Naproxen has been proved to be effective in both experimental and clinical pain like rheumatoid arthritis, osteoarthritis, juvenile arthritis and acute gout without any serious cardiovascular or respiratory side effects (Uziel et al., 2000; Hashkes et al., 2003). The drug is lipid soluble, practically insoluble at low pH and freely soluble at high pH. One of the most important commonly used methods for controlling drug release is to form a matrix system with the help of hydrophilic, inert and hydrophobic polymers. Ethyl Cellulose (EC) is hydrophobic polymer and is essentially tasteless, odorless, colorless and physiologically inert. It has been extensively used as a pharmaceutical vehicle, tablet binder in preparing microcapsules (Sajeev et al., 2002) coating materialfor tablets/granules (Pearnchob and Bodmeier, 2003; Sadeghi et al., 2003; Dashevsky et al., 2004)and matrix forming material for sustained release dosage forms (Zabed et al., 2002; Pruthvipathy et al., 1995). There are few reported studies of matrix tablets prepared by wet granulation for controlling the drug release. In the present study EC in various proportions was used to develop a sustained release matrix system for naproxen. In addition the release data of the optimum formulation was fitted in release kinetic models.
MATERIALS AND METHODS
Materials: Naproxen (Shazoo Labatories, Lahore, Pakistan), ethyl cellulose
45cps (Highnoon Labatories, Lahore, Pakistan), Lactose (BDH, Poole, England),
Magnesium stearate (Fluka, Buchs, Switzerland), Potassium dihydrogen phosphate
(Merck, Darmstadt, Germany) and Disodium hydrogen phosphate (Sigma Aldrich,
St. Louis, Mo, USA) were used as received.
Matrix tablets: For preparing hydrophobic matrix tablet, naproxen (33.33%) and various percentages of EC and lactose as mentioned in Table 1 were first sieved and blended in a Kenwood mixer (Kenwood, Geesthacht, Germany) for 5 min. The powder blend was first granulated with small amount of alcohol (25 mL/100 g) and wet mass was sieved through mesh No. 6 and dried at 60°C for 1 h in an oven (Memmert, Schwabach, Germany). The dried granules were passed through sieve No.10 and the fractions of granules retained on the sieve were discarded. Magnesium stearate in 1.67% w/w was used for lubrication of various granules, which were compressed separately by single punch machine (Emmy, Lahore, Pakistan) using 12 mm punches and dies at fixed compression force of 1500 lb. The weight of tablet was adjusted to 600 mg containing 200 mg naproxen.
Test matrix tablets: Test matrix tablet was prepared using formulation
I (Table 1) by slightly modifying the wet granulation method
as described earlier. The granules were processed with 16.7% w/w of naproxen
instead of 33.33% w/w, 5% EC and 60% lactose. The remaining amount of drug (16.63%
w/w) was then dry mixed with the granules. Magnesium stearate (1.67% w/w) was
then thoroughly mixed with the granules that were tabletted at a fixed (1500
lb) compression force.
Formulations of naproxen matrix tablets (values in percentage)
Weight variation, hardness and friability of tablets: In order to determine the uniformity of tablet weight, twenty tablets of each formulation were randomly selected and weighed using class A weight balance (Precisa, Dietikon, Swizerland) and their percentage variation was determined. Hardness of tablets was determined using automatic hardness tester (Curio, Lahore, Pakistan). Twenty tablets of each formulation were used and the average hardness value was calculated. The tablets of each formulation were also subjected to friability testing employing friabilator (Emmy, Pakistan). Ten tablets were placed in the tumbling chamber and rotated precisely for 4 min at a speed of 25 rpm. The weight of ten tablets prior to their placement in the chamber and at the end of the test was recorded. The percentage weight loss was then calculated. Triplicate measurements were conducted for each formulation.
In vitro release studies: The dissolution studies were performed using USP apparatus type II (Pharma Test, Hainburg, Germany). The dissolution medium consisted of 900 mL of phosphate buffer solution pH (7.4) maintained at 37±0.5°C and stirred at 50 rpm. Samples (5 mL) were withdrawn at predetermined time intervals (0, 0.5, 1, 1.5, 2, 3, 4, 6, 8, 10 and 12 h) with automatic sampling unit (Watson Marlo, Stockholm, Sweden). Samples were filtered through Sinter filter 10 μm (Pharma Test, Hainburg, Germany) to remove suspended and insoluble tablet components and analyzed by UV spectrophotometer (Shimadzoo, Kyoto, Japan) at 332 nm.In addition to the release profile at pH 7.4, the test matrix tablets were also tested at two different pH values namely pH 1.0 (0.1 M HCl) and pH 4.0 (phosphate buffer). The effect of stirring speed on drug release rate was also evaluated on the test matrix tablets. The different stirring speeds used were 50, 100 and 150 rpm. Moreover, test matrix tablets were divided into two portions and packed in an airtight amber glass bottles and kept at 8 and 37°C. The samples of tablets were drawn after 3 months and 6 months and evaluated for stable in vitro release profile. In the data analysis of each formulation, cumulative percentage of drug release was calculated using mean of six samples readings.
In vitro drug release kinetics: The dissolution data of matrix
tablet formulations (Table 1) and the test matrix tablets
were fitted to zero order, Higuchi and Peppas model for determining the release
rates. Equations for (a) zero-order release (Xu and Sunada, 1995) and (b) Higuchi
model (Higuchi, 1963) are given below.
Where Qis the percentage of drug released at time t. k1 and k2 are the release rate constants for zero-order and Higuchi models, respectively. Regression analysis was performed to obtain the release rate constant and the values of coefficient of determination (r) were also compared. Moreover, equation for drug release mechanism from the matrix system was also used as explained by Ritger and Peppa (1987).
Where Mt/Mα is the fraction of drug released at time t, k3
is the kinetic constant and n is the so-called diffusion exponent, indicative
of the mechanism of the drug release. The equation generally holds for Mt/Mα>70%
of drug release (n = 0.45 or 0.45<n<0.89 or n>0.89, indicates Fickian
diffusion or anomalous transport or Case II transport kinetics, respectively).
In addition, the similarity factor f2 (Moore and Flanner, 1996) was
used to compare the difference of dissolution profiles of the test matrix tablets
prepared at different stirring speeds and is given below: Where n is the number
of dissolution samples taken, Rt and Tt are the individual
percentages dissolved at each time point for the reference and test dissolution
profiles, respectively and the f2 value between 50 and 100 suggests
that the data of two dissolution profiles are similar.
RESULTS AND DISCUSSION
The weight variation of all the compressed tablets was well within the acceptable
limits of British pharmacopoeia (BP, 2004) indicating that the filling of the
granules in the die of tablet machine was uniform. The hardness of each tablet
formulation was above 5 kg and negligible weight loss (less than 0.8%) in the
friability test was observed. Furthermore, the tablets exhibited good physical
appearance with no defects such as capping, lamination, picking and sticking
Percentage weight variations, hardness and weight loss
in friability test of various formulations
Effect of various percentages of ethyl cellulose on in
vitro drug release from naproxen tables
The results of weight variation, hardness and friability are presented in Table
2. Dissolution profiles of all the compressed tablets having standard deviation
less than ±2 were determined and explained in the following sections.
Drug release from matrix tablets: Figure 1 shows the
release of naproxen from hydrophobic matrix tablets containing various percentages
of EC. The tablets containing 5, 10, 15 and 20% EC released about 41, 39, 38
and 37.8% in 12 h testing intervals as shown in Table 3, respectively.
EC had pronounced effect in decreasing the drug release rate from hydrophobic
matrix tablets. The slower drug release rate from such tablets was due to formation
of uniform ethyl cellulose coating on the individual drug particles. However,
increasing the percentage of EC had no significant difference in the release
rates of drug. This was due to the fact that the amount of ethanol used during
granulation of various formulations was insufficient to wet all the particles
of EC, which were in granular form and could not provide a uniform coating around
the drug particles. The results found in this study were not in good agreement
with the reported study (Zabed et al., 2002) in which increasing percentages
of micronized EC produced slower drug release rate. As in the reported study,
micronized EC was used which could be more easily wetted by the granulating
liquid and provide more uniform coating around the drug particles.
Moreover, the release profiles of tablet formulations were fitted to zero order,
Higuchi and Peppas model. Regression analysis was performed to obtain co-efficient
of determination (r) and was compared as shown in Table 4.
The values of r obtained from zero order models are almost greater than Higuchi
model and therefore the drug release rate from these formulations followed the
zero order kinetics.
Effect of granulation process on in vitro release
from naproxen table containing 5% ethyl cellulose
Drug release data was also fitted to Peppas model, which showed slope
values in the range of 0.737 to 0.740 indicating a Fickian diffusion release
Influence of granulation process on drug release: Only about 40% of
the drug was released in 12 h from all the hydrophobic matrix tablets containing
various percentages of EC and failed to improve the extent of drug release.
Therefore, the matrix tablets were also prepared by slightly modifying the granulation
process. Drug release profile of the matrix tablets prepared with different
granulation processes is shown in Fig. 2. Surprisingly, the
drug release pattern of the matrix tablets prepared with modified method (test
tablets) was linear and the extent of drug release was also improved in comparison
to standard method used in tablet formulations (Fig. 1). About
65% of drug was released in 8 h from test matrix tablets compared to only 29%
of drug released from the same formulation using standard granulation method
(Table 3). Moreover, about 22% drug was released in 2 h as
burst release from test matrix tablets and was probably attributed to the dissolution
of drug from the surface of tablets. But, further penetration of the dissolution
medium was hindered due to the hydrophobic nature of EC on the drug particles
leading to slower drug release for prolong period of time. The r-values obtained
from test matrix tablets using zero order and Higuchi models were 0.998 and
0.973, respectively and are shown in Table 4. This clearly
indicates that test matrix tablets follow zero order release kinetic. Drug release
data of test matrix tablet was also fitted to Peppas model, which showed the
slope values of 1.074 indicating anomalous diffusion mechanism.
Influence of pH and stirring speed on drug release rate: Figure
3 shows the drug release profile from the test matrix tablets at pH 1, 4
and 7.4. At the lower pH (pH 1 and 4) the release profiles were essentially
similar and negligible amount of drug was released during 12 h testing interval
indicating the poor dissolution of naproxen at lower pH.
Mean in vitro release of naproxen from hydrophobic
matrix tablets containing various percentages of EC (n = 6)
Figure in brackets = Standard deviation
Coefficient of determination r and diffusion exponent of
various formulations using drug release kinetic models
||Effect of pH on in vitro drug release from test matrix
Effect of stirring speed on in vitro drug release
from test matrix tables
Effect of storage at 8°C on in vitro drug release
from test matrix tables
Effect of storage at 37°C on in vitro drug release
from test matrix tables
At pH 7.4 there was apparent difference in drug release profile and almost 100% of the drug was released during 12 h. From the results, it can be concluded that naproxen release from the test matrix tablets is pH dependent. The pH dependency is due to difference in the solubility of naproxen at various pH values. As the naproxen has maximum solubility at pH 7.4 in contrast to pH 1 and 4, therefore greater amount of drug was released from the test matrix tablets at higher pH.
Figure 4 shows the drug release profile from the test matrix
tablets at different stirring speeds. The drug release was seemed to increase
slightly as the stirring speed was increased. The data was statistically compared
using f2 equation and it was found that the dissolution data obtained
at 50, 100 and 150 rpm were similar. The f2 values obtained from
dissolution data at 50 rpm vs 100 rpm and 50 rpm vs 150 rpm were 64.5 and 77.6,
respectively suggesting that the rate of drug release is independent of stirring
Influence of storage conditions on drug release rate of test matrix tablets:
Figure 5 and 6 shows the drug release profile
of test matrix tablets at 8 and 37°C after six months of storage, respectively.
The release profiles were similar and comparable at both temperatures. It seemed
that the drug and the polymer in the test matrix tablets are insensitive to
moisture and temperature during storage. As no significant change in the release
profiles of test matrix tablets was observed at different temperatures, therefore
no further change in the rate of drug release is expected for prolong storage.
The drug release rates from various hydrophobic matrix tablets were comparable and no significant change in the drug release profile was observed even by incorporation of increasing proportions of ethyl cellulose. However, the test matrix tablets prepared by modifying the wet granulation method were found to produce desirable release rate. The release of test matrix tablets was found to be pH dependent due to difference in the solubility of naproxen in various dissolution media. Moreover, the drug release profile of test matrix tablets was independent of stirring speed and stable upon storage at different temperatures.