Drug Delivery

ISSN: 1071-7544 (Print) 1521-0464 (Online) Journal homepage: https://www.tandfonline.com/loi/idrd20

Bilayered transmucosal drug delivery system of

pravastatin sodium: Statistical optimization, in
vitro, ex vivo, in vivo and stability assessment

Suresh K. Maurya, Vikas Bali & Kamla Pathak

To cite this article: Suresh K. Maurya, Vikas Bali & Kamla Pathak (2012) Bilayered transmucosal
drug delivery system of pravastatin sodium: Statistical optimization, in�vitro,�ex�vivo,�in�vivo and
stability assessment, Drug Delivery, 19:1, 45-57, DOI: 10.3109/10717544.2011.644348

To link to this article: https://doi.org/10.3109/10717544.2011.644348

Published online: 23 Dec 2011.

Submit your article to this journal

Article views: 340

Citing articles: 7 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=idrd20
Drug Delivery, 2012; 19(1): 45–57
© 2012 Informa Healthcare USA, Inc.
ISSN 1071-7544 print/ISSN 1521-0464 online
DOI: 10.3109/10717544.2011.644348

ORIGINAL article

Bilayered transmucosal drug delivery system of pravastatin

sodium: Statistical optimization, in vitro, ex vivo, in vivo and
stability assessment
Suresh K. Maurya, Vikas Bali, and Kamla Pathak

Department of Pharmaceutics, Rajiv Academy for Pharmacy, Mathura, Uttar Pradesh, India

Abstract
The objective of the present study was to develop a mucoadhesive sustained release bilayered buccal patch of
pravastatin sodium using Eudragit S100 as the base matrix so as to surmount hepatic first pass metabolism and
gastric instability of the drug. A 32 full factorial design was employed to study the effect of independent variables
viz. levels of HPMC K4M and carbopol 934P on % cumulative drug release, mucoadhesion time and mucoadhesive
force. Amount of carbopol 934P and HPMC K4M significantly influenced characteristics like swelling index, in vitro
mucoadhesive force, drug release, and mucoadhesion time. In vitro evaluation revealed that formulations exhibited
satisfactory technological parameters. The mechanism of drug release was found to be non-Fickian diffusion.
Different permeation enhancers were investigated to improve the permeation of drug from the optimized patches
(F9) across the buccal mucosa. Formulation [F9 (P3)] containing 4% (v/v) dimethyl sulfoxide exhibited desirable
permeation of drug. Histopathological studies performed using goat buccal mucosa revealed no mucosal damage.
Bioavailability studies in rabbits demonstrated that [F9 (P3)] significantly higher Cmax (67.34 ± 3.58 ng/ml) and AUC0-∞
(350.27 ± 9.59 ng/ml×h) (p < 0.05) of pravastatin sodium from optimized patch than IR tablet (Cmax 58.73 ± 4.63 ng/ml
and AUC0-∞ 133.80 ± 8.25 ng/ml×h). Formulation [F9 (P3)] showed sustained drug plasma concentration over a period
of 10 h which was significantly longer than oral tablet (p < 0.05). Stability studies as per ICH guidelines established
physical stability of the patch and chemical stability drug. The present study established potential of the optimized
mucoadhesive buccal patches to circumvent the hepatic first-pass metabolism, gastric instability and to improve
bioavailability of pravastatin sodium.
Keywords:  Mucoadhesive, buccal, pravastatin sodium, patch

1  Introduction of the dosage form from the buccal cavity (Miller et al.,
The concept of mucoadhesive drug delivery system was 2005). Furthermore, larger surface area for drug applica-
introduced into the arena of controlled drug delivery dur- tion and absorption along with good accessibility of the
ing 1980 and since then the concept has been utilized by buccal mucosa over nasal, rectal and vaginal mucosa
many investigators to overcome physiological barriers in also make it an attractive route for systemic drug delivery
the sustained drug delivery (Maurya et al., 2010). Buccal (Rathbone et  al., 1994). Buccal tablets and patches are
route has garnered considerable importance as an alter- the most preferred formulations amongst various buc-
native to oral route amongst the various mucosal routes coadhesive drug delivery systems since most of the other
owing to its several appealing features like: bypassing the systems get easily washed away by the continuous sali-
hepatic first-pass metabolism, precluding instability in vary secretion. Buccoadhesive patches are preferred due
the gastrointestinal milieu, and enhanced patient com- to ease of adherence to buccal mucosa, longer retention
pliance. Moreover, drug delivery can be promptly and period, highly flexibility and ease of removal at any time
conveniently terminated in case of any toxicity by removal during the treatment (Mumtaz and Ch’ng, 1995).

Address for Correspondence:  Vikas Bali, Department of Pharmaceutics, Rajiv Academy for Pharmacy, Mathura -281001, Uttar Pradesh,
Tel: (M): +91-9999964936, Fax: 0565-2530265. E-mail: vksbali@gmail.com
(Received 30 March 2011; revised 29 June 2011; accepted 07 November 2011)

45
46  S.K. Maurya et al.
Pravastatin sodium is reported to be the most effec- A total of experimental nine formulations of (F1 to F9)
tive and best tolerated agent for treating atherosclerotic were prepared as per the design as shown in Table 2.
vascular disease (Shidhaye et al., 2010). It is bitter in taste Mucoadhesion time, mucoadhesive force, and % cumula-
and undergoes extensive hepatic first pass metabolism tive drug release (% CDR) at 10th h were taken as response
(up to 70%) leading to an oral bioavailability of only 17%. parameters.
Moreover, the drug is also reported also to be unstable at
acidic pH (Boyong et al., 2003). 2.2.2  Statistical analysis of data
Hence, the present research attempted to enhance The effect of independent variables on the responses
oral bioavailability of pravastatin sodium by statistically was modeled with the help of Design Expert software
optimizing its sustained release bilayered mucoadhesive version 8.0.2 using following second order polynomial
buccal patches so as to circumvent its hepatic first pass equation:
metabolism and gastric instability. Moreover, bilayered
buccal patches would also help to overcome bitter taste Y = b0 + b1 X 1 + b2 X 2 + b11 X 12 + b22 X 22 + b12 X 1 X 2 (1)
of the drug without involvement of any elaborate taste  Where Y is the measured response, Xi is the level of the ith
masking procedure. It was anticipated that an increased factor, b0 represent the intercept, and bi, bij ….represent
bioavailability of the drug would help to reduce its dose coefficients computed from the responses of the formu-
and dose associated side effects. Literature review shows lations in the design.
that bilayered buccal patches of pravastatin sodium
have not yet reported. Moreover, the studies performed 2.2.3  Preparation of mucoadhesive bilayered buccal patches
until now on the buccal delivery system (tablet) of the 2.2.3.1. Preparation of backing layer  For preparation of
drug are confined to the ex vivo assessment of the devel- backing layer, a glass petriplate of 9 cm diameter was
oped system. The present study, in addition, describes used as the casting surface. The backing membrane of
bioavailability study in New Zealand white rabbits to ethyl cellulose was fabricated by slowly pouring a solu-
specifically delineate the plasma concentration time tion containing 3% (w/v) of ethyl cellulose and 2% (v/v) of
profile of the drug after buccal administration of devel- dibutyl phthalate as the plasticizer in 5 ml ethanol (95%,
oped formulation. The current work also details stabil- v/v) on to the glass petriplate and air drying for 1 h.
ity studies on the optimized patch and an effort has
also been made to calculate shelf life of the optimized 2.2.3.2 Preparation of mucoadhesive layer containing drug 
formulation. 15% (w/v) dispersion of Eudragit S100 was prepared in
ethanol (95%, v/v). 920 mg of drug was added to 5 ml of
ethanol (95 %, v/v) containing 15% (w/v) dispersion of
2  Materials and methods Eudragit S100 with continuous stirring. Ethanolic disper-
2.1  Materials sion containing carbopol 934P and HPMC K4M was pre-
Pravastatin sodium was received as a gift sample from pared separately. Both the dispersions were then mixed
Cipla Pharmaceuticals, Mumbai, India. Eudragit S 100 and 5% (v/v) of propylene glycol was added as the plasti-
was gifted by Degussa India Pvt. Ltd., Mumbai, India. cizer. The resultant dispersion was then poured on to the
Carbopol 934P and HPMC K4M were gifted by Central
Drug House Pvt. Ltd., New Delhi, India. Ethyl cellulose Table 2.  Composition of formulations
and propylene glycol were procured from S.D. Fine Drug / Eudragit Carbopol HPMC Propylene
Chemicals Ltd., Mumbai, India. New Zealand white patch S100 934P K4M glycol
rabbits used for animal studies were procured from the Code (mg) (%, w/v) (%, w/v) (%, w/v) (%, v/v)
animal house of Rajiv Academy for Pharmacy, Mathura, F1 14.46 15 0.5 2 5
India. F2 14.46 15 0.5 3 5
F3 14.46 15 0.5 4 5
2.2  Methods F4 14.46 15 1.0 2 5
2.2.1  Experimental design F5 14.46 15 1.0 3 5
A 32 full-factorial experimental design was used for F6 14.46 15 1.0 4 5
statistical optimization of buccoadhesive patches. The F7 14.46 15 1.5 2 5
amount of carbopol 934P (X1) and HPMC K4M (X2) F8 14.46 15 1.5 3 5
were selected as the independent variables. Each factor F9 14.46 15 1.5 4 5
was set at a high-, medium-, and low-level. The inde- *F10 14.46 15 1.25 3.5 5
pendent and dependent variables are given in Table 1. *Extra design check point formulation

Table 1.  Experimental design: Independent and dependent variables

Levels used
Factors (Independent variables) Low (–1) Medium (0) High (+1) Responses (Dependent variables)
Amount of Carbopol 934P (X1) 0.5% (w/v) 1.0 % (w/v) 1.5% (w/v) Mucoadhesion time (Y1) Mucoadhesive
Amount of HPMC (X2) 2% (w/v) 3% (w/v) 4% (w/v) force (Y2) %CDR at 10th h (Y3)

 Drug Delivery
 Bilayered transmucosal drug delivery system of pravastatin sodium  47
preformed backing layer and allowed to dry undisturbed 2.2.7  Determination of drug content
at room temperature till a flexible film was obtained. Drug content was determined by homogenizing the
Dried films were carefully removed, checked for any patches in 100 ml of phosphate buffer (pH 6.8) and, filter-
imperfections or air bubbles and cut into patches of 1 cm2 ing through 0.45μ filter. The resultant solution was ana-
in size. The patches were then packed in aluminum foil lyzed using HPLC method as described in section 2.2.17.
and stored in an airtight glass container to maintain the The experiment was carried out in triplicate and average
integrity and elasticity of the patches. value was reported.

2.2.4  Determination of weight variation, thickness, and 2.2.8  Determination of ex vivo mucoadhesion time
folding endurance The ex vivo mucoadhesion time was determined by
Twenty patches from a batch were selected randomly for application of the patch onto freshly excised goat buccal
determination of weight variation and thickness of patch mucosa which was fixed in the inner side of the beaker
was measured at ten different randomly selected spots at a distance of 3.0 cm from the bottom using cyano-
using a micrometer screw gauge (Mitutoyo Corporation, acrylate glue. Goat mucosa was used for the study since
Japan). For determination of folding endurance, patches it has been documented in literature that goat mucosa
were repeatedly folded at the same place till it broke or is well suited to study the permeation of drugs which
was folded up to 250 times if not broken. Folding endur- are actively transported (like pravastatin sodium).
ance was considered as the number of times a patch Moreover, it also offers merits of being economical,
could be folded at the same place without breaking. easily available and ethically substitutive (Tembhurne
and Sakarkar, 2008). Mucoadhesive side of each patch
2.2.5  Determination of surface pH was pasted onto the goat buccal mucosa by applying
To rule out the possibility of any irritation during appli- a light force with a fingertip for 25s after wetting the
cation of the patch, the micro environmental pH of the patch with a drop of phosphate buffer (pH 6.8). The
patch was determined by adding 2 ml of distilled water beaker was then filled with 500 ml of phosphate buffer
to its surface and keeping at room temperature for 3 h (pH 6.8) and kept at 37 ± 1°C. A stirring rate of 50 rpm
in glass tubes. The surface pH was noted by bringing was applied to mimic the environment of buccal cav-
the electrode of pH meter (HICON, Grover Enterprises, ity. Patch adhesion was monitored up to 12 h. The
Delhi, India) in contact with the surface of the patch, and time taken by the patch to detach from the goat buccal
allowing it to equilibrate for 2 min. mucosa was recorded as the mucoadhesion time.

2.2.6  Determination of swelling index 2.2.9  Determination of ex vivo mucoadhesive strength

For determination of swelling index, initial weight Assembly for determination of mucoadhesive strength of
(W1) of the patches was determined. The patches were patches is shown in Figure 1. The goat mucosa was fixed
then placed in petridishes containing 5 ml of phos- on to glass support. The backing layer side of the mucoad-
phate buffer (pH 6.8). The patches were removed from hesive patch was fixed on to lower surface of one of pan
the petridishes at 1, 2, 4, 8, and 10 h and excess buffer using cyanoacrylate glue. To facilitate initial hydration and
was removed using filter paper. Weight of the swollen swelling, mucoadhesive surface of the patch was moist-
patches (W2) was then determined and swelling index ened with a drop of phosphate buffer (pH 6.8). Adhesion
(SI) was calculated using the formula given below. The between the patch and goat buccal mucosa was facilitated
experiment was carried out in triplicate and average by placing a preload of 50 gm over the pan for 10 min.
value was reported. After 10 min. the preload was removed from the pan
which was followed by addition of water from the burette
Swelling index = (W2 − W1 ) / W1 × 100  (2) into the beaker at a constant rate of 100 drops per minute

Figure 1.  Setup for determination of mucoadhesive strength.

© 2012 Informa Healthcare USA, Inc.

48  S.K. Maurya et al.
until the patch detached from the mucosa. The weight of 2.2.14  In vitro drug permeation
water required to detach the patch from buccal mucosa Drug diffusion studies were carried out to ascertain the
was noted as the mucoadhesive strength. The experiment permeability of drug across the goat buccal mucosal
was performed in triplicate. Mucoadhesive force was then membrane (obtained from local slaughterhouse and
calculated according to the following formula: used within 2 h of slaughter), by using Franz diffusion
cell. The tissue was stored in phosphate buffered saline
 [Mucoadhesive strength (g) × 9.8] (3)
Force of adhesion (N) = (PBS) pH 7.4 solution upon collection. The epithelium
10000 was separated from the underlying connective tissues
with surgical scissors and clamped in between the
2.2.10  In vitro drug release donor and the receiver chambers of the diffusion cell
The drug release from patch was studied under sink con- for permeation studies. The receptor compartment
ditions using USP apparatus type V (HICON, New Delhi, contained 11 ml of phosphate buffer (pH 7.4) while the
India). Five hundred milliliter of phosphate buffer (pH donor compartment was filled with 3 ml of phosphate
6.8), maintained as 37 ± 0.5°C, was used as the dissolution buffer (pH 6.8). The patch was placed on the mucosal
media. The dissolution media was stirred at 50 rpm. The surface in the donor compartment and 1ml aliquots
patch was applied on to the disc in such a way that the were removed at 1, 2, 3, 4, 5, 6, 7, 8, 9 and 10h from the
mucoadhesive layer of the patch was in contact with the receptor compartment while the solution was being
dissolution media and non-adhesive backing layer was stirred continuously using magnetic stirrer. 1 ml of
fixed on to disc with the help of a cyanoacrylate glue. Five fresh medium was replaced to maintained sink con-
milliliter samples were withdrawn at pre-determined dition each time. The experiment was carried out at
time intervals (1, 2, 3, 4, 5, 6, 7, 8, 9 and 10 h) and replaced 37 ± 0.5°C. The amount of drug permeated was assayed
with 5ml fresh dissolution medium. The samples were fil- using HPLC method as described in section 2.2.17. The
tered through a 0.45 μm filter and analyzed using HPLC graph of % drug permeated v/s time was plotted, flux,
method as described in section 2.2.17. permeability coefficient and enhancement ratio were
determined.
2.2.11  Drug release from backing layer
For determination of drug release from backing layer,
2.2.15  Histopathological evaluation of buccal mucosa
the patch was placed between the donor and the recep-
Comparative histopathological evaluation of tissue
tor compartment of a Franz diffusion cell maintained at
treated with treated with buccal patch (test) containing
37 ± 0.5°C with the backing layer facing the donor com-
4% v/v DMSO for 10 h was performed with respect to tis-
partment. The donor compartment contained 3 ml of
sue incubated in phosphate buffer saline (pH 6.8) for 10 h
phosphate buffer (pH 6.8) and the receptor compartment
(control). The tissue was fixed with 10% formalin, rou-
contained 11 ml of phosphate buffer (pH 7.4). At time
tinely processed and embedded in paraffin wax. Paraffin
interval of 1, 2, 3, 4, 5, 6, 7, 8, 9 and 10 h, 2 ml of sample
sections were cut by microtome and stained with hema-
was withdrawn from the donor compartment and ana-
toxylin and eosin. A pathologist, blinded to the study,
lyzed using HPLC method as described in section 2.2.17.
detected any damage to the test tissue at Department
Two milliliter of phosphate buffer (pH 6.8) was added
of Pathology, Pt. Deen Dayal Upadhaya Pashu Chikitsa
after each sampling to maintain sink condition.
Vigyan vishwavidyalya & Gau Research Centre, Mathura,
2.2.12  Validation of experimental design India by examining sections under a light microscope at
Polynomial equations were generated using Design expert 10X magnification.
software version 8.0.2 (Stat-Ease, Inc, USA) for selected
responses like mucoadhesion time, mucoadhesive force 2.2.16  Bioavailability studies
and % CDR at 10th h. The generated polynomial equations New Zealand white rabbits having weight between 3
were further reduced on the basis of significant terms to 3.5 kg were used for the study after an acclimatiza-
obtained by applying ANOVA. The 32 full factorial design tion period of one week. All animal experiments were
was validated by preparing an extra check point formula- performed according to the guidelines of Institutional
tion (F10). The predicted values of mucoadhesion time, Animal Ethical Committee (Registration No. 882-
mucoadhesive force and % CDR at 10th h for F10 were AC/05/CPCSEA) of Rajiv Academy for Pharmacy,
determined on the basis of respective polynomial equa- Mathura, India. Bioavailability study was performed
tions whereas the experimental values were determined on optimized buccoadhesive patch and in-house pre-
by evaluating F10 for the selected dependent variables. The pared immediate release tablet. A dose equivalent to
predicted and experimental values of the responses were 0.674 mg of pravastatin sodium per kg of body weight
compared for statistical significance using paired t-test. of rabbit was given in each case. IR tablets were pre-
pared in-house by wet granulation method so as to
2.2.13  Selection of optimized formulation administer the dose of the drug depending upon the
Optimized formulation was selected on the basis of maxi- body weight of the animal. Animals were divided into
mum mucoadhesion time, maximum mucoadhesive force, two groups with three animals in each group. Animals
maximum % CDR at 10th h and with good desirability. were lightly anesthetized using an i.m. injection of a 1:5

 Drug Delivery
 Bilayered transmucosal drug delivery system of pravastatin sodium  49
mixture of xylazine (1.9 mg/kg) and ketamine (9.3 mg/ aliquot was injected into the HPLC system. Recovery
kg). In one group, in-house prepared IR tablet con- (mean±S.D., n = 3) of pravastatin sodium from pooled
taining the required dose based on the body weight of rabbit plasma was found to be 72.44 ± 2.34 %.
the rabbit was administered. In another group, buccal
patch containing the required dose based on the body 2.2.18  Pharmacokinetics analysis of data
weight of rabbit was adhered onto buccal mucosa. At Pharmacokinetics analysis of the plasma concen-
an interval of 1 h, 1 ml of blood was withdrawn upto tration versus time profile of pravastatin sodium
10 h from marginal ear vein using a 26 gauge needle. was performed to calculate various parameters
Withdrawn blood samples were centrifuged at 8000 rpm like maximum plasma concentration (C max), time
for 10 min. at 15°C (REMI Pvt. Ltd., Vasai, India) to to reach maximum concentration (T max), and area
separate plasma. Plasma samples were stored at -20°C under the plasma concentration–time curve (AUC 0-t
until further analysis. and AUC 0-∞). The pharmacokinetic parameters were
calculated by using one compartment open model
2.2.17  HPLC analysis of pravastatin sodium with the help of Quick calc. software (developed by
Concentration of pravastatin sodium in the samples Dr. Shivprakash, Plexus, Ahmedabad, India). The
was determined using HPLC method (Otter and values of C max and T max were read directly from the
Mignat, 1998) with slight modification as per system arithmetic plot of time and plasma concentration of
requirement. The system used was Shimadzu LC-10AT pravastatin sodium. The relative bioavailability of
VP having a UV detector (Shimadzu, Kyoto, Japan) pravastatin sodium after buccal administration vs
and the software used was Class VP, version 5.032. A the oral administration was calculated by using the
C18 column (5 µm, 250 × 4.6 mm I.D., Phenomenex, following formula:
Aschaffenburg, Germany) was used as the station-
ary phase and 20 mM sodium dihydrogen phosphate F% = [AUCbuccal × Doseoral × 100] / [AUCoral × Dose buccal ] (4)
solution containing 1 mM dodecyl sulfate: acetonitrile  The pharmacokinetic data was analysed for statistical
(65: 35, v/v) was used as the mobile phase. The pH of significance using one way ANOVA, wherein p values
the mobile phase was adjusted to 2.0 with phosphoric <0.05 were considered to be statistically significant.
acid. The flow rate was maintained at 1ml/ min. and
detection was carried out at 238 nm. The retention 2.2.19  In vitro In vivo correlation (IVIVC)
time for pravastatin sodium was 7.2 min. The method To establish IVIVC for optimized patch, in vitro drug
was found to be linear in the concentration range of release curve of the patch was compared to its in vivo
10–100 ng/ml of pravastatin sodium. Coefficient of input performance i.e., the curve produced by suitable
correlation was found to be 0.9999. The least-squares deconvolution of the plasma level data using mass bal-
regression analysis provided the following equation: ance model-independent technique.
Y= 1860.99X+594.30 for pravastatin sodium. % RSD for
the precision study was found to be less than 2% indi- 2.2.20  Stability in simulated saliva
cating precision of the method. Percent recovery in The stability of developed patch was tested in simu-
the range of 98–102% indicated accuracy of the HPLC lated saliva pH 6.8 (sodium chloride 4.5 g, potassium
method. LOD and LOQ was found to be 1.5 ng/ml and chloride 0.3 g, sodium sulfate 0.3 g, ammonium acetate
5 ng /ml of pravastatin sodium respectively. It was also 0.4 g, urea 0.2 g, lactic acid 3 g and distilled water up to
observed that independent of concentration (10, 40, 1,000 ml, adjusting pH of solution to 6.8 by 1 M NaOH
and 60.0 ng/ml equivalent of pravastatin sodium), the solution). The patch was placed in a petridish contain-
excipients did not interfere in the analysis indicating ing 5 ml of simulated saliva and kept in a temperature-
selectivity of the method. After chemical oxidation controlled oven at 37 ± 2°C. At time intervals of 1, 2, 3, 4,
test, acid and alkaline hydrolysis, chromatograms 5, 6, 7, 8, 9 and 10 h, the patch was observed for change
showed additional peaks, due to the presence of deg- in color, thickness, and assayed for pravastatin sodium
radation products. However, in none of the case, the content using HPLC method as described in section
additional peaks were found to interfere with the peak 2.2.17.
of pravastatin sodium. Hence, the present method
was found to be appropriate for quantitative deter- 2.2.21  Accelerated stability testing as per ICH Q1A (R2)
mination of pravastatin sodium in the presence of its guidelines
degradation products. For analysis of plasma samples, Accelerated stability testing as per ICH Q1A (R2) guide-
protein was separated from the plasma by adding lines were performed by keeping the patches at a tem-
equivalent amount of methanol and centrifuging at perature of 40 ± 2°C and humidity of 75 ± 5% RH. The
10,000 rpm (REMI Pvt. Ltd., Vasai, India) for 10 min at samples were withdrawn at 0, 30, 60, and 90 days and the
15°C. Then, 200μl of protein-free plasma was collected physical characteristics (surface pH, weight gain/loss,
in amber colored microcentrifuge tubes, and evapo- mucoadhesive force, folding endurance, drug content
rated to dryness at 40–50°C. The resultant residue was and % drug remaining) were determined. The zero time
reconstituted with 200 μl of mobile phase and a 20 μl samples were used as controls.

© 2012 Informa Healthcare USA, Inc.

50  S.K. Maurya et al.

3  Results and discussion Mucoadhesive force (× 10−2N) = 38.45 + 8.26 X1 + 2.90X2

+ 1.02 X12 +0.62 X22 + 2.30 X1X2
3.1  Determination of weight variation, thickness, and The “Model F-value” of 31.39 obtained on applying
folding endurance ANOVA to the model for % cumulative drug release at
A total of nine experimental formulations were pre- 10th h implied the model was significant [Table 4]. There
pared employing different ratios of carbopol 934P and was only a 0.01% chance that a “Model F-Value” this
HPMC K4M in accordance with 32 full factorial design. large could occur due to noise.Value of “Prob > F” less
Physicochemical parameters of the patches were deter- than 0.05 indicated model terms were significant. In the
mined as depicted in Table 3. Thickness of the patches present case X1, X2, X1X2, X12 and X22 were found to be sig-
was found to range between 198.4 ± 0.85 and 255.2 ± 0.69 nificant model terms. The final equation for in vitro drug
µm. Total weight of the formulated patches ranged release in terms of coded factors was found to be as fol-
between 66.33 ± 1.15 and 89.53 ± 0.25mg. Assay values lows (r2 = 0.9907):
of drug content closer to 100%. Folding endurance was % Cumulative drug release at 10th hr = 96.53 + 1.60 X1 +
found to increase proportionately with an increase in the 0.045 X2 + 0.32 X12 + 0.03X22 + 0.04 X1X2
amount of carbopol 934P from formulation F1 to F9. The A square power of the factors in polynomial equations
increase in folding endurance may be attributed to the shows quadratic response where as two factors in a term
highly cross linked structure of carbopol 934P. The sur- indicate the interaction effect.
face pH of all the experimental formulations was within Three-dimensional response surface plots were
the range of 6.21 ± 0.31to 7.15 ± 0.10. Since the obtained drawn to estimate the effects of the independent vari-
values of surface pH were closer to the pH of buccal ables on each response. The response surface plots for
mucosa (pH = 6.8), it was anticipated that the experimen- mucoadhesion time and mucoadhesive force revealed
tal formulations would not cause any mucosal irritation. a corresponding increase in the mucoadhesion time
and mucoadhesive force of patches with an increase in
3.2  Optimization of Formulation amount of HPMC K4M [Figure 2(a) & 2(b)]. This may
Response surface methodology was employed for optimiza- be due to the fact that HPMC K4M is capable of forma-
tion of formulation. The “model F-value” of 675.60 obtained tion of hydrogen bonding with the glycoprotein rich
on applying ANOVA to the model for mucoadhesion time mucous (Desai and Pramodkumar, 2004). An increase in
implied that the model was significant. There was only a the amount of HPMC K4M in the experimental formu-
0.01% chance that a “Model F-value” this large could occur lations led to an increase in the possibility of formation
due to noise. Value of “Prob>F” less than 0.05 indicated of hydrogen bond between the polymer and the mucus
that the model terms were significant (Table 4). In this case thereby increasing the mucoadhesion time and mucoad-
X1, X2, X1X2, X12 and X22 were found to be significant model hesive force. The response surface plots also revealed
terms. The final equation for mucoadhesion time in term of a corresponding increase in the mucoadhesion time
coded factors was found to be as follows (r2 = 0.9932): and mucoadhesive force of patches with an increase
amount of carbopol 934P [Figure 2(a) & 2(b)]. This can
Mucoadhesion time (min.) = 634.32 + 13.66 X1 + 4.16 X2
be explained on the basis of fact that carbopol 934P
+ 0.78 X12 + 0.17 X22 + 1.00 X1X2
and HPMC K4M are mucoadhesive in nature. Thus an
As seen from Table 4, the “Model F-value” of 53.57 increase in the amount of mucoadhesive polymers in the
obtained on applying ANOVA to the model for mucoad- patch would facilitate enhanced contact of the system
hesive force implied the model was significant. Value of with the mucus leading to enhanced mucoadhesion time
“Prob>F” less than 0.05 indicated that X1, X2, X1X2, X12 and and mucoadhesive force.
X22 were significant model terms. The final equation for The % cumulative drug release was found to increase
ex vivo mucoadhesive force in the term of coded factor with increase in the amount of the hydrophilic polymers
was found to be as follows (r2 = 0.9917): carbopol 934P and HPMC K4M [Figure 2(c)]. which could

Table 3.  Pharmacotechnical evaluation of different formulations

Ex vivo Ex vivo
Weight of Folding Swelling % Drug mucoadhesion mucoadhesive %CDR
Formulation patch (mg) ± Thickness endurance± index (%) content Surface time force (×10−2N) after 10h
code S.D. (µm) ± S.D. S.D. ± S.D. ±S.D. pH ± S.D. (min.) ± S.D. ±S.D. ±S.D.
F1 66.33 ± 1.15 198.4 ± 0.85 295 ± 4.51 15.72 ± 0.62 99.50 ± 0.35 7.15 ± 0.10 593 ± 2.42 24.12 ± 0.62 91.12 ± 0.19
F2 70.34 ± 0.93 205.0 ± 0.92 315 ± 6.03 16.05 ± 0.58 99.21 ± 0.53 7.10 ± 0.06 597 ± 2.54 27.22 ± 0.44 93.57 ± 0.06
F3 73.25 ± 0.88 214.1 ± 0.75 325 ± 3.54 16.97 ± 0.94 99.10 ± 0.81 7.02 ± 0.14 603 ± 1.56 33.43 ± 0.43 95.28 ± 0.17
F4 76.66 ± 0.57 243.5 ± 0.67 336 ± 7.05 18.31 ± 0.75 99.33 ± 0.25 6.83 ± 0.09 610 ± 2.83 37.10 ± 1.20 96.02 ± 0.52
F5 81.67 ± 0.60 217.6 ± 0.75 339 ± 4.03 18.93 ± 0.79 99.53 ± 0.67 6.71 ± 0.11 615 ± 2.91 39.02 ± 1.32 96.14 ± 0.42
F6 84.40 ± 0.82 246.4 ± 0.63 342 ± 5.51 18.41 ± 0.63 99.80 ± 0.33 6.59 ± 0.04 619 ± 1.70 40.51 ± 0.57 96.67 ± 0.45
F7 85.66 ± 0.73 231.7 ± 0.57 358 ± 6.26 21.47 ± 0.27 99.79 ± 0.71 6.32 ± 0.25 622 ± 1.84 42.62 ± 0.72 97.18 ± 0.82
F8 87.12 ± 0.68 234.6 ± 0.87 360 ± 2.74 21.71 ± 0.18 99.78 ± 0.42 6.28 ± 0.08 625 ± 1.52 44.42 ± 0.53 97.75 ± 0.41
F9 89.53 ± 0.25 255.2 ± 0.69 362 ± 8.11 22.13 ± 0.52 99.91 ± 0.62 6.21 ± 0.31 628 ± 0.56 47.31 ± 0.58 98.97 ± 0.45

 Drug Delivery
 Bilayered transmucosal drug delivery system of pravastatin sodium  51

Table 4.  Analysis of variance (ANOVA) for factorial models: mucoadhesion time, mucoadhesive force and % cumulative drug release at
10th h
Source Sum of squares df Mean square F value p value Prob>F
Mucoadhesion time:
Model 1251.11 5 250.22 675.60 <0.0001 Significant
X1-Carbopol 934P 1120.67 1 1120.67 3025.80 <0.0001
X2-HPMC K4M 104.17 1 104.17 281.25 <0.0005
X1 X2 4.00 1 4.00 10.80 <0.0462
X12 22.22 1 22.22 60.00 <0.0045
X22 0.056 1 0.056 0.15 <0.0190
Residual 1.11 3 0.37
Cor total 1252.22 8
Mucoadhesive force:
Model 477.45 5 95.49 53.57 <0.0039 Significant
X1-Carbopol 934P 409.70 1 409.70 229.85 <0.0006
X2- HPMC K4M 50.52 1 50.52 28.34 <0.0129
X1 X2 5.34 1 5.34 2.99 <0.0164
X12 11.11 1 11.11 6.23 <0.0354
X22 0.79 1 0.79 0.44 <0.0450
Residual 5.35 3 1.78
Cor total 482.79 8
% cumulative drug release at 10th h:
Model 17.24 5 3.45 31.39 <0.0016 Significant
X1-Carbopol 934P 15.42 1 15.42 140.40 <0.0043
X2- HPMC K4M 1.60 1 1.60 14.58 <0.0116
X1 X2 0.008 1 0.008 0.074 <0.0313
X12 0.20 1 0.20 1.86 <0.0424
X22 0. 001 1 0.001 0.11 <0.0470
Residual 0.33 3 0.11
Cor total 17.57 8

Figure 2.  Response surface plots for 2(a) Y1 (mucoadhesion time), 2(b) Y2 mucoadhesive force) and 2(c) Y3 (% cumulative drug release).

© 2012 Informa Healthcare USA, Inc.

52  S.K. Maurya et al.
be attributed to the higher swelling of carbopol 934P and patches started eroding after 10 hours. Swelling index of
HPMC K4M, the higher swelling could attributed due to the buccoadhesive patch was found to be in the range of
higher water uptake by carbopol 934P and HPMC K4M. 15.72 ± 0.62 to 21.71 ± 0.18. Swelling index was found to be
Thus, higher amount of water uptake by carbopol 934P increase proportionately with an increase in the amount
and HPMC K4M may lead to considerable swelling of the of polymers in formulation F1 to F9 as well as with the
polymer matrix, allowing drug to diffuse out at a faster duration of study. The increase in swelling index may
rate from patch. The release rate of drug increased with be attributed to the hydrophilic nature of the polymers
increasing amount of HPMC K4M in the patch. It has owing to which the polymers absorbed water leading to
been postulated that relaxation and swelling of Eudragit their swelling. Thus, an increase in the amount of poly-
matrix increased as the amount of HPMC K4M in the mers resulted in an increased swelling index. Moreover,
patch. HPMC K4M is believed to hydrate and subse- since swelling of polymer is a gradual process, the swell-
quently produce water-logged regions (pores) within the ing index increased with the duration of study.
patch (Sakellariou et  al., 1988). Optimized formulation
(F9) was selected on the basis of higher mucoadhesion 3.5  Ex vivo mucoadhesion study
time, higher mucoadhesive force and higher in-vitro % Figure 4 and 5 shows ex vivo mucoadhesion time and ex
cumulative drug release after 10 h, with good desirability vivo mucoadhesive force for pravastatin sodium buccal
(0.989) with the help of Design Expert 8.0.2 software. patches respectively. Except formulation F1, the pre-
pared patches (F2 to F9) exhibited good mucoadhesive
3.3  Validation of experimental design force in the range of 37.10 ± 0.44 to 39.77 ± 0.43 x10−2 N,
In order to check the validity of the generated equations in which was more than the minimum value (>35 × 10−2N)
the optimization procedure, a new batch of tablets (extra reported to be required by a dosage form to adhere to
check point formulation) was prepared. Comparative buccal mucosa (Adel et  al., 2004). The formulations
analysis of the predicted value and experimental values (F1 to F9) exhibited mucoadhesion time in the range
using paired t-test indicated no significant (p < 0.05) of 601 ± 2.83 to 623 ± 1.56 min. The mucoadhesion time
difference between the two values thereby establishing and force were found to increase with an increase in
validity of the generated model (Table 5). the amount of mucoadhesive polymer. A correspond-
ing increase in the mucoadhesive strength of tablets was
3.4  Swelling behavior of patches observed with increase in concentration of carrageenan
Swelling index of a buccoadhesive patch is an impor- gum by shidhaye and coworkers (Shidhaye et al., 2010).
tant parameter of for uniform and prolonged release of Same authors have also observed increase in mucoadhe-
the drug and effective mucoadhesion (Chen and Cyr, sive strength of buccal patches with increase in amount
1970). Swelling of the prepared patches was found to of chitosan (Shidhaye et al., 2008). Similarly, Tehrani and
occur without any lag although the rate was slow ini- co-workers report that as the concentration of carbopol
tially and increased with time. Time required to achieve 934P and polycarbophil increased in the formulations,
maximum swelling was observed to be 10 hours as the the adhesion force also increased due to higher attractive

Figure 3.  Swelling behavior of patches.

Table 5.  Predicted and experimental values of responses for extra design check point formulation
Response Parameters Predicted value Experimental value Paired t-test
Mucoadhesion time (min.) 634.32 631.23 ± 1.58 p>0.05
Mucoadhesive force (× 10−2N) 45.55 42.11 ± 0.72 p>0.05
% CDR at 10th h 98.57 97.06 ± 0.35 p>0.05

 Drug Delivery
 Bilayered transmucosal drug delivery system of pravastatin sodium  53
force exerted by these polyacrylic acid polymers (Therani amount of mucoadhesive polymers. Maximum amount
et al., 2002). Similar observation for increase in mucoad- of mucoadhesive polymers in F9 provided the maximum
hesive strength with increase in mucoadhesive polymers polymer for contact as well as for interpenetration with
including Carbopol 934P for Eudragit matrix has also mucin, leading to maximum mucoadhesion time and
been made by Wong and co-workers (Wong et al., 1999). mucoadhesive force.
The patch containing 1.5% (w/v) of carbopol 934P and 4%
(w/v) of HPMC K4M i.e. F9 exhibited the highest mucoad- 3.6  In vitro drug release study
hesion time and mucoadhesive force of 623 ± 1.56 min In vitro drug release profile of the formulation F1 to F9
and 39.77 ± 0.43 × 10−2N respectively. This could be due to in comparison to immediate release (IR) tablet is shown
the reason that formulation F9 contained the maximum in Figure 6. Immediate release formulation was found
exhibit 100% of drug release within 2h of the study. In
comparison to immediate release tablet, all the prepared
patches (F1 to F9) were able to sustain the release of
drug up to 10th h releasing majority of the drug (90%) by
10h. this could be attributed to the presence of a matrix
of Eudragit S100 in the prepared patched. Eudragit S100
being hydrophobic in nature provides an impermeable
barrier and there by controls the release of the drug from
the patches. Formulation F1 exhibited the minimum
drug release of 93.83 ± 0.19 % during 10h of the drug
release study. The drug release was found to increase
from formulation F1 to F9. Increase in drug release with
increase in amount of hydrophilic polymer in mucoad-
hesive patches has also been observed by Shidhaye
and coworkers (Shidhaye et  al., 2008). Desai and co-
Figure 4.  Ex vivo mucoadhesion time. workers report increase in amount of drug release with
an increase in amount of HPMC K4M in the formulated
novel buccal adhesive system (Desai et al., 2004).This can
be explained on the basis of fact that carbopol 934P and
HPMC K4M are embedded in the hydrophobic matrix
of Eudragit S100. Carbopol 934P and HPMC K4M are
hydrophilic in nature. HPMC K4M has been reported
to generate to water-logged region (pores) on hydration
(Sakellariou et  al., 1988) whereas carbopol 934P has
been reported to undergo ionization at pH 6.8 (pKa = 6).
Ionization of carbopol 934P leads to the generation of
negative charges along the backbone of the polymer.
Repulsion between these like charges leads to uncoiling
of the polymer to produce an extended structure capable
of higher uptake of water. Thus owing to generation of
pores by HPMC K4M and uncoiling of carbopol 934P at
pH 6.8, the system absorbs more water and there by pro-
Figure 5.  Ex vivo mucoadhesive force.
motes dissolution which in turn leads to an increase in
release of water-soluble drug pravastatin sodium.
To investigate the release kinetics of pravastatin
sodium from bilayered buccal patches, the release data
was applied to various models and best fit was deter-
mined with help of PCP Disso V2.08 software (PCP
DissoV2.08, Pune, India). All the formulations showed
values of n in the range of 0.5 to 1.0 indicating that all
formulations followed non-Fickian diffusion mechanism
of drug release. The values of r2, k and n for optimized
formulation (F9) were found to be 0.9946, 23.0559 and
0.6597 respectively.

3.7  Drug release from backing layer

To ascertain the efficiency of backing membrane in
Figure 6.  In vitro drug release study of patches. providing unidirectional release of pravastatin sodium

© 2012 Informa Healthcare USA, Inc.

54  S.K. Maurya et al.
through the patch, drug release from backing layer was fluidity of stratum distendum leading to enhanced drug
investigated. Results of the study showed that no drug diffusivity. The formulation [F9 (P3)] optimized on the
was released in the donor compartment of diffusion basis of amount of penetration enhancer was subjected
cell during 10h of the study. This indicated that backing to pharmacotechnical characterization to confirm that
layer of ethyl cellulose was able to prevent the release of penetration enhancer did not affect the other pharma-
pravastatin sodium and swelling of the mucoadhesive cotechnical characteristics and the results shown in
layer did not change the integrity of the backing layer. Table 7. No significant difference (p<0.05) was found
between the evaluated pharmacotechnical parameters
3.8  Permeation studies of F9 and [F9 (P3)] establishing intactness of integrity of
Pravastatin sodium is hydrophilic in nature with a log the optimized formulation in presence of DMSO.
P value of 1.77 (Shidhaye et al., 2010). Due to its hydro-
philic character, the drug is not able to across the lipo- 3.9  Histopathological evaluation of buccal mucosa
philic membrane of the cells of buccal mucosa and The microscopic observation of the final formulation
hence exhibits low permeability through buccal mucosa. containing 4% (v/v) of DMSO [F9 (P3)] showed no del-
Thus, there is a need to enhance buccal permeation of eterious effect on the microscopic structure of mucosa.
the drug. To achieve enhanced permeation of pravas- As seen from Figure 2, no cell necrosis was observed and
tatin sodium by F9 through buccal mucosa, different cellular membrane was also found to be intact. Thus,
penetration enhancers (DMSO and Transcutol P at vari-
ous concentrations) were investigated. Figure 7 shows
comparative permeation of permeation of pravastatin
sodium through goat buccal mucosa for formulations
containing different penetration enhancers. The per-
cent drug permeated through optimized formulation
[F9 (P3)] was found to be 68.75 ± 0.84% in comparison to
47.73 ± 0.43% for formulation containing no permeation
enhancer. Permeability coefficient was calculated from
the graph and the results are listed in Table 6. Results of
the trials with 2% (v/v), 3% (v/v) and 4% (v/v) Transcutol
did not show any significant (P>0.05) improvement in
the permeation of pravastatin sodium as compared to
trials with DMSO. DMSO increased the permeability of
drug significantly (P<0.05) at level 4% (v/v) demonstrat-
ing highest permeation of the drug. This could be due to
Figure 7.  Comparative permeation of pravastatin sodium
disruption of the organized lipid structure in the intercel- through goat buccal mucosa in absence and presence of different
lular region of stratum distendum, resulting in increased penetration enhancers.

Table 6.  Permeation of pravastatin sodium through goat buccal mucosa in presence of different penetration enhancers
Permeability Flux Enhancement Statistical significance with respect
Formulation F9 containing coefficient (Kp) (mg/cm2. h) ratio to P0 using Student t- test
No penetration enhancer (P0) 0.667 1.620 – –
(2% v/v) DMSO (P1) 0.584 2.755 1.70 P<0.05
(3% v/v) DMSO (P2) 0.653 3.170 1.95 P<0.05
(4% v/v) DMSO (P3) 0.632 3.440 2.12 P<0.05
(2% v/v) Transcutol (P4) 0.554 2.543 1.50 P<0.05
(3% v/v) Transcutol (P5) 0.575 2.70 1.66 P<0.05
(4% v/v) Transcutol (P6) 0.613 2.943 1.81 P<0.05

Table 7.  Evaluation of optimized (F9) and final formulation [F9 (P3)]
Results
Parameter (F9) [F9 (P3)] Paired t-test
Weight (mg) 89.53 ± 0.25 90.05 ± 1.21 P>0.05
Thickness (µm) 255.2 ± 0.69 254.6 ± 0.95 P>0.05
Drug content (%) 99.91 ± 0.62 99.07 ± 1.82 P>0.05
Surface pH 6.21 ± 0.31 6.89 ± 0.94 P>0.05
Swelling index (%) 22.13 ± 0.52 21.87 ± 1.94 P>0.05
Drug release (%) 98.97 ± 0.45 98.08 ± 1.12 P>0.05
Mucoadhesion time (min.) 628 ± 0.56 626 ± 1.26 P>0.05
Mucoadhesive force (x10−2 N) 47.31 ± 0.58 46.04 ± 1.43 P>0.05

 Drug Delivery
 Bilayered transmucosal drug delivery system of pravastatin sodium  55

Figure 8.  Histopathological evaluation of sections of goat buccal mucosa treated with patch containing 4% v/v DMSO and control.

Figure 9.  Mean (±S.D.) plasma concentration of PS in New

Zealand white Rabbits (n = 3) following oral administration of
immediate release (IR) tablet and buccal patch.

formulation containing 4% (v/v) of DMSO [F9 (P3)] was Figure 10.  IVIVC of pravastatin sodium buccal patch [F9 (P3)].
found to be safe for buccal administration.
indicates increased (2.61 times) bioavailability of PS from
the buccal formulation (Table 8). Ka in case of IR tablet
3.10  Bioavailability studies was found to be 1.864 ± 0.08 h−1. Significant reduction in
The plasma concentration-time profile obtained after Ka was observed in case of buccal patch i.e.0.750 ± 0.08 h−1
administration of oral IR tablet and buccal bilayered in comparison to tablet. This may attributed to the sus-
patch to rabbits is shown in Figure 9. Tmax for the tablet tained release behavior of drug from the patch. Thus
was found to be 1 h whereas Tmax in case of buccal patch comparative pharmacokinetic studies of buccal patch
was found to be 4.00 ± 0.57 h, indicating delayed absorp- and IR tablet of pravastatin sodium in New Zealand white
tion of the drug from the optimized patch in comparison rabbits revealed delayed Tmax, increased Cmax, increased
to IR tablet. Moreover, plasma level reduced to zero in AUC and reduced absorption rate constant indicating
5 h in case of oral tablet but drug could be detected in enhanced absorption of the drug from the patch in a sus-
plasma up to 10 h in case of buccal patch corroborat- tained manner. Superior absorption of the drug from the
ing the in vitro sustained drug release behavior from patch is anticipated to reduce the dose and hence dose
patch in vivo (animal model) as well. Cmax was found related side effect of the drug. Sustained release of the
to be significantly (P<0.05) higher for buccal bilayered drug from patch is expected to reduce frequency of dos-
patch (67.34 ± 3.58 ng/ml) than IR tablet (58.73 ± 4.63 ng/ ing of the drug. Both these factors would in turn lead to
ml) implying better absorption of the drug from patch enhanced patient compliance towards the therapy.
comparison to IR tablet. Improved absorption of drug
from patch could be due to avoidance of hepatic first 3.11  In Vitro-In Vivo Correlation (IVIVC)
pass metabolism and gastric instability of the drug when The result of IVIVC study is shown in Figure 10. The value
the drug in given by buccal route in comparison to oral of regression coefficient (r2) was found to be 0.9690.
administration. There are four levels of correlation established by the
The AUC0-∞ in case of buccal bilayered patch USP Biopharmaceutics subcommittee for setting up
(350 ± 9.59 ng/ml×h) was found to be significantly higher the correlation between in vitro dissolution and in vivo
(P<0.05) of IR tablet (133.80 ± 8.25 ng/ml×h) which absorption.

© 2012 Informa Healthcare USA, Inc.

56  S.K. Maurya et al.

Table 8.  Mean (±S.D.) pharmacokinetic parameters for in-house immediate release tablet (IR tablet) and bilayered buccal patch of
pravastatin sodium in New Zealand white Rabbits (n = 3)
Type of Formulation
Pharmacokinetic parameters IR tablet Buccal patch
Cmax (ng/ml) 58.73 ± 4.63 67.34 ± 3.58
T max (h) 1.00 ± 0.35 4.00 ± 0.57
Ka (h )−1
1.864 ± 0.08 0.75 ± 0.08
AUC0-∞ (ng/ml/h) 133.80 ± 8.25 350.8 ± 9.59
Relative bioavailability – 262.18%

3.12  Stability study of final optimized patch [F9 (P3)]

The stability study of optimized patch [F9 (P3)] was per-
formed in simulated saliva (pH 6.8) and characteristics
such as color, thickness, and drug content were evalu-
ated. No change was observed in color of the patch until
the end of the study, indicating physical stability of the
patch. % drug remaining in the patch at the end of 10 h
was found to be 99.37 ± 0.57%, indicating maximum utili-
zation of the drug incorporated.

3.13  Accelerated stability testing as per ICH Q1A (R2)

guidelines
Accelerated stability studies were performed as per ICH
Q1A (R2) guidelines at 40 ± 2°C and 75 ± 5% RH. The effect
Figure 11.  Shelf life determination of buccal patch [F9 (P3)]. The black of storage conditions on average weight gain/loss, surface
line shows the curve obtained by plotting percent drug remaining pH, mucoadhesive performance, folding endurance, and
versus time where as the grey line shows the plot of 95% one sided drug content was investigated. No significant (P>0.05)
confidence limit for the percent drug remaining values versus time.
change was observed in the physical characteristics of the
buccoadhesive patches. This indicated physical stability
1. Level A: The in vitro dissolution curve matches with of the patch during the period of study. Percent drug con-
the in vivo absorption curve in a point to point (1:1) tent at the end of 90 days was found to be 99.12 ± 0.006.
manner and the two curves nearly or completely Shelf life was estimated as the time at which the 95% one-
overlap each other. sided confidence limit for the mean curve intersected the
2. Level B: The in vitro mean dissolution time (MDT), acceptance criterion of 90% percentage label claim. The
calculated by moment analysis, shows correlation data was analyzed using Sigmaplot ™ 10 software (Cranes
with the in vivo mean residence time (MRT). Software International, Bangalore, India). Percentage
3. Level C: Among t50%, t90%, etc one of the dissolution label claim (% drug remaining) was plotted against time
parameters shows correlation with one of the phar- in months to determine the shelf life (Figure 11). Shelf
macokinetic parameters like AUC, Cmax, Tmax, etc. life was found to be 20 months on the basis of the plot
4. Level D: The disintegration correlates qualitatively obtained using the software.
with the in vivo behavior or the in vitro data corre-
lates qualitatively with the in vivo data. Conclusions
Mucoadhesive bilayered patches of pravastatin sodium
 Level A correlation is point to point correlation and thus were successfully formulated for buccal application using
is the highest degree of correlation which is applicable 32 full factorial design. The formulation [F9 (P3)] exhib-
to modified release systems than any other level of cor- ited satisfactory swelling index, mucoadhesive proper-
relation. The in vitro dissolution data was determined ties, sustained drug release for 10 h and permeation of
and compared with the in vivo drug absorbed. A value the drug through goat buccal mucosa without causing
of r2 = 0.9690 indicated linear relationship between the any tissue damage. Model fitting of the in vitro release
percent drug released and the percent drug absorbed kinetics revealed non-Fickian mechanism of drug release
establishing a level A correlation between the two data as from the optimized patch. Pharmacokinetic studies
shown in figure 10, where revealed more than two folds increase in bioavailability
% drug absorbed from patch and sustained release behavior of the drug from in com-
 [F9 (P3)] = 0.9690 × % drug released (5) parison to IR tablet. Stability studies in simulated saliva
and under accelerated conditions established physical
 This signifies that in vivo absorption of the drug from the stability of the patch and chemical stability of drug. Shelf
patch can be predicted from in vitro drug release data. life of the patch was found to be 20 months. Thus, the

 Drug Delivery
 Bilayered transmucosal drug delivery system of pravastatin sodium  57
present investigation demonstrates the potential of the Desai KGH, kumar TMP. (2004). Preparation and evaluation of a novel
developed pravastatin sodium bilayered buccal patch buccal adhesive system. AAPS PharmSci Tech, 5, Article 35.
Maurya SK, Pathak K, Bali V. (2010). Therapeutic potential of
as an alternative dosage form in surmounting poor oral mucoadhesive drug delivery systems−An updated patent review.
bioavailability and gastric instability of the drug. Recent Pat Drug Deliv Formul, 4(3), 256–265.
Miller NS, Chittchang M, Johnston TP. (2005). The use of mucoadhesive
polymers in buccal drug delivery. Adv Drug Deliv Rev, 57,
Acknowledgement 1666–91.
Mumtaz AM, Ch’ng HS. (1995). Evaluation of bioadhesive buccal
The authors are thankful to Panax Matrix Research tablets containing triamcinolone acetonide in healthy volunteers.
Laboratory, Meerut, Uttar Pradesh, India and Pt. Deen Int J Pharm, 121, 249–254.
Dayal Upadhaya Pashu Chikitsa Vigyan, Mathura, Uttar Otter K, Mignat C. (1998). Determination of pravastatin in human
Pradesh, India for their assistance in HPLC analysis and plasma by high-performance liquid chromatography with
performing histopathological study respectively. Mr. ultraviolet detection. J Chromatogr B Biomed Sci Appl, 708,
235–241.
Suresh is thankful to the All India Council for Technical Rathbone M, Drummond B, Tucker B. (1994). The oral cavity as a site
Education (AICTE), India, for providing financial support for systemic drug delivery. Adv Drug Deliv Rev, 13, 1–22.
to carry out the project. Sakellariou P. (1988). A study of the leaching/retention of water-
soluble polymers in blends with ethylcellulose using torsional
braid analysis. J Control Rel, 7, 147–157.
Declaration of interest Shidhaye SS, Saindane NS, Sutar S, Kadam VJ. (2008). Mucoadhesive
bilayered patches for administration of sumatriptan succinate.
All the authors report no declaration of interest. AAPS PharmSci Tech, 9, 909–916.
Shidhaye SS, Thakkar PV, Dand NM, Kadam VJ. (2010). Buccal
drug delivery of pravastatin sodium. AAPS PharmSci Tech, 11,
References 416–424.
Adel N, Ismail F, Boraie N, Mortada L. (2004). Mucoadhesive delivery Tembhurne SV, Sakarkar DM. (2008). Transport of glucose in isolated
systems. II. Formulation and in-vitro/in-vivo evaluation of buccal rat ileum preparation: An alternative research tool for absorption
mucoadhesive tablets containing water soluble drugs. Drug Dev study. Indian J Physiol Pharmacol, 52, 211–213.
Ind Pharm, 30, 995–1004. Tehrani MR, Jazayeri G, Toliyat T, Bayati K, Khalkhali K, Shamimi
Boyong LI, Nangia A, Dely A, Wong D. (2003). Stable pharmaceutical K, Miremadi A, Dorkoosh FA. (2002). Formulation and
compositions containing pravastatin. WO Patent 03/000239. in vitro evaluation of novel buccoadhesive tablet formulation of
Chen J L, Cyr GN. (1970). Compositions producing adhesion through prednisolone. Acta Pharm, 52, 121–127.
hydration. Manly RS, ed. In: Adhesion in Biological Systems. Wong CF, Yuen KH, Peh KK. (1999). Formulation and evaluation of
Academic Press, New York, 163–167. controlled release Eudragit buccal patches. Int J Pharm, 178, 11–22.

© 2012 Informa Healthcare USA, Inc.