-h after preparation.
0
400
800
1200
1600
2000
Ethosome Transferosome Niosome PEG-PPG-PEG
niosome
Proniosome Conventional
Liposome
Size (nm)
day1
4 weeks (20ºC)
4 weeks (4ºC)
8 weeks (20ºC)
8 weeks (4ºC)
Fig. 8.1 Stability of lipid formulations: Mean diameter of vesicle formulations (nm) with
time, (n=3).
128
8.3.2 Determination of Encapsulation Efficiency
Fig. 8.2 represents the encapsulation efficiency of the vesicles. There was no
significant difference in the amount of drug encapsulated in all formulations (p >
0.05). Proniosomes and conventional liposomes entrapped the most drug of 63.43 ±
9.31% and 67.15 ± 5.84%, respectively (p> 0.05). The preparation methods required

these vesicles to be hydrated at above 60
o
C while the rest of the formulations were
hydrated at room temperature. The hydration temperature was thought to influence
the extent of drug encapsulation, therefore less drug was entrapped at temperatures
below the lipid transition point (Hao et al., 2002). This could explain the low drug
encapsulation observed for other vesicle. Lipid vesicles were thought to effectively
entrap both hydrophobic and hydrophilic drugs. However, studies have also
demonstrated that the entrapment of hydrophilic molecules could be less efficient
than that of hydrophobic molecules (Touitou et al., 2001; López-Pinto et al., 2005).
The latter supports our findings as relatively small amounts of diclofenac sodium
were encapsulated in the vesicles when compared to those of hydrophobic drugs in
the literature. Encapsulation efficiencies of the drug in our formulated systems after
12-week storage at 20
o
C and 4
o
C showed no difference from that of a freshly
prepared sample. These formulations were relatively stable with minimal drug
leakage (p > 0.05).
8.3.3 In vitro Skin Permeation Studies
The effects of 6 vesicular formulations on the in vitro percutaneous permeation of
diclofenac sodium through human epidermis are shown in Fig. 8.3.