Design, statistical optimization of Nizatidine floating tablets using natural polymer

The present research was aimed in developing gastroretentive tablets of Nizatidine, in order to increase the bioavailability of the drug. Nizatidine belongs to BCS class 3 and thus formulating into gastroretentive tablets helps to achieve a better therapeutic effect. There were no reports available on the use of Mimosa gum in the design of gastroretentive drug delivery systems. Response surface methodology was employed to optimize the formulation with suitable experimental design. The goal of the response surface methodology was to obtain a regression model and to find a suitable approximation for the true functional relationship between the response and the set of independent variables. Hence, the statistical approach like full factorial design was utilized to obtain optimized formulation with a smaller number of experiments. DSC study justified no interaction of the drug with excipients. The floating lag time was observed to be less than 20 s, total floating time was in the range of 8–24 h, hardness ranges from 4 to 5 kg/cm2, and friability was less than 1%. Dissolution data indicated that the higher viscosity of Mimosa (2%) delayed the drug release for extended period of time up to 23 h when compared to lower viscosity Mimosa (1%), which controlled the release of the drug up to 12 h only. The ‘n’ values of all the prepared formulations were found to be 0.59 to 0.81 indicating that the release mechanism followed anomalous (non-Fickian) diffusion. The optimal values of independent test variables were obtained from the overlay plots. The optimized formulation of Mimosa gum (2%) (M2%opt) contained 170 mg of polymer and 25.5 mg (15%) of sodium bicarbonate. Similarly, the optimized formulation of Mimosa (1%) (M1%opt) contained 255 mg of polymer and 34 mg (10%) of sodium bicarbonate. The results clearly indicated that the optimized formulations followed zero-order release kinetics with diffusion mechanism as per the predicted theoretical release rate confirming the suitability of the predicted theoretical release profile.


Background
The primary objective of gastroretentive drug delivery systems (GRDDS) is to ensure safety of drugs as well as patient compliance. Gastric floating tablets have less density than gastric fluid and consequently they are able to float in the stomach for prolonged period of time releasing the drug slowly at the required rate from the system which results an elevated gastric retention time.
EGFMT of Nizatidine were designed to retain the tablets in the stomach for longer periods of time and deliver Nizatidine effectively to the absorption window for maintaining the effective plasma levels for a prolonged time thereby decreasing the frequency of administration of drug. Nizatidine is a histamine H 2 receptor antagonist used to treat and prevent the recurrence of ulcers and occasional heartburn, acid indigestion, or sour stomach. It decreases the amount of acid made in the stomach.
Many naturally available polymers [8] were being studied for their future prospects in the development of prolonged active ingredient release. It was observed that there was difference in properties from one batch to the other when natural excipients were used. The observed changes were due to change in the physicochemical properties. Hence in the present investigation, the properties of Mimosa gum were evaluated and its applicability in the design of floating tablets was studied.
One of the statistical optimization techniques, response surface methodology (RSM), was utilized for the development and optimization of EGFMT. RSM was employed to observe the empirical relationship between one or more measured responses and a number of independent variables in the form of polynomial equations, mapping of the response over the experimental domain, with the ultimate goal of obtaining an optimal formulation [9][10][11].

Materials
Nizatidine was received as a gift sample from Aurobindo Pharma (Hyderabad, India). Mimosa gum was kindly provided by Govt. Co-operative stores (Mumbai). Sodium bicarbonate was provided as gift sample from Kartikeya Chemicals. (Hyderabad, India). All other chemicals and solvents were of analytical grade or highest quality and were used as such as obtained.

Physicochemical characterization of mimosa gum
Mimosa seed gum hydrates well and swells swiftly when comes in contact with water. Mimosa gum has greatest advantage of degrading into biologically acceptable molecules that are easily metabolized and removed from the body. It sustains the release of drug from the dosage form by following diffusion mechanism at higher proportions. For this reason, it was selected for the present study.
The physiochemical properties of Mimosa gum such as particle size distribution, surface characteristics, bulk density, tapped density, compressibility, moisture content, pH, volatile acidity, swelling, and water absorption properties were measured [12,13].

Preparation of tablets
All the ingredients sufficient for a batch of 100 tablets according to the formulae shown in Tables 1 and 2 were passed through #30 mesh (600 μm). Active ingredient was mixed geometrically with specified excipients in order to get a uniform blend and the produced blend was lubricated with magnesium stearate and aerosil and compressed into tablets on a 16 station rotary tablet punching machine (M/s. Cadmach Minipress Machinery Co. Pvt. Ltd, India) using 12 mm round, flat, plain punches using sufficient compression force to obtain a hardness of 4 to 5 kg/cm 2 containing 85 mg of Nizatidine per tablet.

Evaluation of tablets
In vitro floating characteristics So far, prepared tablets of Nizatidine were studied for floating lag time (FLT) and total floating time (TFT). FLT and TFT were determined for 3 tablets of each batch in the 1 L glass beaker containing 900 ml of 0.1 N HCl [14]. Swelling index The drug release from any tablet depends upon the % of intake of medium; here, the medium used was 0.1 N HCl. The medium temperature was maintained at 37 ± 0.5°C throughout the study.
Uniformity of weight test As per official pharmacopeia, 20 tablets were taken in random, studied for difference in weight both individually and in group. The mean and percent deviations were determined [15].
Hardness test The strength of each tablet was measured using tablet hardness tester (Monsanto type, MHT-20). The mean hardness was determined and expressed in kg/cm 2 . Five tablets were taken to perform the above phenomenon [16].
Friability test The friability test was carried out in Roche Friabilator (PANOMEX Inc., PX/FTA-201). The tablets equivalent to weight of 6.5 g were selected randomly and initial weight (w o ) was noted down after de-dusting and placed in a rotating drum. They were subjected to 100 falls of 6 in height (25 rpm for 4 min) [17]. The percent loss in weight (or friability) was calculated by the formula given below.
Uniformity of content test To study this, 10 tablets were taken and crushed; from this, 50 mg was taken in to the volumetric flask. The drug was extracted into 25 ml of 0.1 N HCl with vigorous shaking on a mechanical shaker for 1 h and the volume was made up to the mark with 0.1 N HCl. The solution was filtered through 0.45 μm Millipore nylon filter disc and appropriate dilutions were further made with 0.1 N HCl. The dilutions were measured for the absorbance by UV spectrophotometer (UV-1800, Shimadzu, Japan) at 325 nm against blank (0.1N HCl). Content of each individual preparations were determined and the average of 10 was calculated.
In vitro drug release studies The drug release from the prepared floating tablets was studied using USP XXIV dissolution rate test apparatus (LABINDIA). Then, 900 ml of 0.1 N HCl was used as dissolution medium maintained at a temperature of 37 ± 0.5°C and the paddle was rotated at 50 rpm. The procedure was studied and the samples were suitably diluted and the absorbance was measured by UV spectrophotometer (UV-1800, Shimadzu, Japan) at 325 nm. Drug release from commercial release formulation of Nizatidine was also studied.

Comparison of dissolution data
The differences in the rate and extent of drug release due to formulation and process variables can be studied by model independent and model dependent approaches [18][19][20][21].
Model independent approaches Model independent approaches are based on dissolution efficiency (DE) or on mean dissolution time (MDT) or on time to release certain percentage of drug like T X (time to release X% of drug), difference factor (ƒ 1 ), and similarity factor (ƒ 2 ), etc.
In the present investigation, three responses; floating lag time (Y 1 ), swelling index at first hour (Y 2 ), and time to release 100% of drug (T 100 ) (Y 3 ) were studied.
Another model independent approach is based on comparing the similarities of experimental formulations with reference formulation. Comparing the parameters obtained similar to methods proposed by Moore and Flanner which involves calculation of ƒ 1 and ƒ 2 . The ƒ 1 and ƒ 2 were calculated using the equations given below.
where n is sampling number, R j and T j are respectively % drug dissolved from reference and experimental formulations at time j [22,23].
Model dependent approaches The order of drug release from matrix systems was described by using zeroorder [24] or first -order kinetics [25,26]. The mechanism of drug release from matrix systems was studied by using Higuchi diffusion model [27] and Hixon-Crowell erosion model [28]. Korsemeyer-Peppas [29,30] support the drug release mechanism for further judgments.

Data analysis, optimization, and cross-validation of model
Data analysis DESIGN EXPERT (Stat-Ease Inc., Minneapolis, USA) software was used for analyzing the data. It selects and suggests the highest order polynomial model as a suitable model based on coefficient of determination (R 2 ) and predicted residual sum of squares (PRESS) values where the additional terms are significant. Analysis of variance (ANOVA) was performed on the suggested model for the responses Y 1 , Y 2 , and Y 3 to identify significant effect.
Multiple regression analysis was performed on the dependent variables to know the significance of the regression coefficients on the model. The models generated were used to construct contour (2D) and response surface (3D) plots for floating lag time, swelling index at first hour, and time to release 100% of drug responses of Mimosa gum (2%) and Mimosa gum (1%) based formulations to understand the main and the interaction effects of these three factors [31][32][33].
Optimization Desirability and graphical optimization technique (overlay plots) were employed to optimize the formulations with the desired responses (responses from theoretical profile values).
Optimization was performed with constraints of Y 1 Floating lag time = 9 s, Y 2 swelling index at first hour = 16%, and Y 3 time to release 100% of drug = 16.2 h, which were obtained from the theoretical profile. For finalizing the optimum formulation, targets were set for these constraints for getting respective desirability function response and overlay plots.
Cross-validation of model Optimized EGFMT of FNM opt and FNW opt were evaluated for uniformity of weight, hardness, friability, uniformity of content, in vitro floating, and in vitro dissolution. Pictures were taken for optimized formulations during in vitro floating. The ƒ 1 and ƒ 2 values were determined for optimized formulations using theoretical release profile as reference formulation.
The experimental values of the responses (floating lag time, swelling index at first hour, and time to release 100% of drug) were determined from the in vitro dissolution data of the optimized EGFMT.
The percentage relative error between predicted values and experimental values of each response was calculated using the below equation.

Drug-polymer interaction studies
Fourier transform infrared spectroscopy Fourier transform infrared spectroscopy (FTIR) spectra of samples were obtained on a Perkin Elmer 2000 FTIR system (Perkin-Elmer, Norwalk, CT) using the KBr disk method (2 mg sample in 200 mg KBr). The scanning range was 450-4000 cm −1 and the resolution were 1 cm −1 .
Differential scanning calorimetry Differential scanning calorimetry (DSC) was performed using a differential scanning calorimeter (DSC 220C, Seiko, Japan) at a heating rate of 10°C/min from 30 to 300°C in nitrogen atmosphere. X-ray diffraction studies X-ray diffraction patterns of powdered samples were recorded on a Philips powder X-ray diffractometer (with Philips, PW 1140/90 X-ray generator) using Ni-filtered, CuKα radiation, at 45 KV and 25 mA between 5 and 60°2θ values with 2°/2 cm/ 2θ chart speed.

Physicochemical characterization of mimosa gum
The physicochemical properties of gum are shown in Table 3.

Flow properties
Nizatidine showed an angle of repose value of 52.1°indicating poor flow and flow characteristics changed to excellent flow with increase in polymer content. The results of angle of repose values of all drug-polymer physical mixtures are represented in Table 4.

In vitro floating characteristics
In the present work, EGFMT were designed using hydrophilic polymer (Mimosa gum) and a gas generating agent (sodium bicarbonate). Mimosa is a low-density hydrophilic polymer, rapidly hydrates, and produces hydrogel to control the drug release. Upon contact with gastric contents, sodium bicarbonate in the tablets liberates carbon dioxide which is entrapped in hydrocolloid causes a decrease in the density and results an upward movement of the dosage form and keeps it afloat. The results of in vitro floating behavior of EGFMT are summarized in Table 5.   The floating lag time was observed to be less than 20 s for all the prepared formulations. Total floating time was observed to be in the range of 8-24 h.

Uniformity of weight, hardness, friability, and uniformity of content
The results of uniformity of weight, hardness, friability, and uniformity of content are represented in Table 5.

In vitro drug release studies
The percent of Nizatidine released data of Mimosa (2%) and Mimosa (1%) based EGFMT drug release profiles are shown respectively in Fig. 1.
The results indicated slow and controlled release of Nizatidine from Mimosa (2%) and Mimosa (1%) based EGFMT. During the first hour, the % drug released values were found to be in the range of 7-25% from the Mimosa (2%) based EGFMT and 15-34% from Mimosa (1%). The drug release was extended from 10 to 23 h for Mimosa (2%) formulations. About 100%     T 100 values were determined as model independent approaches and summarized in Table 6 and they were found to be in the range of 10-23 h and 5-12 h for Mimosa (2%) and Mimosa (1%) based formulations respectively and results are shown in Table 6.

Model dependent approaches Drug release kinetics
The zero-and first-order correlation coefficient (r) values of EGFMT are presented in Table 6. In all the cases, the appropriate correlation coefficient (r) values were in favor of zero-order release rather than first order release.

Drug release mechanisms
The correlation coefficient (r) values of Higuchi, Hixon-Crowell and Korsmeyer-Peppas models are represented in Table 7. It was found that EGFMT prepared with both the percentages of Mimosa gum showed predominating diffusion mechanism than erosion mechanism as indicated by higher correlation coefficient values of Higuchi model.
Plots of log fraction of Nizatidine released versus log time of all EGFMT were found to be linear. The 'r' values of these matrices were found to be 0.9809 to 0.9978 indicating that the release followed Korsmeyer-Peppas model also. The exponential 'n' values of all the prepared formulations were found to be 0.59 to 0.81 indicating that the release mechanism followed anomalous (non-Fickian) diffusion, i.e., the polymer swelling and polymer and drug dissolution governs the drug release from the matrix. This behavior indicating that the release of the drug depends simultaneously on the matrix swelling and diffusion phenomena.

Data analysis, optimization, and cross-validation of model Data analysis
Three responses, i.e., Y 1 (floating lag time), Y 2 (swelling index at 1 h), and Y 3 (T 100 ) were selected for statistical optimization and fitted to linear, interactive, and quadratic models. The summary of statistics was presented in Table 7 and the comparative R 2 , adjusted R 2 , predicted R 2 , PRESS, s.d., F values, and p values were determined using DESIGN EXPERT (Stat-Ease Inc., Minneapolis, USA). A suitable polynomial model for describing the data was selected based on correlation (R 2 ) and PRESS values. Response Y 1 , response Y 2 , and response Y 3 followed linear model for Mimosa (2%) based EGFMT. Quadratic models were followed by responses Y 2 and Y 3 respectively for Mimosa (1%) based EGFMT, whereas linear model was followed by Y 1 .
The results of the second-order response surface model fitting in the form of ANOVA are given in Table 8 respectively for Mimosa (2%) and (1%) based formulations. These parameters were used to construct the independent variables on the responses.
The detailed summary of results of multiple regression analysis of dependant variables for both polymer grades is shown in Table 10.
The contour plots were built to evaluate the relationship between polymer content and % of sodium bicarbonate and their effect on formulation parameters such as FLT, SWI, and T 100 for both Mimosa 2% and Mimosa 1% based EGFM T (Figs. 2, 3, and 4). Similarly, response surface plots were generated to determine the role of effect of polymer content and % of sodium bicarbonate on FLT, SWI, and T 100 for both Mimosa 2% and 1% based EGFMT (Figs. 3 and 5).

Optimization
The higher desirability value indicates the more suitability of the formulation and the optimized formula can directly be obtained from the desirability function response surface plots and (or) overlay plots. The desirability function (as shown in Fig. 6) was found to be higher (near to 0.9) for the optimized formula indicating the suitability of the formulations. The optimal values of independent test variables were obtained from the overlay plots (Fig. 7).

Cross-validation of model
The model predicted that the formulation with floating lag time 9.1 s, swelling index at 1 h is 16.77% and T 100 in 12 h can be obtained using the above optimum concentrations. Hence, formulations were prepared with the above optimized concentrations of polymer and sodium bicarbonate with other ingredients viz. Aerosil and magnesium stearate. The prepared optimized EGFMT fulfilled all the evaluation tests described and the results are shown in Table 11. The floating lag time for M(2%)opt was found to be 9.1 s and that of M(1%)opt was found to be 10.3 s. Both the optimized formulations floated up to 14 h and 13 h respectively for M2%opt and M1% opt .
The dissolution data of optimized EGFMT is represented and comparative dissolution profiles of the optimized EGFMT and theoretical release profile is shown in Fig. 8.
The correlation coefficient values of release order kinetics and release mechanism models along with ƒ 1 and ƒ 2 values are presented in Table 12.  Upon comparison of the observed responses with that of the anticipated responses, the prediction error was lower than 5.0% (Table 13).
The FTIR spectrum of Mimosa (2%) showed hydroxyl stretching at 3440.82 cm −1 , C-O-C asymmetric stretching at 1289.33 cm −1, and C-O-C symmetric stretching at 1103.89 cm −1 .  The FTIR spectrum of M2% opt showed all the characteristic peaks of Nizatidine with minor shifts indicating the undisturbed drug in the formulation. This spectrum showed alcoholic -OH stretch at 3504. 37  DSC analysis The DSC thermograms of pure drug Nizatidine, pure polymer Mimosa gum, and optimized formulations M2% opt and M1% opt are shown in Fig. 10.
Nizatidine showed a single sharp endothermic peak at 170.08°C corresponding to the melting range of Nizatidine. Mimosa gum showed broad endothermic peaks at 77.35°C and 73.43°C respectively. Nizatidine melting peak was slightly shifted to left for M2% opt and M1% opt at 168.64°C and to 165.71°C respectively.
XRD studies X-ray diffractograms of pure drug Nizatidine, polymer Mimosa gum, and their optimized formulations M2% opt and M1% opt were shown in Fig. 11.
X-ray diffraction patterns revealed that pure Nizatidine was clearly in crystalline state as it showed sharp distinct peaks notably at 2θ diffraction angles of 5. 8 M1% opt formulation showed characteristic peaks of pure drug, Nizatidine without shift at 18.1, 21.0, 22.4, and 26.6°(2θ). One peak disappeared at 17.5°(2θ) and some peaks showed lower intensity or shifted slightly.

Discussion
The values of angle of repose [34], bulk density, and compressibility index indicated that the Mimosa gum powder has good flow properties and compressibility.  The optimized formulation of Mimosa gum (2%) (M2%opt) contained 170 mg of polymer and 25.5 mg (15%) of sodium bicarbonate. Similarly, the optimized formulation of Mimosa (1%) (M1%opt) contained 255 mg of polymer and 34 mg (10%) of sodium bicarbonate.
The predicted formulations were prepared and compared their dissolution profile with the theoretical profile. The optimized formulations were very close to '0' (< 2) and ƒ 2 values were more than '50' (> 90) indicating the similarity between the optimized formulations and theoretical profile. The results clearly indicated that the optimized formulations followed zero order release kinetics with diffusion mechanism as per the predicted theoretical release rate confirming the suitability of the predicted theoretical release profile.
Lower values of the relative error indicated that there was a close agreement of experimental values with predicted values for both the polymers. This proved the predictability and validity of model and ascertained the effects of polymer and the amount of sodium bicarbonate on drug release.
The FTIR spectra of M2% opt and M1% opt showed all the characteristic peaks of Nizatidine confirms the undisturbed drug in the formulation.
Compared to pure drug, the melting peak was broadened to some extent in the formulations which may be due to changes in crystalline form. In addition, the studied polymers were hydrophilic in nature with melting points less than that of Nizatidine. The low melting point of the polymers might have influenced the shift in the melting point of drug in the formulation.

Conclusion
GRDDS of Nizatidine was prepared using Mimosa gum 1% and 2% as rate retarding polymer. The results clearly indicated that the optimized formulations followed zeroorder release kinetics with diffusion mechanism as per the predicted theoretical release rate confirming the suitability of the predicted theoretical release profile.