Release characteristics of gliclazide in a matrix system

Abstract

In this study, the release characteristics of gliclazide in a polymeric matrix system, which is used for controlled drug release purposes, are conducted experimentally and numerically. A code using the finite element method predicting the drug release behavior of gliclazide matrix system in an aqueous medium is developed. The parameters having significant importance in drug release kinetics, such as structure factor, the slab’s size and shape are varied systematically. The consistent reduction in the solid drug during the dissolution process is evaluated. The numerical data agree well with the experimental results. Therefore, the controlled drug release of gliclazide is accurately modeled by the present numerical code. The results imply that the porosity of the matrix system has the most significant effect on the drug dissolution rate. The reduction in the tablet’s diameter and utilization of cylindrical slab geometry increases the speed of the drug dissolution in the aqueous medium.

Introduction

For the past decades, it is reported that extending drug release in the human body has numerous benefits; such as decreased fluctuations of the active ingredient in the bloodstream, which leads to reduced adverse effects (Jaimini and Kothari 2016), lower administration frequency (Borgquist et al. 2006), increased patient compliance (Kuentz et al. 2016). Several therapeutic and chemical precautions are proposed for prolonging the efficiency of the active ingredient within the human body. For instance; absorption duration can be extended by altering the administration form of the drug (Patel et al. 2011), biotransformation can be avoided by adding enzymatic inhibitors to the drug formula (Tan et al. 2008), extraction of the active ingredient from the body can be postponed by vasoconstrictor substances (Pitt and Schindler 1995). However, the most reliable and healthy way is adjusting the pharmaceutical dosage form. For this purpose, modification of the drug release matrix is one of the most common methods, which is utilized for preoral drug administration (Frenning et al. 2005). Besides, drug particles can be coated with hydrogel membranes, which is investigated in numerous studies such as Bettini et al. (1994), Choi et al. (2009) and Lamberti et al. (2011).

The controlled drug release aims to realize drug delivery in an organized and predictable fashion over time (Thombre et al. 2010). The monolithic matrix system is one of the most utilized controlled drug delivery systems. This system is constructed from a polymer matrix, which includes compressed active ingredient together with the additives and the retarding materials. With this technique, active ingredient particles can be efficiently embedded within the retardant material's porous matrix core. The matrix’s skeleton is constructed from an insoluble polymer, and the ingredients dissolve in an aqueous or a gastric media slowly in a homogeneous matrix system without using a hydrophilic material. This type of matrix delivery system becomes popular in recent decades due to the ease of processing with low manufacturing costs (Reddy et al. 2003). Active ingredients can be released from a delivery system using three various ways, such as diffusion, degradation, and swelling followed by diffusion. Generally, the active ingredient diffusion is observed when a drug or the other active agent passes through the polymer forming the sustained release system. Therefore, the primary driving mechanism in the matrix systems is diffusion, which is affected by the porosity and availability of hydrophilic ingredients. One of the most used active ingredients in matrix systems is gliclazide, an essential medicine in the World Health Organization (WHO) (Lu et al. 2017). This drug is utilized in the treatment procedure of insulin non-dependent type II diabetes mellitus, and it is a derivative of second-generation sulfonylurea, which can be a better candidate for a prolonged sulfonylurea therapy (Ambrogi et al. 2009). Gliclazide is a well-tolerated active ingredient and has a low incidence of hypoglycemia (Lu et al. 2017). Although it is reported that gliclazide has a low dissolution profile, the controlled release of gliclazide in water and gastric fluids is desired to manage the blood glucose level in a long period of time with consistent dosage forms for diabetic patients, as well as to decrease the side effects (Ambrogi et al. 2009). Therefore, sustained drug release designs, such as the matrix systems (Lu et al. 2017; Thombre et al. 2010), hydrogel capsules (Bajpai et al. 2015) are widely utilized for controlled release purposes.

The experimental methods for the drugs' dissolution behavior are a better way to understand the drug release kinetics. However, these techniques sometimes are not financially affordable and require a dedicated laboratory with expert laborers. Therefore, numerical modeling becomes a good alternative in the absence of the experiment for the scientists and the industry to design a newly developed drug and optimize the existing product (Barba et al. 2009). Therefore, several numerical codes are proposed in the open literature to predict drug dissolution/release behavior by considering real experimental and/or working conditions. Cobby et al. (1974) propose an expression with a basic cubic form for foreseeing the results of in-vitro testing of slowly dissolved tablets in a homogeneous insoluble matrix. Their mathematical model assumes that the tablets have identical compositions with various shapes and sizes. A theoretical approach is proposed by Siepmann et al. (1999), who investigate the HPMC (hydroxypropyl methylcellulose) matrix’s drug release behavior to make quantitative predictions for sustained drug release. It is reported that a good match between experimental and theoretical results are observed. Frenning and Strømme (2003) present a mathematical model for the tablets' drug release with a polymer matrix, which considers drug dissolution, diffusion, and immobilization. Their model includes Noyes and Whitney equation for the drug dissolution and Langmuir–Freundlich adsorption isotherm for drug adsorption. Their mathematical model also has a good agreement between the results of the theory and the experiment. After this model, Frenning et al. (2005) suggest a finite element approach for the drug release of matrix systems with a cylindrical shape, whose outcomes are valid for all solubility rates dissolution for the drugs. Their model demonstrates that a finite dissolution rate is an important parameter affecting the release profile and might produce an initial delay. A two-dimensional numerical model with finite element analysis is also reported by Blagoeva and Nedev (2008) for presenting the release profile of the drug with the matrix system, who considered the finite dissolution rate. Barba et al. (2009) demonstrate a general code for calculating the corresponding parameters of the matrix systems’ drug release with various shapes. Their numerical solution is to understand the drug release profile of matrices manufactured from swellable/erodible polymers. In particular, it should be stated that solid drug dissolution problems are usually solved with moving boundaries for the previous studies (Lee et al. 1998; Lemaire et al. 2003; Pitt and Schindler 1995). On the other hand, this problem is also consistently modeled via combining modified Noyes–Whitney and diffusion equations, which does not require moving boundaries in the studies of Frenning et al. (2005) and Blagoeva and Nedev (2008).

In the present study, the drug release kinetics of gliclazide tablets made of the polymeric matrix is investigated both experimentally and numerically. A finite element model (FEM) is conducted by combining modified Noyes–Whitney and diffusion equations. Therefore, drug diffusion, drug dissolution, and increase in diffusivity are considered in our numerical model. In order to perform a parametric investigation, the critical parameters affecting the release profile of the matrix system, such as the size and shape of the tablet and the effective diffusion coefficient, are systematically varied. The data from this model is compared to the experimental results. It can be reported that the outcomes of the numerical model agree well with that of the experiment. This numerical model, which is dedicated to the release kinetics of a tablet with a polymer matrix, is unique for this active ingredient, gliclazide, in the open literature.

Materials and methods

Experimental study

Materials and suppliers

The materials used in this study, and the suppliers are indicated in Table 1.

Table 1.

Materials and suppliers in the current experimental study

Materials	Supplier
Gliclazide	Servier, Turkey
Kollidone K25	Eczacıbaşı, Turkey
Lactose	Eczacıbaşı, Turkey
Avicel pH 101	Eczacıbaşı, Turkey
Magnesium stearate	Eczacıbaşı, Turkey

Tablet preparation

During the preparation of 150 mg sustained-release gliclazide tablets, Kollidone K 25 was used as the matrix material. The ratio of the matrix material was determined by performing pre-formulation studies. The ratio of kollidone K 25 in the formulations was 21.53%. Magnesium stearate was used as the lubricant with a ratio of 2.15% in the formulations. Two filling and diluting agents were used to reach a total tablet weight of 325 mg. One of the filling and diluting agents was lactose, which is hydrophilic, and the other was Avicel pH 101, which is not hydrophilic. The ratio of lactose was 21.53%, and the ratio of Avicel pH 101 was 8.61%. The effect of filling and diluting agents on the dissolution profile of the active ingredient was evaluated.

The active ingredient and additives included in the tablet formulation were mixed using a progressive mixing technique. Tablet forms were made using the direct compression method utilizing 1.18 cm diameter flat-faced tablet punches. The tablets were characterized immediately after the formulation according to physical properties like hardness, thickness, friability and drug contents. The drug release of individual six tablets was measured using USP XXII (pedal type) apparatus (Prolabo, England USP XXII) using the medium of 900 ml pH 7.5 phosphate buffer. The dissolution media were maintained at a temperature of 37 ± 0.5 °C. The sample was withdrawn at the interval of 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17 and 18 h. At every withdrawal, the sample was replaced with 5 ml of fresh media. The sink conditions were fulfilled during the whole process. All the solutions of samples were analyzed spectrophotometrically.

Theory

The drug tablet is designed as a circular slab which has a diameter of Dia = 2.37 cm and a height of H = 0.215 cm, which is shown in Fig. 1a. However, the computational domain (Ω) is considered two-dimensional, as seen in Fig. 1b. Therefore, drug transport through the depth of the tablet is neglected. Release of dissolved drug and dissolution of solid drug is solved with diffusion equation and modified Noyes–Whitney equation in rectangular coordinates, respectively;

$$\frac{\partial c}{\partial t}=D^{\mathit{eff}}{\mathrm{\nabla }}^{2}c+k〈s^{2/3}\left(c_s-c\right)〉$$ 1

$$\frac{\partial s}{\partial t}=-k〈s^{2/3}\left(c_s-c\right)〉$$ 2

Fig. 1.

a Tablet shape and dimensions, which are used in this study, b a typical mesh structure in the present finite element solution

where $⟨⟩$ symbol represents Macaulay bracket, which corresponds to

$$〈x〉=\frac{x+\left|x\right|}{2}=\left\{\begin{array}{cc}x& ifx\ge 0\\ 0& ifx<0\end{array}\right)$$ 3

c and s are the concentrations of dissolved drug and solid drug, respectively. These parameters are normalized according to the initial total concentration, C_tot. Therefore, c = C/C_tot and s = S/S_tot. In addition, k is the dissolution rate constant and c_s is the solubility ratio of the drug. It is assumed that the drug is completely solid in the matrix. Therefore, initial conditions of the drug concentrations are determined as c = 0 and s = 1. Perfect sink condition is applied to the solving domain's outer boundaries, i.e., c = 0 for outer boundaries at all times. Because diffusion is the dominant mechanism of the matrix's dissolved regions, the drug’s diffusion coefficient is assumed constant for every condition. Moreover, fractional drug release (FDR) from the matrix is calculated according to the formula below:

$$FDR=1-\frac{1}{V}\underset{\mathrm{\Omega }}{\overset{}{\int }}\left(c+s\right)d\mathrm{\Omega }$$ 4

Due to the non-linearity nature of Eqs. (1) and (2), there is no analytical solution (Frenning and Strømme 2003). Therefore, the problem must be simplified. To this end, the dissolved drug concentration is integrated by means of time:

$$u=\underset{0}{\overset{t}{\int }}c\left(t^{\prime }\right)d{t}^{\prime }$$ 5

Then, Eq. (2) is solved:

$$s={〈s_0^{1/3}+\frac{k}{3}\left(u-c_{s}t\right)〉}^3$$ 6

Therefore, the first equation becomes:

$$\frac{\partial u}{\partial t}=D^{\mathit{eff}}{\mathrm{\nabla }}^{2}u+\left(1-{〈s_0^{1/3}+\frac{k}{3}\left(u-c_{s}t\right)〉}^3\right)$$ 7

By this means, the problem can be solved a simple non-linear partial differential equation.

Numerical study

A numerical study is performed with finite-element solver software COMSOL Multiphysics 5.3a (Burlington, MA, USA). Initially, mesh independence is investigated for the purpose of increasing the accuracy of the solutions. Simulations are consecutively carried out under the same conditions with meshes having 1212–2956–5854–11,662–23,188 elements. The relative error between 5854 and 11,662 mesh elements is observed lower than 1%. Therefore, 5854 is determined as the optimum mesh element number. The dimensions and geometry of the tablet are shown in Fig. 1a, b, respectively. The code is validated with our experimental results as demonstrated in the plots of Fig. 2. After reaching high correlation values with the experiment, some critical parameters, which are the effective diffusion coefficient, the size, and shape of the tablet, are changed to evaluate the release kinetics further. The effective diffusion coefficient is defined as $D^{\mathit{eff}}=\left(\raisebox{1ex}{$\epsilon $}\!\left/ \!\raisebox{-1ex}{$\tau $}\right.\right)D^{\mathit{liq}}=f{D}^{\mathit{liq}}$ where ε, τ, f are porosity, tortuosity and structure factor, respectively. D^liq is the actual diffusion coefficient of the material. According to this equation, there is a direct proportionality between the effective diffusion coefficient and the actual diffusion coefficient. The reader is referred to the study of Pisani (2011) for further information. In this study, the effective diffusion coefficient is changed according to f values. Moreover, the diameter of the tablet is increased, and decreased twofold, consecutively, and corresponding release profiles are calculated. Moreover, the rectangular slab and cylindrical tablet geometries' release behaviors, which have equal volumes, are compared to fundamental circular slab geometry. For cylindrical tablet geometry, governing equations are implemented in cylindrical coordinates. The key parameters, which are used in this numerical simulation, are listed in Table 2.

Fig. 2.

The comparison of the numerical simulation results in the current study with the experimental data

Table 2.

Parameters used in the present numerical study

Parameter	Definition	Value
D	Effective diffusion coefficient of solute drug	8 × 10–10 m2/s
Cs	Solubility ratio of the drug	0.4
K	Drug dissolution rate	0.4 s−1
Co	Initial dimensionless solute drug concentration	0
So	Initial dimensionless solid drug concentration	1

Results and discussions

Prior to the primary systematic investigation, the present finite element code is validated with our experimental data, which is depicted in Fig. 2. It is shown that numerical data agree well with the experimental results. This implies that our code in Comsol finite element solver makes perfect predictions. This comparison is also essential to find out the magnitudes of useful parameters, such as effective diffusion coefficient, solubility ratio, and drug dissolution rate. These parameters can be predicted using the equations, which are available in the works of Higuchi (1963) and Cussler (2009). However, these values are different from the ones, which are utilized in this work. Similar to the previous studies, these values are estimated from the curve fit with the experimental results as presented in Fig. 2. As a result, the effective diffusion coefficient, solubility ratio, drug dissolution rate in this study are estimated as D = 8 × 10^–10 m²/s, c_s = 0.4 and k = 0.4 s⁻¹, respectively. Regarding these values, a strong correlation is achieved between experimental and numerical results.

In order to perform a qualitative investigation about the matrix behavior, the solid drug concentration maps are depicted in Fig. 3. The top left-hand side map shows the case of a 100% solid drug at the initial state of the drug dissolution study, which is entirely painted red. As seen in the figure, the solid drug is dissolved by the time, and tablet size is gradually shrunk. The figure demonstrates that there is a consistent reduction in the tablet area from all directions. The solid drug almost reaches the tablet's center point after 840 min; however, a certain amount of solid drug remains. This difference can be attributed to the thickness of the tablet because a small amount of distinction is available between numerical and experimental results for the thick tablets, which is also stated in the work of Frenning et al. (2005). Solid drug is fully dissolved after 840 min in the experiment; however, the numerical simulation demonstrates that approximately 95% of the solid drug is dissolved within this duration, as seen in Fig. 2. The solid drug concentrations along a horizontal line starting from the center of the tablet, which is coincident with the diameter, are plotted versus the tablet radius. The purpose of this graph is to provide quantitative data, which supports qualitative representation in Fig. 3. According to Fig. 4, the solid drug concentration gradually diminishes as the duration of the dissolution progresses. It also does not entirely dissolve after 840 min. The tablet's diameter reduction is almost the same for all reference durations, starting from t = 120 min to t = 840 min.

Fig. 3.

Concentration maps of the solid drug at various time steps for 840 min

Fig. 4.

Concentration distribution along a horizontal line coincident with the diameter of the tablet

Figure 5 depicts the concentration distribution along the centerline of the solid drug. The structure factor, f is varied between f/4 ≤ f ≤ 2f to show the effect of the ratio of porosity and tortuosity on the gliclazide tablet's drug dissolution profile. It is seen that the drug fractional release rate greatly influences the structure factor. The release rate attenuates with decreasing the structure factor and accelerates with increasing the structure factor. The drug wholly dissolved after approximately 600 min when the structure factor is doubled. In contrast, the drug is dissolved with the ratios of 40% and 53% approximately for the cases of f/4 and f/2 after 840 min, respectively. Namely, it can be interpreted that the porosity of the tablet has a more significant effect on the fractional drug release rate than the tortuosity. As a result, our findings are in line with the corresponding theory. Figure 6 shows the influence of the tablet diameter on the drug release rate. In this systematic evaluation, the diameter of the tablet is increased and decreased twofold. Namely, the Dia = 0.59 cm, 1.185 cm, 2.37 cm, 3.55 cm and 4.74 cm are tested. When our reference diameter (Dia = 2.37 cm) is increased twice, only 70% of the drug is released after 840 min. In contrast to this situation, the drug is entirely released after 720 min, when the tablet's diameter (Dia = 1.185 cm) is reduced by twofold. The further reduction in tablet diameter (Dia = 0.59 cm) results in the tablet's quicker dissolution, approximately within 216 min. It is concluded that faster drug release can be obtained by reducing the tablet dimensions, which agrees well with the outcomes of Siepmann et al. (1999).

Fig. 5.

Variation of fractional drug release with the duration of the release for different structure factors

Fig. 6.

Variation of fractional drug release with the duration of the release for different tablet diameters

The effect of geometry on the release kinetics of the gliclazide matrix system is presented in Fig. 7. The reference geometry is circular; however, rectangular slab and cylindrical geometries are also tested along with circular slab geometry. Although a slight difference is observed for fractional drug release between the circular and rectangular slab geometries, the solid drug is wholly dissolved in 30mins for cylindrical geometry. These findings may be attributed to the increase in the effective surface area of the tablet. Please recall that these geometries have equal volumes. Due to a significant increase in surface area for cylindrical geometry, higher diffusive flux density results in faster dissolution and drug release in contrast to rectangular geometry. A similar trend for cylindrical geometry is also obtained in the study of Barba et al. (2009).

Fig. 7.

Variation of fractional drug release with the duration of the release for different tablet shape

Conclusions

The experimental and numerical investigations for the release kinetics of a monolithic matrix system, which is one of the most used techniques in controlled drug release studies, are conducted in the present work. The active ingredient in this system is gliclazide, which is the essential drug in the World Health Organization (WHO) list. The following conclusions can be drawn from this study.

1. The results of the present numerical code are in good agreement with the experimental data.

2. The diameter of solid drug consistently reduces during the 840 min. Therefore, the increase in the dissolved drug concentration in the liquid medium is stable, which indicates the occurrence of controlled drug release.

3. The systematic evaluation of the structure factor implies that the matrix system's porosity is more effective than its tortuosity on the release kinetics.

4. There is an indirect proportion between the diameter of the tablet and the speed of the drug release. Namely, small tablets dissolve faster than large tablets.There is an indirect proportion between the diameter of the tablet and the speed of the drug release. Namely, small tablets dissolve faster than large tablets.

5. Cylindrical slab geometry exhibits faster release characteristics than rectangular and circular geometries with similar trends, which is also reported in the open literature.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Compliance with ethical standards

Conflict of interest

The authors declare no conflict of interest.