INTRODUCTION

Today, oral extended-release (ER) and controlled-release (CR) formulations become more and more important because of their unique advantages, such as better patient compliance, less fluctuation of blood drug concentrations, reduced side effects, and increased safety (1,2). Among various ER/CR dosage systems, the most famous one is the hydrophilic matrix tablet (3–5). Hydrophilic matrix tablets are usually assembled by hydrophilic polymer and/or other excipients with homogeneous dispersion of drug. The mechanism of drug release from the matrix is mainly through the polymer swelling and dissolution (6,7). Thus, the hydrophilic polymer in the formulations is very important to control drug release (8). There are two mechanisms to affect the drug release: one is the water absorption and then drug diffusion through gel layer; the other is water absorption and then drug release by gel erosion. Water uptake is particularly important to both mechanisms. After water absorption reaches a critical value, a gel layer will be formed at the matrix surface. In the gel layer, folding polymer chains gradually open and dissolve, then the process gradually moves to the dry core of the matrix until the tablet core completely disappears. Drug solubility is an important factor to this process. For water-soluble drugs, drug release is mainly influenced by the water and drug diffusion, while for water-insoluble drugs are more likely controlled by the swelling and erosion rate of the matrices (9–11).

In addition to drug solubility, drug release can also be modified by the polymer type and polymer viscosity (4). The Stokes-Einstein equation and Wilke correlation has shown the relationship between the viscosity of polymers and the drug overall release rate (12). Polymer erosion is determined by polymer concentration at the gel layer. One important parameter is the critical concentration (C crit ). The C crit is the lowest polymer concentration in the formulations at which the polymer chains can withstand the surrounding share forces without being released (13). For a linear polymer, the C crit is determined only by the molecular weights or viscosity grades of the polymers. One of widely used hydrophilic linear polymer is polyethylene oxide (PEO) because of its good compatibility, mobility, and chemical stability (14). When PEOs contact with water, because of their high water affinity, they can form strong hydrogen bonding force with the water. With the diffusion of water into the matrix, the force becomes stronger and the molecular chains become swelling and erosion (15,16). Although many scientists have studied the extended release hydrophilic matrices, fewer results demonstrated constant zero-order release in hydrophilic matrix tablets for water-insoluble drugs (17–20).

The purpose of this study was to design a PEO binary hydrophilic matrix controlled system and investigate the key influencing factors of PEO hydrophilic matrix systems on in vitro release rate for poorly soluble drugs. PEO was chosen as the main polymer added in the hydrophilic matrix systems and other low-viscosity hydrophilic materials were also added in this system to change the drug release rate. Different parameters of these matrix systems should be characterized, including the matrix hydration, matrix swelling, erosion rate, and the rheology of hydrophilic polymer so that the controlled release mechanism could be well elucidated. These PEO binary hydrophilic systems could provide a good zero-order drug release and offer us an effective guidance to design the PEO matrix tablet formulation.

MATERIALS AND METHODS

The following materials were used in the manufacture of the matrix tablets: gliclazide (GLZ, TianJin, China), nimodipine (NIM, WuHan, China), and theophylline (TPL, WuHan, China); WSRN 301 and PEON 80 were gained from Dow Chemical Co. (Piscataway Township, NJ); polyethylene glycol 4000 (PEG 4000, TianJin, China); poloxamer 188 (F68) was provided by BASF Co. (Guangzhou, China); microcrystalline cellulose was obtained from (MCC, AnHui, China); and magnesium stearate was supplied from (MS, Shanghai, China). Other chemicals were of analytical grade.

Study of GLZ Matrix Tablets

The PEO binary hydrophilic matrix system was needed to further study the critical factors that influence the water-insoluble drug release as the zero-order kinetics. In this section, GLZ was chosen as a model drug to obtain the optimized zero-order release formulations.

Preparation of GLZ Matrix Tablets

Seven formulations of GLZ (30 mg) matrix tablets were prepared to adopt three different kinds of the small molecular material (PEON 80, F68, and PEG 4000). Table I showed the composition of the studied formulations. All the materials were blended for 10 min in a mortar with the exception of MS. After addition of MS, the mixing procedure was continued for another 5 min. Tablets (273 ± 3 mg) of each group were made by direct compression with a single-punch tablet press machine (SOP-TDP-5, Shanghai, China) using a 9-mm diameter die and manual feeding. The compression force has been measured by TBH 20 (ERWEKA, Germany) to ensure that these matrix tablets’ hardness was in 9–10 kP.

Table I Composition of the Different Formulations of GLZ Prepared
Full size table

In Vitro GLZ Release Studies

Release experiments (six tablets) were studied in an automatic paddle with USP apparatus type 2 by employing ZRCD6-B (Shanghai Huanghai Medicine Checking Instrument Co., Ltd., Shanghai, China) dissolution tester with a rotation speed of 100 rpm at 37 ± 0.5°C. The matrix tablets were locked in the basket which laid the bottom of the cup to avoid them adhering to the cup. The dissolution media (1000 ml) were phosphate-buffered solution (pH 7.4). A sample (5 ml) of the solution was withdrawn from the dissolution apparatus at specific time intervals (1, 2, 4, 6, 8, 10, and 12 h), and then analyzed for drug release at a wavelength of 226 nm using a UV-9100 (Ruili, China). The cumulative percentages of the drug released were analyzed according to zero-order kinetics, first-order kinetics, and Korsmeyer-Peppas (21) equations.

$$ \mathrm{Zero}\hbox{-} \mathrm{order}:\kern0.5em Mt/{M}_{\infty }=Kt $$
(1)
$$ \mathrm{First}\hbox{-} \mathrm{order}:\ \ln \left(1-Mt/M\infty \right) = -Kt $$
(2)
$$ \mathrm{Korsmeyer}\hbox{-} \mathrm{Peppas}:\kern0.5em Mt/M\infty = K{t}^n $$
(3)

Where M t is the fractional drug release at time t, M ∞ is the amount of drug loading at infinite time; K is a kinetic constant that measures the release rate; and n is the diffusion exponent characteristic that depends on the release mechanism and the geometry of the system. In Eq. (3), for a cylindrical matrix, 0.89 < n < 1.0 indicates a zero-order release, while 0.45 < n < 0.89 states anomalous release kinetics. The release profiles of these formulations were compared by the similarity factor f 2 (22,23). The values of f 2 were calculated by Eq. (4):

$$ {f}_2=50\times \lg \left\{{\left[1+\left(1/T\right){{\displaystyle \sum_{i=1}^T\left({\overline{x}}_{ti}-{\overline{x}}_{ri}\right)}}^2\right]}^{-1/2}\times 100\right\} $$
(4)

The similarity factor f 2 was recommended by the FDA to evaluate the similarity of the in vitro dissolution curves. Profiles were considered to be similar if 50 < f 2 < 100. And, the f 2 values were far more than 50 which indicate drug release profiles at the same level or have the same release kinetics. The closer the f 2 value is to 100, the more similar or identical the release profiles are.

Based on studying GLZ formulations, the optimized PEO binary matrix systems were achieved and then their characteristics were investigated.

Study on the Applicability of the PEO Hydrophilic Matrix Systems

In order for the applicability of these systems to be used in designing controlled drug delivery formulation, the study on the range of dose and drug solubility was performed in the following experiments. In this section, the optimized binary matrix system was chosen as the model matrix; different doses of GLZ and different solubility drug, such as NIM and TPL, were investigated.

The Range of the Dose of Water-Insoluble Drug in These Systems

According to the above experimental methods, the different doses of matrix tablets were prepared for further study. Table II shows different GLZ dose (30, 60, 90, 120 mg) formulations. The in vitro release profiles at different time points were calculated. And, zero-order, first-order, and Korsmeyer-Peppas equations were used to analyze drug release profiles. The similarity factor f 2 was used to compare the similarity of these drug release profiles.

Table II Composition of the Different Doses Formulations of GLZ Prepared
Full size table

The Range of the Solubility of the Insoluble Drug in These Systems

In this study, two groups of PEO matrix tablet formulations were investigated using the method of the GLZ preparation and the composition of this study formulation shown in Table III. To find out the range of the drug solubility, the low-solubility drug NIM (2.29 μg/ml, at 37 ± 0.1°C) (24) and high-solubility TPL (8.3 mg/ml, at 25 ± 0.1°C) (25) were chosen as model drugs.

Table III Composition of the Formulations of NIM and TPL Prepared
Full size table

The whole process of preparation of NIM matrices must be protected from light because the degradation of NIM will occur easily in the solution (26). This dissolution study was carried out in a 900-ml phosphate-buffered solution (PBS, pH 7.4) including 0.5% sodium dodecyl sulfate (SDS). And, the water bath condition was set up to 37 ± 0.5°C and the dissolution medium was sequentially stirred at 100 rpm. In pre-decided interval (1, 2, 4, 6, 8, 10, and 12 h), 5 ml of dissolution medium was taken out and then measured under a UV-9100 spectrometer at 360 nm.

In vitro dissolution tests of TPL matrices were employed in 1000-ml distilled water at 37 ± 0.5°C, with locking in the basket and the paddle speed of 100 rpm. Sample of TPL solution (5 ml) was taken out at the interval (1, 2, 4, 6, 8, 10, and 12 h). The concentration of TPL was measured under a UV-9100 spectrometer at 236 nm.

NIM and TLP release profiles were analyzed according to the zero-order kinetics, first-order kinetics, and Korsmeyer-Peppas equations. In addition, the similarity of these drug release profiles was assessed by using the similarity factor f 2.

Study of the Characteristics of this Hydrophilic Matrix System

In order to further explain the mechanism of this system control’s drug release, the water uptake, swelling rate, and erosion rate of PEO binary matrix tablets should be studied. And, the effect of the pH of dissolution medium and rotation speed of the dissolution apparatus on the PEO matrix erosion rate were studied to research the stability of these matrix systems. In addition, to find the nature of the polymer material which affects the water uptake rate, the swelling rate, and the erosion rate of matrices, the rheology of polymers in this system was also studied.

The Study of Water Uptake, Swelling, and Erosion of PEO Matrices

Each tablet which contains all excipients without drugs was placed in a beaker with 500 ml of distilled water, and then measured by using a dissolution apparatus (ZRCD6-B) with the basket method (USP Apparatus I). The paddles were rotated at 50 rpm at 37 ± 0.5°C. The tablets were taken out from the medium at different time intervals (5, 15, 30, 45, 60, 90, 120, 240, 360, and 480 min) and removed the excess liquid and then weighed (W2). And, dried in the drying oven for 48 h and weighed (W1). The percentage of the water uptake was determined by using Eq. (5):

$$ \%\ \mathrm{water}\ \mathrm{uptake} = \left({\mathrm{W}}_2\hbox{-}\ {\mathrm{W}}_1\right)/{\mathrm{W}}_1\times 100 $$
(5)

At the end of the water uptake study, the swelling index (SI) was tested and calculated according to Eq. (3), and values of SI were recorded and then plotted against time on a graph. W0 is the initial tablet weight, and the swelling profiles were plotted by measuring tablets from three different batches at 1, 2, 4, 6, 8, 10, and 12 h.

$$ \%\ \mathrm{S}\mathrm{I} = \left({\mathrm{W}}_2 - {\mathrm{W}}_0\right)/{\mathrm{W}}_0\times 100 $$
(6)

At the end of each water uptake study, the erosion index (EI) was calculated using the following Eq. (4) and then plotted against time (1, 2, 4, 6, 8, 10, 12 h) in a graph:

$$ \%\ \mathrm{E}\mathrm{I} = \left({\mathrm{W}}_0\hbox{-}\ {\mathrm{W}}_1\right)/{\mathrm{W}}_0\times 100 $$
(7)

Study of the Effect of the pH and Rotation Speed on PEO Matrix Erosion Rate

The optimized binary matrix was used to investigate the effect of the dissolution medium on the PEO matrix erosion rate. In order to simulate the in vivo environment, the PEO binary matrix was run at 37 ± 0.5°C using the following 500-ml dissolution media and residence times: pH 1.2 (0.1 N HCl, 1.5 h); pH 2.5 (phosphate buffer, 1.5 h) pH 4.5 (phosphate buffer, 1.5 h); pH 7.0 (phosphate buffer, 3 h); pH 7.4 (phosphate buffer, 1 h) (27). Rotation speed was selected at 50 rpm. To evaluate the influence of the hydrodynamic conditions on the erosion rate, basket rotation speeds at 25, 50, 75, and 100 rpm (28) were tested at the dissolution of pH 7.0 and 37 ± 0.5°C. Cumulative erosion data were analyzed according to the zero-order equation.

Study Rheology of the Matrices

As studied above, we investigated solution viscosities of several kinds of matrix system polymers to find the key influencing factors of the PEO-controlled water-insoluble drug release and guide us in finding the best combination. Flow curves and viscosities of the solution which were the matrices dissolved in releasing medium were studied by using a TA Instruments AR 2000 ex. The flow curves were obtained by the shear rate which increased from 0.1 to 120 s−1. This test was measured at 37 ± 0.5°C in order to assess the body temperature.

RESULTS

Study of GLZ Release From PEO Matrices and Release Kinetics

As can be seen in Table IV, when the proportion of WSRN 301 and low viscosity hydrophilic materials was 1:1, the perfect dug release behavior was obtained. It was clearly observed that higher correlation coefficients of zero-order, first-order, and Korsmeyer-Peppas equations (r 2 > 0.99) were acquired from the formulations G-2, G-4, and G-6 compared with formulations G-1, G-3, G-5, and G-7. And in order to obtain a good zero-order drug release behavior, the formulations G-2, G-4, and G-6 were selected as the optimized formulation to further study. Then, the similarity factor f 2 was used to compare these three optimized formulations. As a result, there was no significant difference between them (f 2  > 50). It demonstrated that these low-viscosity hydrophilic materials slightly influence the drug release rate. Table IV has shown that the n values of these optimal formulations were higher than 0.89 and their correlation coefficients (r 2) of zero-order kinetic equation were higher than 0.99, which indicated that they all acquired good controlled release.

Table IV The Kinetic Parameters of Formulations
Full size table

The Applicability of the PEO Matrix System

In order to study the range of PEO matrices’ drug loading, we studied the dose of GLZ in the matrices. And, the in vitro release profiles of 30, 60, 90, and 120-mg matrices were investigated. As it can be seen in Fig. 1, these formulations’ drug release profiles have no significant difference (f 2 > 50). The kinetic parameters of these formulations all showed a good zero-order (r 2 > 0.99).

Fig. 1
figure 1

The effects of dose of GLZ on drug release of three optimized PEO binary systems: WSRN 301 and PEON 80 (a); WSRN 301 and F68 (b); WSRN 301 and PEG 4000 (c)

Full size image

Therefore, the GLZ loading of PEO matrices was from 10.98 to 43.95%. Based on this drug loading, we prepared the 60-mg NIM and the 100-mg TPL matrix tablets to study the range of the drug solubility in these PEO matrix systems. As shown in Fig. 2, the three optimized formulations were all able to control NIM and TPL release with zero-order kinetics. In other words, these three PEO binary matrix systems can hold the water-insoluble drug solubility from several micrograms (μg/ml) to several milligrams (mg/ml). Of course, this solubility range includes GLZ (29).

Fig. 2
figure 2

The effects of different solubility of water insoluble drug on three optimized PEO binary systems release: WSRN 301 and PEON 80 (a); WSRN 301 and F68 (b); WSRN 301 and PEG 4000 (c)

Full size image

The Characteristics of These Three PEO Matrix Systems

Water Uptake Behavior

The study of the water absorption was performed to further discuss the swelling behavior and erosion rate of PEO tablets. The results (in Fig. 3) showed that the hydrophilic materials of low viscosity exhibit higher water uptake than pure PEO (WSRN 301), and this resulted in a rapid increase in the weight. In the first 90 min, these matrices showed a high water uptake rate, and there were no significant differences between the three PEO binary systems except pure WSRN 301.

Fig. 3
figure 3

Water uptake curve of the three optimized formulations and pure WSRN 301

Full size image

Swelling Behavior

As can be seen from Fig. 4, pure PEO was bigger than the swelling of the other three optimized matrix tablets. This could be explained that when the low viscosity polymer is added in the PEO matrix system, the interaction force of PEO (WSRN 301) chains would become weak after that matrix tablets dissolves. Therefore, as gel strength decreased, little swelling behavior exhibited on the surface of these PEO binary matrices.

Fig. 4
figure 4

Swelling curve of the three optimized formulations and pure WSRN 301

Full size image

Erosion Behavior

As can be seen from Fig. 5, the erosion rate of pure WSRN 301 was very slow, the dissolution amount at 12 h was only 50%, and T50 of zero-order was 21 h (in Table V), although the plot showed a zero-order kinetics (r 2 > 0.99). However, when WSRN 301 combined with hydrophilic materials of low viscosity, the dissolution curve not only had zero-order behavior but also dissolution amount at 12 h was more than 75% (in Fig. 5), and the time of matrix erosion of 50% was about 13 h. Because these three binary matrix systems had good zero-order kinetics and fast release rate, they can provide a favorable condition to control the insoluble drugs as zero-order release. We can design poorly soluble drug release behavior by preparing these binary matrices based on the characteristics of the PEO binary matrix system.

Fig. 5
figure 5

Erosion curve of the three optimizing formulations and the pure WSRN 301

Full size image
Table V The Dissolution Kinetics of PEO Binary Matrix System
Full size table

The Effect of pH and Rotation Speed on Matrix Erosion Rates

Characteristics of the PEO binary matrices showed us that the mechanism to control insoluble drug release was the erosion behavior of the binary matrices. And, the different erosion behaviors would produce a different in vitro drug release profiles. The matrix tablet erosion played a more important role than water uptake in the whole process of drug release.

Figure 6 indicated that when the pH of dissolution medium changed from 1.2 to 7.4, the matrix erosion behavior had no significant difference (r 2 > 0.99, good linear in Table VI). That is to say, these PEO binary matrix system erosion rates were not affected by pH of the dissolution medium. However, Fig. 7 ( r2 < 0.99) and Table VII (f 2 < 50) suggested that PEO binary matrix erosion was influenced by the rotation speed, and this result was in accordance with the characteristics of hydrophilic matrix tablets which several authors have examined (30,31).

Fig. 6
figure 6

The effects of different pH of dissolution medium on drug release from three optimized PEO binary systems

Full size image
Table VI The Dissolution Kinetics of PEO Binary Matrix System in Different pH and Different Rotation Speed
Full size table
Fig. 7
figure 7

The effects of different rotation speed on drug release from the three optimized PEO binary systems

Full size image
Table VII The Effect of Different Rotation Speeds on Erosion Rate of PEO Binary Systems
Full size table

The Rheology

Along with the water entering the matrix, the gel layer gradually formed. Because of the higher viscosity of the polymer, the stronger gel layer would be formed. This would prevent polymer erosion and then the drug release rate would become slow. The polymer dissolving rate was dependent on their chemical structure, molecular weight, the interactions with dissolution medium, and the gel layer strength. As a linear polymer, such as PEO, its molecular weight and viscosity had a significant effect on the drug release rate. When the pure PEO was unable to control drug release very well, the other kinds of hydrophilic materials were needed to add in prescriptions. Figure 8 shows us that even though these four pure materials all fitted the equation of the Herschel–Bulkley model (32) (SD < 20, the value of SD was usually used to evaluate the fitting degree of rheology equation and this value less than 20 was regarded as fitting very well), their viscosities was different. Table VIII suggests that adding the low-viscosity hydrophilic materials to WSRN 301 would reduce the viscosity of WSRN 301. Then, these three PEO binary matrix systems would not only exhibit three similar flow patterns but also provide a similar apparent viscosity. With the increase of shear rate, the share stress trend of these PEO binary matrices was very similar (in Fig. 9) and was changing within the same magnitude. If similar patterns and viscosities of PEO binary matrix solution were shown, the similar PEO matrix erosion and in vitro water-insoluble drug would be observed.

Fig. 8
figure 8

The rheology curve of four hydrophilic polymers, pure WSRN 301 (a); pure PEON 80 (b); pure F68 (c); pure PEG 4000 (d)

Full size image
Table VIII The Rheology of Materials
Full size table
Fig. 9
figure 9

The shear stress changes of these optimizing PEO binary systems when the shear rate from 30 (1/s) increases to 120 (1/s)

Full size image

DISCUSSION

Hydrophilic matrices are the most famous DDS for the controlled drug release. There are many factors that influence the drug release from hydrophilic matrices, including drug solubility, types of hydrophilic materials, and proportion of these hydrophilic materials in the matrix systems (33). Water-insoluble drugs (GLZ, NIM, and TPL) with different solubilities have been used to prepare the PEO binary matrix systems which include low-viscosity hydrophilic materials (PEON 80, F68, PEG 4000). The results have shown that when the drug solubility is within a certain range, the most influencing factor of the drug release rate is the viscosity of PEO binary matrix systems.

WSRN 301 has a high molecular weight and viscosity to form very strong gel layers, which would decrease the drug release rate (34,35). On the contrary, if the polymer viscosity is reduced, the strength of the gel layer will become weak and the drug release rate will become fast. Forming a gel layer includes two steps: water uptake and swelling. These two steps take place at the whole process of drug release. When water absorption and swelling reach a critical value, the matrices start to dissolve and release the drugs (36). If the water absorption and polymer swelling cooperate very well, ideal drug release kinetics will be obtained. Therefore, by changing the viscosity of binary polymer in hydrophilic matrix tablets, a constant zero-order release rate could be obtained for water-insoluble drugs.

The hydrophilic matrices, including WSRN 301 and the low-viscosity hydrophilic materials such as PEON 80, F68, and PEG 4000, produce a gel layer on the surface of the tablets. When the ratio of PEO and hydrophilic materials changes, the viscosity of the binary matrix will be changed, which would result in different erosion rates and drug release rates. Hydrophilic materials can change the drug release rate because their viscosity directly affects the strength of the gel layer and C crit . In our studies, four hydrophilic polymers have different viscosities. However, the binary matrix system containing the same proportion of WSRN 301 and these low viscosity polymers showed same pattern of rheology and similar viscosity. Therefore, viscosity of hydrophilic systems played a key role in controlling drug release rate in this binary matrix system for water-insoluble drugs, which help us to design the hydrophilic matrix formulations.

The critical concentration also plays a very important role in hydroxypropyl methylcellulose (HPMC) matrix system dissolution (13). Moreover, the drug release from the HPMC matrix system is also affected by the substituted heterogeneity of HPMC. The substituted heterogeneity facilitated hydrophobic interactions to increase viscosity and therefore changed the gel layer strength and drug release rate. Thus, both molecular weight and substituted heterogeneity should be considered in HPMC matrix formulation design, while for PEO formulation, one should only consider the molecular weight.

CONCLUSIONS

Binary PEO hydrophilic matrix systems with other hydrophilic materials were developed to zero-order controlled release of water-insoluble drug with the solubility range from 2.27 to ∼8.3 mg/ml. The controlled drug release rate was mainly determined by the viscosity of the hydrophilic systems. System viscosity is the most important factor to control drug release in the hydrophilic matrix system. Our research provides an effective guidance to choose appropriate formulations for hydrophilic matrix tablets, which brings benefit to reduce trial-and-error in formulation design.