Preparation and Characterization of IPN Microspheres Containing Miglitol byUsing in House Synthesised Acrylamide Grafted Locust Bean Gum

Introduction

Interpenetrating polymer network (IPN) is a polymer that consists of two or more partially interlaced networks. A delivery system may be a preferable alternative for controlled distribution of high water soluble drugs like miglitol.^1,2,3 Many of the first controlled-release systems aimed for a delivery profile that would result in a high blood level of the drug for an extended period of time.⁴ A significant number of IPN microspheres have recently been synthesized for drug delivery purposes employing various polymer combinations.

Polyvinyl alcohol (PVA)-based IPN systems have been extensively researched.⁵ A combination of PVA with a natural polymer like locust bean gum (LBG) may improve the system’s mechanical strength and stability, allowing it to satisfy the key goals of controlled release drug delivery. Locust bean gum (LBG) is a highmolecular-weight branch polysaccharide made up of a 1,4-linked b-D mannopyranose backbone with branch points attached to a-D-galactose at their 6-positions. LBG has a molecular weight range of 300,000 to 1,200,000 Da. It is not easily soluble in water and must be heated in order to dissolve.⁶ In this study, locust bean gum was grafted with Acrylamide to increase the water solubility of the resulting polymer, which was then combined with PVA for hybridization to create IPNbased microspheres. The hybridization aids in retaining the IPN matrix’s integrity, making it cross linkable to the delivery device during its GI residence.

Obesity is a global health issue that leads to diabetes, cardiovascular disease, stroke, and cancer, among other ailments. Many pharmaceutical companies have spent significant resources and time developing anti-obesity medications; nevertheless, majority of the anti-obesity drugs that have been licensed and sold to date have been discontinued due to serious side effects.

Miglitol is often administered to diabetic patients because it inhibits alpha-glucosidase in the small intestine, reducing postprandial hyperglycemia and thereby prolonging carbohydrate absorption. Miglitol is an antidiabetic medicine that has been offered in Japan, United States, Australia, France, Germany, Spain, Switzerland, and Mexico since 1996. Furthermore, based on both animal and human trials, there is accumulating evidence that miglitol has an anti-obesity impact.

A significant issue in the creation of a dosage form is to construct an optimum pharmaceutical formulation in a short amount of time with few trials. Because the development of pharmaceutical formulations is complicated, various computer-based optimization strategies based on response surface methodology (RSM), which involves the use of appropriate experimental designs and the application of polynomial equations, have become popular.

When employing nominal trials to estimate the influence of individual variables and their interactions, factorial designs, which deal with factors in all possible combinations, are the most efficient. The use of factorial design in the production of pharmaceutical formulations has aided in the comprehension of the relationship between independent factors and reactions to them. The independent factors can be controlled, but the answers are not. The procedure allows just only a small amount of trial and error.⁷

The IPN microspheres of acrylamide grafted locust bean gum (Am-g-LBG) and PVA containing Miglitol were created by emulsion crosslinking method with the help of the full factorial design, and the microspheres were tested for their drug entrapment efficiency, swelling, and Fourier transform infrared spectroscopy (FTIR) profile. To better understand the drug release mechanism, researchers conducted an in vitro drug release investigation [in acidic media (pH 1.2) and phosphate buffer (pH 6.8)] as well as kinetic modelling. Response surface plots and contour plots created by the DesignExpert programme were used to investigate the effect of all the independent variables on the dependent variables. The desirability function was utilized to optimize the response variables.

Materials and Method

Materials

Miglitol, Locust bean gum, poly vinyl alcohol (PVA), Hydrochloric acid (HCl), Light liquid paraffin, Glycine, Acetone, Span 80 and Glutaraldehyde were purchased from Yarrow chemicals. All other chemicals and reagents used were of analytical grade.

Methods

Preparation of Acrylamide grafted Locust bean gum

120 mL distilled water was used to dissolve one gram of locust bean gum (LBG) and was stirred for 30 minutes using magnetic stirrer. Then specified amount of acrylamide (Table 1) was dissolved in 30 mL of water and then mixed with LBG solution and stirred for about one hour. Then ceric ammonium nitrate (CAN) solution was added to the above solution (300 mg/150 mg in 30 mL of water). It was then irradiated at 480 W with alternate 1 min heating and 1 min cooling for specified time. After irradiation, the dispersion was allowed to reach the ambient temperature by leaving it overnight. The grafted gum was precipitated using 200 mL of acetone and washed with absolute alcohol and 100 mL of 30% aqueous ethanol. The grafted gum was then dried in hot air oven at 60° C and converted into fines for further use.^8,9 The grafting efficiency (% GE) was then calculated by using the formula:

% Grafting efficiency = [{Mass of graft co polymer} / {Mass of (Acrylamide + LBG)}] x 100

Full factorial design for the preparation of acrylamide grafted LBG

For the synthesis of acrylamide grafted LBG, a full factorial design was utilised. A factorial design is a prominent and extensively used experimental design in which different amounts of a variable factor are mixed with all other factors of all other variables in the experiment. For the optimization of acrylamide grafting onto Locust bean gum, a two-level, three-factor, full factorial design (8 batches) was used. The amount of acrylamide, ceric ammonium nitrate, and microwave irradiation time were chosen as independent variables, and the percentage of acrylamide, ceric ammonium nitrate, and microwave irradiation time were chosen as the dependent variables. The dependent variable was chosen to be grafting efficiency.¹⁰ (Table 2).

Each dependent factor was investigated on two levels: high (+1) and low (+1). (-1). Multiple linear regression analysis (MLRA) was used to create polynomial models for the dependent variable, which included interactions and quadratic factors. Factorial models were used to analyse the responses generated by the experimental designs using Design Expert software (Trial version 11.1.2.0 64-bit, Stat-Ease, Inc., Minneapolis, USA).

Preparation of Miglitol-containing Acrylamidegrafted LBG-PVA IPN microspheres

Miglitol encapsulated PVA and Am-g-LBG microspheres were created by water-in-oil. Miglitol will be introduced to the polymeric dispersion is a term used to describe a mixture of polymers (Table 3).

Under steady mechanical stirring at 500 rpm, the drugpolymer blend must be slowly emulsified with light liquid paraffin containing 1 percent (w/w) Tween-80. It will be possible to make a milk white emulsion (w/o). Glutaraldehyde (GA) (2.5 and 5 mL) containing 0.5 mL of 1 N HCl must be gently added to this emulsion, and stirring must be continued for three hours.

To remove excess liquid paraffin, unreacted GA, and surfactant, the crosslinked microspheres were filtered and washed with acetone, 0.1 M glycine solution, and water. By treating the filtrate with Fehling’s reagent, the complete elimination of unreacted GA was confirmed. A negative result indicated that there was no unreacted GA. Hardened microspheres were vacuum-dried for 24 hours at 40°C and kept in a desiccator until needed. By producing an aqueous dispersion of crushed dried particles and treating it in the same way as previously mentioned,¹¹ the lack of unreacted GA in the dried particle matrix was confirmed.

Full factorial design for the preparation of Am-gLBG-PVA IPN microspheres containing Miglito

The optimum formula for the manufacture of Amg-LBG-PVA IPN microspheres containing Miglitol was determined using a factorial design similar to that employed in the previous experiment for the optimization of acrylamide grafting onto LBG. The optimization method in this study was carried out using a two-level [High level (+1) & Low level (-1) three-factor, full factorial design (8 batches). As independent variables, we used the LBG:PVA ratio, glutaraldehyde, and drug loading percent (Table 3). % Drug entrapment efficiency, % Swelling (pH 1.2 & pH 6.8), and % Cumulative drug release at 12 hours were chosen as dependent variables (responses). Using the multiple linear regression analysis (MLRA) technique, polynomial models with interactions and quadratic terms were created for the dependent variable. The outcomes of the response generated by the experimental designs were analysed by factorial models using Design Expert software (Trial version 11.1.2.0 64- bit, Stat-Ease, Inc., Minneapolis, USA).

FTIR studies

The FTIR spectrums of Miglitol, PVA, Am-g-LBG, drugpolymer physical mixture, blank microspheres, and drug loaded IPN microspheres were carried out to confirm the formation of Am-g-LBG and compatibility of different ingredients of the IPN formulations. A small amount of each material was mixed with potassium bromide (KBr) (1% w/w sample content), taken into sample holder and scanned in the range of 600-4000 cm^-1.

Percentage yield of microspheres

The prepared microspheres were collected and weighed from different formulations. The measured weight was divided by total amount of drug and polymers which were used for the preparation of the microspheres to obtain percentage yield.^12,13

% Yield = (weight of floating microspheres/ weight of drug+ weight of polymer) x 100

Drug entrapment efficiency (% DEE)

IPN microspheres were crushed in mortar and pestle and a required amount (10 mg) was taken into 50 mL of phosphate buffer solution (pH 6.8), heated at 50 ºC for effective drug extraction. After 24 hours, the suspension was allowed for filtration and centrifugation for the removal of polymeric debris. The supernatant was then analysed with a spectrophotometer (UV-1800, Shimadzu, Japan) at λ_max of 282 nm. All samples were analysed in triplicates. The drug entrapment efficiency (%) was calculated by the formula given below:

% DEE = (Actual drug content / Theoretical drug content) x 100

Equilibrium swelling studies:

Equilibrium swelling study of IPN microspheres was done in different media. An accurately weighed amount of microspheres (W₁ ) was immersed in 50 mL buffer (pH 1.2 and pH 6.8) and allowed to swell for 24 hours at 37 ºC. The swollen microspheres were collected and the adhered liquid droplets on the surface of the particles was removed carefully with tissue paper and reweighed (W₂ ) to an accuracy of ±0.01 mg on an electronic microbalance. All the samples were analysed in triplicates. The swelling index was calculated by using the following equation:^14,15

% Swelling = (W₂ – W₁ / W₁ ) x 100

Where, W₁ and W₂ are the dry weight and swollen weight of the IPN microspheres.

In vitro drug release study

In vitro drug release was performed in triplicate in a dissolution tester equipped with eight baskets (glass jars) at the stirring speed of 50 rpm. The drug release from the IPN microspheres were investigated in acidic medium (pH 1.2) for the initial two hours, to be followed by using phosphate buffer of pH 6.8. Throughout the experiment, an accurately weighed quantity of each sample (equivalent to 100 mg Miglitol) was placed in 900 mL of dissolution medium maintained at 37.5 ºC. At regular intervals of time, sample aliquots were withdrawn and analysed using UV spectrophotometer (UV-1800, Shimadzu, Japan) at the fixed λ_max of 282 nm.

Release kinetics

To understand the mechanism of drug release, in vitro drug release data have been analysed using the empirical kinetic equations. The regression factor (R² ) of zero order, first order, Higuchi & Korsmeyer peppas plot was calculated along with the n value for Korsmeyer peppas plot.

Design-Expert software was used to do several response surface methodology simulations for the current optimization investigation (Trial version 11.1.2.0 64-bit, Stat-Ease, Inc., Minneapolis, USA). After examining the intensity of the coefficient and the mathematical sign it bears, positive or negative, the polynomial equation was employed to derive a conclusion. Synergy is indicated by a positive indication. The statistical validity of the polynomials was determined using the Design Expert software’s ANOVA. p< 0.05 was used as the threshold for significance. The Stat-Ease Design-Expert software also produced three-dimensional response surface graphs and contour plots.

Stability study

The purpose of stability testing is to provide evidence on how the quality of a drug substance or drug product varies with time under the influence of a variety of environmental factors. To assess the drug and formulation stability, stability studies were done as per ICH guidelines. The formulated IPN microspheres were wrapped in aluminium foil and stored at 45 ± 0.5ºC for a period of twelve weeks. After the period of three month, the prepared IPN microspheres were tested for drug entrapment efficiency.¹⁶

Results

Grafting efficiency of prepared Acrylamide grafted LBG (% GE)

Table 5 represents the grafting efficiency of different prepared acrylamide grafted gum. The grafting efficiency of the prepared formulations (L1-L8) ranges from 77.22 ± 0.69 to 92.96 ± 0.36. As said earlier in the methodology, factorial design was used to optimize the GE using three different variables at two levels. Their mathematical relationship with GE was generated using multiple linear regression analysis (MLRA) and was expressed as:

% GE = 84.92 + 1.85 A – 2.98 B + 3.36 C – 0.575 AB

where, A is amount of CAN (mg), B is irradiation time (min) and C is amount of Acrylamide (g). The threedimensional plot is shown in figure 1.

side chains. The irradiation time had the negative impact on GE, which was due to frequent breakage of chain under microwave irradiation.

ANOVA analysis indicated that the factorial model was significant (p = 0.0044, i.e. p < 0.05) having R² value of 0.9854. The Predicted R² of 0.9659 was in reasonable agreement with the Adjusted R2 of 0.8960; i.e. the difference was less than 0.2. (Table 6)

Adequate precision measures the signal to noise ratio of the model. A ratio greater than 4 is desirable. The ratio of the model 21.29 indicates an adequate signal. Thus, model can be used to navigate the design space.

After generating the model polynomial equations to relate the dependent and independent variables, the best optimized amount of independent variables were selected to achieve the maximum grafting efficiency.

The software generated solution shows the formulation having 288.84 mg of CAN, 2.54 minutes of irradiation time and 9.85 grams of acrylamide can satisfy the required conditions, which was in good and close agreement with the L2, which had good desirability of 0.98.

Hence, the formulation L2 was considered for the preparation of the IPN microspheres containing Miglitol.

FTIR Studies

The FTIR studies of the Miglitol, LBG, Am-g-LBG (L2), PVA, Physical mixtures (Miglitol with Am-g-LBG and Miglitol with PVA), Blank microspheres (Placebo), Optimized drug loaded microspheres (F4) was done to study the compatibility of different formulation components and to confirm the formation of IPN matrix.

Fourier transform infrared (FTIR) spectra of Miglitol was shown in the figure 2.

The FTIR spectrum of miglitol standard consists of characteristic band values at 3279 cm^-1 due to C-H stretching and 1261cm^-1due to C-O stretching. It was confirmed as miglitol.

The LBG FTIR spectra showed the strong O-H stretching peak of hydroxyl group at about 3377.47 cm^-1, and it was due to hydrogen bonding involving the hydroxyl groups on the LBG molecules. The C-H stretching vibrations were observed at 2926.11 cm^-1, and additional characteristic bands of LBG appearing at 1381.08 and 1147.68 cm^-1 were attributed to C-H and O-H bending vibrations, respectively. The broad peak at about 1031 - 1147 cm^-1 was mainly attributed to CH-O-H stretching/ bending. Marked changes were observed in spectra of Am-g-LBG (L4) compared to LBG. The bands at 1654.98 and 1560.46 cm^-1 were attributed to amide-I (C-O stretching) and amide-II (N-H bending) conferred by Acrylamide. The peak at 2800- 3510 cm⁻¹ in Am-gLBG was attributed to overlap of N–H stretching band of amide group and O-H stretching band. A shouldering at about 1452.45 cm-1 was due to the C-N stretching. Peak at 1060.88 cm-1 was due to CH-O-CH₂ group which occurs during grafting reaction between -OH group of C₂ and π bond of acrylamide. FTIR analysis indicated that LBG was successfully modified into Am-g-LBG.

In case of PVA, major peaks related to hydroxyl and acetate groups were present in FTIR spectra. The large band observed between 3650 and 3228 cm^-1 are linked to the O-H stretching of the intermolecular and intramolecular hydrogen bonds. The vibrational band observed between 2854.74 and 2924.18 cm^-1 referred to the C-H stretching from alkyl groups and the peaks between 1735.99 and 1681.98 cm^-1 were due to the stretching C=O and C-O from acetate group residues from PVA. A sharp peak obtained near 1456 cm^-1 indicated the bending vibration of CH₂ groups. The intensity of the 1735.99 - 1681.98 cm^-1 strongly suggested that it was less hydrolysed.

The placebo Am-g-LBG-PVA IPN formulation showed peaks around 1672.34 and 1624.12 cm^-1 may be due to C=O group of PVA acetate and amide-II (N-H bending) conferred by Am-g-LBG, respectively. A small absorption peak at about 1240.27 cm^-1 was observed in the spectrum which was due to ether linkage (C-O-C stretching) formed between the -OH group of PVA and Am-g-LBG by the help of glutaraldehyde. Thus, it can be said that Am-g-LBG and PVA were successfully crosslinked by GA to form the network structure.

The physical mixture showed the peaks of all components (Miglitol with PVA and Miglitol with Am-g-LBG) in its spectra without significant changes, indicating compatibility of the ingredients. The peak intensities were observed to be reduced for all components due to the dilution effect.

The surface morphology of the particles prepared by acrylamide grafted locust bean gum (Am-g-LBG) having (GA) glutaraldehyde in the lesser amount exhibited porous surface (figure 3a) but in case of particles prepared with more amount of GA (figure 3b), the absence of porous surface was observed.

The miglitol XRD report exhibited a peak at 21.220 showing its nature of crystallinity i.e., figure 4a; the microspheres loaded with drugs exhibited a peak at 21.810 though microspheres with miglitol displayed the disappearance of other peaks in the figure 4b.

Percentage yield of IPN microspheres

The percentage yield of prepared Am-g-LBG-PVA IPN microspheres containing miglitol was calculated by using the formula, as described above in the methodology. The percentage yield of IPN microspheres ranges from 95.56 (F2) to 98.51 (F5). (Table 7)

Drug entrapment efficiency (% DEE)

Entrapment of drug in any matrix system is considered as important criteria for selection of suitable batch formula as amount of drug retained in matrix indicates the overall efficiency of drug delivery system showing sustainability and ability to prolong drug availability in site of action. In the current study, the drug entrapment efficiency percentage ranged from 70.12 ± 1.13 to 90.12 ± 0.95 (Table 8).

The percentage of entrapment efficiency of Miglitol was increased with the raise in polymer concentration as shown in the present study. Drug entrapment capacity had a strong dependence on particular proportion of polymeric complex.

The three-dimensional plot of the impacts of different independent variables on % DEE is shown in figure 5.

The mathematical relationship of % DEE with the independent variables (MLRA) was generated and expressed as:

% DEE = 80.24 + 5.92 A +1.01 B + 3.09 C – 0.0963 AC

where, A is Am-g-LBG:PVA ratio, B is glutaraldehyde amount and C is % drug loading. Here we observed that all the independent variables were having impact on % DEE as was observed experimentally.

The model was shown to be significant (p=0.0001) by ANOVA, with an R2 value of 0.9994. The adjusted R2 of 0.9955 was reasonably close to the Predicted R2 of 0.9985; that is, the difference was less than 0.2. (See Table 6)

It is preferable to have a precision ratio of >4. The signal-to-noise ratio of 91.30 suggests that the signal is adequate. As a result, the current model can help you navigate the design space.

Swelling Studies

Swelling study is also considered as important criteria for selection of suitable batch formula for the optimum drug release. The % equilibrium water uptake data of the IPN microspheres (Table 9) suggests that it was dependent upon major factors like the amount of crosslinker, the polymeric mix ratio, and the amount of drug loading. The swelling in the current study spans from 165.23 1.98 to 212.52 1.25 at pH 1.2, and from 203.33 0.98 to 253.22 1.66 at pH 6.8.

Crosslinker effect: When the amount of crosslinker was increased from 2.5 mL to 5 mL in set formulation parameters, the amount of medication released reduced. In formulation F3 and F6, polymer composition (1:2) and drug loading (50%) was same but amount of crosslinker was different (2.5 mL in F3 & 5.0 mL in F6). At pH 1.2, F6 showed more swelling (169.33 ± 0.28) than F3 (165.23 ± 1.98); likewise, at pH 6.8 also, F6 showed more swelling (209.36 ± 1.96) than F3 (203.33 ± 0.98). It was due to the formation of a rigid network structure at higher cross-link density. Thus, crosslinking has an effect on water uptake equilibrium as well as in vitro release profiles.

Effect of Am-g-LBG:PVA blend ratio: At pH 1.2, when the blend ratio of IPN particles were changed from 1:2 to 1:1 in fixed crosslinker and drug loading percentage, the swelling increased from 165.23 ± 1.98 (F3, 1:2, 2.5 mL, 50%) to 175.26 ± 1.66 (F7, 1:1, 2.5 mL, 50%) and 201.22 ± 1.56 (F4, 1:2, 5 mL, 25%) to 212.52 ± 1.25 (F2, 1:1, 5 mL, 25%). At pH 6.8, the swelling increased from 306.57 ± 2.15 (F3, 1:2, 2.5 mL, 50%) to 318.65 ± 1.15 (F7, 1:1, 2.5 mL, 50%) and 299.68 ± 4.03 (F4, 1:2, 5 mL, 25%) to 308.59 ± 1.59 (F2, 1:1, 5 mL, 25%).

The enhanced water uptake values were attributable to the increased hydrophilic character of Am-g-LBG in the mix IPN hydrogel microspheres.

Impact of drug loading: Formulations having different drug loading in a fixed gum- PVA ratio and crosslinker amount, showed different extent of drug release. At pH 1.2, F3 (50% drug loading, 165.23 ± 1.98) showed less water uptake property than F8 (25% drug loading, 183.65 ± 0.98) though other parameters for both the formulations were same. Likewise at pH 6.8, F3 (50% drug loading, 203.33 ± 0.98) showed less water uptake than F8 (25% drug loading, 224.21 ± 3.11). This variation in water uptake and the rise in the drug loading reduces the swelling efficiency of the matrix. Uptake is because of the dissolution of Miglitol along with the matrix swelling. Hence, the rise in drug loading reduces the swelling efficiency of the matrix. The effect of independent variables on swelling is shown in the figure 6 & 7.

The mathematical relationship is expressed as:

Swelling % (pH 1.2) = 189.11 – 9.25 A + 5.74 B – 12.58 C + 0.5538 BC

Swelling % (pH 6.8) = 227.82 – 9.26 A + 5.29 B – 11.69 C – 0.5188 AC

where, A is Am-g-LBG:PVA ratio, B is glutaraldehyde amount and C is % drug loading.

ANOVA analysis showed that both models (Swelling at pH 1.2 & 6.8) were significant with p values 0.0295 & 0.0020 (p < 0.05) with R2 value 0.9471 & 0.9915 respectively (Table 6).

The modified R2 (0.8765, 0.9801) and the expected R2 (0.6838, 0.9395) were in reasonable agreement; the difference was less than 0.2. The signal-to-noise ratio was measured by enough precision. It was necessary to have a ratio of more than four. An appropriate signal was shown by a ratio of 10.85 (pH 1.2) to 27.71 (pH 6.8). As a result, the model can be utilized to find its way about the design area.

In vitro drug release studies

The cumulative percentage drug release vs. time plot of different batches of Miglitol loaded Am-g-LBG-PVA IPN microspheres are presented in the figure 8.

In the present study, after 12 hours drug dissolution, the drug release percentage ranged from 61.45 to 92.87 (Table 10).

The amount of crosslinker, polymeric blend ratio, and drug loading were found to influence drug release from IPN microspheres. Effect of gum: PVA blend ratio - When the Am-g-LBG: PVA ratio of IPN particles were changed from 1:2 to 1:1 in fixed crosslinker and drug loading percentage, at 12 hours, the drug release improved from 64.51% (F6, 1:2) to 79.74% (F1, 1:1) and 75.02% (F8, 1:2) to 87.00% (F5, 1:1).

This was due to the hydrophilic nature of the grafted gum, which reacted with the media and caused it to inflate and erode at a faster pace, allowing for faster drug release.

Crosslinker effect: Drug release was variable in formulations with different drug loading in a fixed gum: PVA ratio and percent drug loading. F7 (2.5 mL, 83.21% drug release) showed increased drug release property than F1 (5 mL, 79.74% drug release) though other parameters for both the formulations were same. This may be due to the higher crosslinking of the IPN matrix which prevents solvent imbibition leading to less retardant property.

Effect of drug loading: When the amount of drug loading was increased from 25% to 50% in fixed formulating parameters, the amount of drug released reduced. The polymer composition (1:1) and glutaraldehyde (5 mL) in formulations F1 and F2 were the same, however the amount of drug loading was varied. F2 (92.87 percent, 25% drug loading) had a higher drug release rate than F1 (79.74 percent , 50 percent drug loading). It is possible that this is because the concentration gradient and driving force are stronger in high-drug-load formulations, promoting faster drug release. Furthermore, a low drug load matrix would have a higher gum and polymer component to operate as a drug release barrier.

The 3-dimensional plot of the effects of different formulation variables on % CDR (at 12 hours) is shown in the figure 8. The mathematical relationship of % CDR (at 12 hours) with the independent variables was generated and expressed as:

% CDR = 76.22 – 9.48 A - 0.44 B – 3.99 C

where, A is Am-g-LBG:PVA ratio, B is glutaraldehyde amount and C is % drug loading.

ANOVA analysis indicated that the model was significant with p values 0.0100 (p < 0.05) with R2 value 0.9260. (Table 6). The Predicted R2 of 0.8704 was reasonably close to the Predicted R2 of 0.7039; that is, the difference was less than 0.2. A precision ratio of more than four was sought. The signal-to-noise ratio of 9.562 suggested that the signal was adequate. As a result, you can use this model to navigate the design space.

Drug release Kinetic Study

The in vitro release data were fitted into various empirical kinetic equations and presented in the Table 11.

After plotting zero order, first order, higuchi plots for the optimized formulation F4, it was observed that the best fit was with the Higuchi model, which suggests the release of drug from matrix was diffusion controlled. (Figure 9)

The Korsmeyer peppas equation was also used for the kinetics study, except F5, the n value ranged from 0.8473 to 0.9961, which shows that all the formulations follow Super case-II transport, whereas, F5 had n value of 1.0375, which indicated that it follows non fickian diffusion.

Optimization data analysis

The numerical optimization technique was used to obtain an optimal configuration for the formulation using the desirability approach. The procedure was tuned for the dependent (response) variables, with the goal of maximising swelling at pH 1.2 and pH 6.8, as well as drug entrapment efficiency percentage, and cumulative drug release at 12 hours stayed in the 70 to 80 percent range. The formulation F4 met almost all of the criteria established during the desirability search (Figure 10).

The low % prediction error of 0.3542 to 8.1896 indicated the high prognostic ability of the factorial model (Table 12).

Stability studies

Stability studies were conducted for the optimized formulation as per ICH guidelines for a period of 90 days which revealed that the formulation (F4) was stable. The results (Table 13) suggests that the developed IPN microspheres containing Miglitol were stable for storage for long period of time.

Conclusion

In conclusion, for a period of 90 days, stability studies were undertaken for the optimized formulation according to ICH requirements, revealing that the formulation (F4) was stable. On employing glutaraldehyde as a crosslinker, Am-g-LBG-PVA IPN microspheres containing Miglitol were successfully synthesized utilising the emulsion crosslinking method. The IPN microspheres were created using a design of experiment and optimised using response surface methods, which included both independent and variable responses. The F4 formulation was determined to have the best drug entrapment efficiency and had a desirable controlled release property with moderate pH sensitivity. When taken together, these findings imply that the currently constructed IPN microspheres containing Miglitol may be recreated with high predictability and may be advantageous to patients with hyperglycaemia. This type of formulation could also be a promising biomaterial for dealing with the major issues of controlled release of extremely water-soluble medicines.

Conflicts of Interest

None. 

Acknowledgment

The authors are thankful to Rajiv Gandhi University of Health Sciences, Karnataka, Bangalore for providing financial assistance under Advanced Research Projects (Project code 17P028, University notification number: RGU /ADV. RES/BR/001/2017-18 dated 19.04.2017) to carry out the research work.