Optimization of Metronidazole Emulgel

The purpose of the present study was to develop and optimize the emulgel system for MTZ (Metronidazole), a poorly water soluble drug. The pseudoternary phase diagrams were developed for various microemulsion formulations composed of Capmul 908 P, Acconon MC8-2, and propylene glycol. The emulgel was optimized using a three-factor, two-level factorial design, the independent variables selected were Capmul 908 P, and surfactant mixture (Acconon MC8-2 and gelling agent), and the dependent variables (responses) were a cumulative amount of drug permeated across the dialysis membrane in 24 h (Y 1) and spreadability (Y 2). Mathematical equations and response surface plots were used to relate the dependent and independent variables. The regression equations were generated for responses Y 1 and Y 2. The statistical validity of the polynomials was established, and optimized formulation factors were selected. Validation of the optimization study with 3 confirmatory runs indicated a high degree of prognostic ability of response surface methodology. Emulgel system of MTZ was developed and optimized using 23 factorial design and could provide an effective treatment against topical infections.


Introduction
When gels and emulsions are used in a combined form the dosage forms are referred to as emulgels [1,2]. As the name suggests they are the combination of emulsion/microemulsion and gel. In recent years, there has been great interest in the use of novel polymers with complex functions as emulsi�ers and thickeners because the gelling capacity of these compounds allows the formulation of stable emulsions by decreasing surface and interfacial tension and at the same time increasing the viscosity of the aqueous phase. In fact, the presence of a gelling agent in the water phase converts a classical emulsion into an emulgel. Both oil-in-water and water-in-oil emulsions are used as vehicles to deliver various drugs to the skin [3]. Emulsions possess a certain degree of elegance and are easily washed off whenever desired. ey also have a high ability to penetrate the skin. Emulgels for dermatological use have several favorable properties such as being thixotropic, greaseless, easily spreadable, easily removable, emollient, nonstaining, water soluble, having longer shelf life, biofriendly, transparent, having and pleasing appearance [4].
Several antifungal and antibacterial agents are available in the market in different topical preparations (e.g., creams, ointments, and powders for the purpose of local dermatological therapy). One of these antibacterial agents is Metronidazole (MTZ), which has antibacterial properties. MTZ is used for the treatment of acne vulgaris, skin lesions, wound drainage, and wound odor. MTZ possesses poor water solubility and hydrophobicity; hence such drugs pose problems in a topical drug delivery. Hence, for solubilization of MTZ, formulation of microemulsion-based gel appeared to be a viable approach.
In the development of emulgel dosage form, an important issue is to design an optimized formulation with an appropriate drug diffusion rate in a short period of time and minimum number of trials. Many statistical experimental designs have been recognized as useful techniques to optimize the process variables. For this purpose, a computerbased optimization technique with a 2-level factorial design utilizing a polynomial equation has been widely used. is technique requires minimum experimentation and time, thus is far more effective and cost-effective than the conventional methods of formulating emulgel dosage forms. e aim of this investigation was to develop an emulgel system of MTZ using xanthan gum as a gelling agent and optimization of the formulation by applying the 2-level factorial design.

Materials and Methods
Metronidazole was obtained as a gi sample from Indochem Health Specialities Pvt. Ltd. Daman (India). Xanthan gum was a gi sample from SD. Fine Chemicals, Mumbai, and Capmul 908-P, Acconon MC-8-2 EP (polyoxyethylene (8) caprylic/capric glycerides) were gied by Abitec Corporation, USA. Propylene glycol was purchased from SD. Fine Chemicals, Mumbai, India. Methanol (AR grade) was purchased from Loba Chemical Mumbai (India). Double distilled water was used for all experiments.

Screening of Oils and Surfactants for Microemulsions.
To �nd out suitable oil, surfactant, cosurfactant phase in microemulsions, the solubility of MTZ in various oils, surfactants, and cosurfactants were screened like Capmul MCM L, Capmul MCM L8, Capmul MCM C8, Capmul 908 P, Acconon MC8, Tween 80, Span 80, Tween 20, Caproyl 90, and Propylene glycol. An excess of MTZ was added individually to the oils, surfactants, and cosurfactant (5 g each) in screw capped tubes. en the mixture was vortexed using a cyclomixer for 10 min in order to facilitate proper mixing of drug with the vehicles. Mixtures were then shaken for 48 h in a mechanical shaker (Remi, Mumbai, India) maintained at 25 ± 2 ○ C. Aer 72 h, each sample was centrifuged at 5000 rpm for 10 min. e supernatant (0.5 mL) was diluted suitably, and the amount of MTZ present in the supernatant was analyzed by UV-spectrophotometer at 277 nm. e oil, surfactant, and cosurfactant phase that showed high solubility for MTZ were used in the preparation of microemulsions containing 1% MTZ.

Construction of Phase Diagrams and Formulation of MTZ-Loaded
Microemulsions. MTZ showed maximum solubility in Capmul 908 P as compared to other oils; hence it was selected for further studies. Acconon as a surfactant and propylene glycol as a cosurfactant showed better solubility for MTZ and good emulsifying properties with Capmul 908 P as oil phase. Pseudoternary phase diagrams were constructed using water titration method. Surfactants and cosurfactants ( mix ) were mixed in different volume ratios (1 : 1, 2 : 1, 3 : 1, and 4 : 1). Oil and mix mixture were mixed thoroughly in different volume ratios (1 : 9, 1 : 8, 1 : 7, 1 : 6, 1 : 5, 1 : 4, 1 : 3, 1 : 2, and 1 : 1). Distilled water was added dropwise to the different mixtures of oil/ mix till a cloudy dispersion was obtained. Pseudoternary plots were constructed using Chemix school trial version soware 3.00, and microemulsions were prepared based on ternary phase diagram.

Preparation of Emulgel.
From preformulation studies xanthan gum was selected as the gel matrix to prepare the emulgel formulation. Xanthan gum was dispersed in puri�ed water with constant stirring; the pH was adjusted from 6 to 6.5 using tri ethanol amine (TEA) [5]. Xanthan gum was slowly mixed with microemulsion in 1 : 1 ratio with constant stirring. Aer xanthan gum was entirely dissolved in the microemulsion, milky white emulgel was obtained (Table 1).

Experimental
Design and Statistical Analysis. 2 3 full factorial design was used to statistically optimize the formulation factors and evaluate main effects, interaction effects on the amount of MTZ permeated in 12 h, and spreadability [1,2]. A 3-factor, 2-level factorial design was used to explore response surfaces and constructing second-order polynomial models with Design Expert soware (Version 7.1, Stat-Ease Inc., Minneapolis, MN). e 2-level factorial design was speci�cally selected since it requires fewer runs than other experimental designs. A design matrix comprising 8 experimental runs was constructed. e nonlinear computer, generated quadratic model is given as where is the measured response associated with each factor level combination; 0 is an intercept; 1 to 123 are regression coefficients computed from the observed experimental values of ; and 1 , 2 , and 3 are the coded levels of independent variables. e terms 1 , 2 , and 3 ( = 1, 2 or 3) represent the interaction and quadratic terms, respectively. e dependent and independent variables were with their low and high levels, which were selected based on the results of pseudoternary phase diagrams. e proportion of oil ( 1 ), mix ( 2 ), and gelling agent ( 3 ) used to prepare the 8 experimental trials and the respective observed responses are given in Table 2. All other formulation and processing variables were kept invariant throughout the study. 2.1.7. Rheological Studies. e viscosity of the different emulgel formulations was determined at 25 ○ C using a �rook�eld viscometer (LV2, �rook�eld Inc., USA) equipped with the tbar spindle number 92, and the viscosities were recorded at different rotational speeds of 10, 20, 50, and 100 RPM.

Spreading
Coefficient. e spreading coefficient (spreadability) of the formulations was determined using an apparatus described by Jain et al. e apparatus consisted of two glass slides (7.5 × 2.5 cm), one of which was �xed onto the wooden board and the other was movable, tied to a thread which passed over a pulley, carrying a weight. Formulation (1 g) was placed between the two glass slides. Weight (100 g) was allowed to rest on the upper slide for 1 to 2 minutes to expel the entrapped air between the slides and to provide a uniform �lm of the formulation. e weight was removed, and the top slide was subjected to a pull obtained by attaching 30 g weight over the pulley. e time (sec) required for moving slide to travel a premarked Propylene glycol 16.14 ± 1.07 distance (6.5 cm) was noted and expressed as spreadability. Spreadability is calculated by using the following formula: where is weight tied to upper slide, is length of glass slides, and is time taken to separate the slides 2.1.9. Drug Content Determination. MTZ content in emulgel was measured by dissolving known quantity of emulgel formulation in methanol by sonication. Absorbance was measured aer suitable dilution at 277 nm using UV-Vis spectrophotometer.

In Vitro Diffusion
Studies. Franz diffusion cell was used for the drug diffusion studies. Emulgel (1 g) was evenly applied onto the surface of dialysis membrane. e dialysis membrane was clamped between the donor and the receptor chamber of diffusion cell. e receptor chamber was �lled with freshly prepared phosphate buffer (pH 7.4). e receptor chamber was stirred by magnetic stirrer. e aliquots (1 mL) were collected at time intervals of 1 h up to 12 h. Samples were analyzed for drug content by UV-Vis spectrophotometer aer appropriate dilutions. Cumulative corrections were made to obtain the total amount of drug release at each time interval.
2.1.11. Ex Vivo Diffusion Studies. Ex vivo diffusion study was carried out by using rat skin, and procedure was similar to that of in vitro diffusion study. Cumulative corrections were made to obtain the total amount of drug diffused at each time interval and ex vivo parameters were calculated. e average cumulative amount of drug permeated per unit surface area of the skin was plotted versus time [6]. e slope of the linear portion of the plot was calculated as �ux ss ( g/cm 2 /h), and the permeability coefficient was calculated using the following formula: where is the permeability coefficient and is the total amount of drug.
e enhancement of drug penetration due to microemulsion formulation compared with marketed gel Metrogyl (J. B. Pharmaceuticals) was noted as enhancement factor (EF) [7] which was calculated using the following formula:

Selection of Excipients for Formulation of Microemul-
sions. e solubility of MTZ in various oils, surfactants and cosurfactant was analyzed in order to select components for microemulsions. MTZ is a BCS IV drug having extremely poor water solubility of (1 mg/mL). Due to poor solubility and permeability, microemulsions are attractive approaches to overcome bioavailability problems. Solubility of MTZ in various oils and mix was determined (Table 3). It was found that MTZ was found to have maximum solubility in Capmul 908 P, Acconon, and propylene glycol (35.14, 51.5, and 16.14 mg mL −1 , resp.). Hence Capmul 908 P was selected as oil phase and Acconon, and propylene glycol was selected as surfactant and cosurfactant for further studies.

Construction of Pseudoternary
Diagrams. For the construction of pseudoternary phase diagrams, MEs, from the selected oil and surfactants, were prepared in different volume ratios (1 : 9, 1 : 8, 1 : 7, 1 : 6, 1 : 5, 1 : 4, 1 : 3, 1 : 2, and 1 : 1). Figure 1 presents the pseudoternary phase diagrams with various weight ratios of Acconon/propylene glycol (1 : 1, 2 : 1, 3 : 1, and 4 : 1). From Figure 1, it was found that the ME area was maximum at mix ratio of 2 : 1. Hence this ratio was selected for preparation of drug-loaded MEs. At 1 : 1 ratio, the concentration of Acconon may not be sufficient to form a tightly packed barrier �lm. e ME region for 3 : 1 and 4 : 1 was signi�cantly lesser than 1 : 1 and 2 : 1. Acconon is a C 8 PEG-caprylic glyceride with HLB of 14 and molecular weight of 400 daltons. At higher concentration of Acconon, some of the molecules may be involved in formation of micelles. Micelles lie in the colloidal size range and hence may be contributing to the cloudiness of the dispersion. us we may presume that, for 3 : 1 and 4 : 1 ratios, reduced ME region is seen in the ternary plots.

Physical Appearance and pH
Determination. e MTZ emulgels were white viscous creamy preparation with a smooth homogeneous appearance. e pH values of all prepared formulation ranged from 6.0 to 6.9, which are considered acceptable to avoid the risk of irritation upon application to the skin because adult skin pH is 5.5.

Rheological Studies.
Rheological behavior of the emulgels indicated that the systems were shear thinning in nature showing decrease in viscosity at the increasing shear rates. e viscosity data has been summarized in Table 4. As the shear stress is increased, the normally disarranged molecules of the gelling material are caused to align their long axes in the direction of �ow. Such orientation reduces the internal resistance of the material and hence decreases the viscosity [8]. An increase in the concentration of xanthan gum (1 to 3%) was expected to show increase in viscosity. However the microemulsions incorporated into the gel contained varying amounts of oil/ mix which could be contributing to the viscosity of the formulations. Hence no particular trend was evident, though all formulations exhibited shear thinning properties.

Spreading Coefficient.
One of the essential criteria for an emulgel is that it should possess good spreadability. Spreadability depends on the viscosity of the formulation and physical characteristics of the polymers used in the formulation. A more viscous formulation would have poor spreadability. Spreadability is a term expressed to denote the extent of area on which the gel readily spreads on application to the skin. e therapeutic efficacy of a formulation

RPM
Viscosity (mPas) also depends upon its spreading value. e spreadability of different emulgel formulations is shown in Figure 2. It shows that the 3 formulation shows higher spreading coefficient as compared to other formulations.

In Vitro Diffusion
Study. e in vitro diffusion pro�les of MTZ from various emulgel formulations are represented in Figure 3. It was observed that all the formulation had become li�ue�ed at the end of experiments, indicating water diffusion through the membrane. In general, it can be observed from the �gures that all emulgels showed better release as compared to plain drug formulation. e higher drug release was observed with formulations 3 and 5 . is �nding may be due to presence of Capmul 908P in its low level and emulsifying agent in its high level. is led to an increase in the hydrophilicity of the emulgel, which in turn facilitated penetration of the release medium into the emulgel F 3: Cumulative percent of MTZ released from 1 to 8 emulgel formulations through dialysis membrane using Franz diffusion cell.

Formulation Optimization by Experimental Design.
A three-factor, two-level full factorial experimental design was used to optimize the formulation variables as the response surface methodology requires 8 experiments. e independent variables and the responses for all 8 experimental runs are given in Table 2. e 3D response surface plots drawn using Design Expert soware are shown in Figure  4. Based on the results of pseudoternary phase diagrams, appropriate ranges of the components were chosen. e oil phase concentration that could form microemulsion was found to be 10-70% and was selected as oil concentration to identify the optimum proportion of oil. Previous reports revealed that there was a really tight relationship between the hydration effect of the stratum corneum and the dermal permeation [9], and the thermodynamic activity of drug in microemulsions was a signi�cant driving force for the release and penetration of drug into skin [9]. Based on pseudoternary phase diagrams, the surfactant mixture (surfactant, cosurfactant, and mix 2 : 1), that could form clear microemulsion with large area was selected as variable and was found to be 25-55%. Design Expert soware was used to optimize the formulation and to develop the mathematical equations which are depicted in (5) and (6). e responses, percent drug diffusion ( 1 ) and spreadability ( 2 ) were found to be signi�cantly higher ( 1 , 93.16-83.06%; 2 , 25.87-19.54 gm⋅cm/sec) only when the oil and mix were used at 10% (v/v) and 55% (v/v) concentration level, respectively. e ranges of other responses, 1 and 2 were 68.06-93.16% and 7.76-25.87 gm⋅cm/sec, respectively. e responses of these formulations ranged from a low drug diffusion of 68.60% ( 8 , high level of oil and mix and of high level of gelling agent) to a higher penetration of 93.16% ( 3 , low level of oil, high level of mix , and low level of gelling agent). For estimation of quantitative effects of the different combination of factors and factor levels on percent drug diffusion and spreadability, the response surface models were calculated 3.8. Fitting of Data to the Model. Formulation 3 showed a signi�cantly higher amount of percent drug diffusion ( 1 ) and higher spreadability ( 2 ) among the formulations. e responses observed for 8 formulations prepared were simultaneously �t to design model, �FI and 3FI models using Design Expert 7.�.5. It was observed that the best �t model was 3FI model, and the comparative values of 2 , standard deviation, and coefficient of variation (%) are given in Table 3 along with the regression equation generated for each response. A negative value represents an effect that favors the optimization, while a positive value indicates an inverse relationship between the factor and the response. It is evident that the independent variable 3 (concentration of gelling agent) was found to have a negative effect on the responses: percent dug diffusion ( 1 ) and spreadability ( 2 ). e independent variable 2 was found to have a positive  effect on the percent dug diffusion ( 1 ) and spreadability ( 2 ). e three-dimensional response surface plots ( Figure  4) were drawn to estimate the effects of the independent variables on response and to select the optimal formulation.
3.9. Data Analysis. Formulations 3 and 5 had the higher percent drug diffusion and spreadability. e percent drug diffusion and spreadability obtained at various levels of the 3 independent variables ( 1 , 2 , and 3 ) were subjected to multiple regression to yield a second-order polynomial equation. e value of the correlation coefficient ( 2 ) of (5) was found to be 0.9985, indicating good �t ( e value of 2 of (6) was found to be 0.9740, indicating good �t (Table 5). e "Pred R-Squared" of 0.8151 is in reasonable agreement with the "Adj R-Squared" of 0.9393. e spreadability values of 3 and 5 were found to be more among the formulations. e spreadability values were found to be increased from high to low levels of 1 , low to high levels of variable 2 , and low levels of 3 . e spreadability values measured for the different formulations showed wide variation (i.e., values ranged from a minimum of 7.76 in 3.10. Validation of Response Surface Methodology. ree checkpoint formulations were obtained from the RSM, the composition, and predicted responses which are listed in Table 6. To con�rm the validity of the calculated optimal parameters and predicted responses, the optimum formulations were prepared according to the above values of the factors and subjected to ex vivo permeation studies. From the results presented in Table 5, the predicted error was below 5%, indicating that the observed responses were very close to the predicted values. Percentage prediction error is helpful in establishing the validity of generated equations and to describe the domain of applicability of RSM model. Linear correlation plots between the actual and the predicted response variables were shown in Figure 5. e linear correlation plots drawn between the predicted and experimental values demonstrated high values of 2 (percent drug diffusion, 0.9919� spreadability, 0.9231) indicating goodness of �t.
3.11. Ex Vivo Diffusion Study. e ex vivo release study of optimized emulgel (10% oil, 55% mix , and 1% gelling agent) compared with the 1% MTZ gel formulation. e optimized emulgel and MTZ gel showed the 83.14% and 34.63% release at the end of 12 h, respectively. e emulgel exhibited higher �ux and permeation coefficient as compared to the MTZ gel formulation ( Figure 6). e results showed that the MTZ emulgel has the steady state �ux ( ) 351.78 ( g/cm 2 /h) and apparent permeation coefficient ( ) 35.17 (cm/h) × 10 −3 ( Table 7). e permeability enhancement factor for emulgel when compared with marketed formulation Metrogyl was found to be 3.65.
3.12. Skin Irritation Test. e skin irritation studies were carried out to evaluate the tolerability of the emulgel components aer application. It was observed that emulgels were very well tolerated by the rabbits, and no signs of erythema and/or edema were seen even aer 3 days.
3.13. Stability Studies. Short-term accelerated stability of emulgel was found aer 3 months at 40 ○ C/75% RH and 4 ○ C. Emulgels were found to be white viscous creamy preparation with the smooth homogenous appearance which is similar to the day on which it was formulated. pH and the drug release of formulation were found to be 6.3 ± 0.4 and 82.7 ± 1.02%, respectively. us the formulations were found to be stable under accelerated conditions. ere was no evidence of syneresis in the emulgels which is a common drawback of gels.

Conclusion
e present study conclusively demonstrates that the use of a 2 3 full factorial design is valid for predicting the percent drug diffusion and spreadability in optimization of emulgel formulations. e derived polynomial equations and contour plots aid in predicting the values of selected independent variables for preparation of optimum emulgel with desired properties. e developed emulgels were efficacious for the delivery of lipophilic and poorly soluble drugs such as Metronidazole. e results demonstrated that the formulations were stable and showed improved permeation of the drug from the emulgel compared to conventional gel.