Microenvironmental pH and microviscosity inside pH-controlled matrix tablets: An EPR imaging study

Journal of Controlled Release 112 (2006) 72 – 78 www.elsevier.com/locate/jconrel

Microenvironmental pH and microviscosity inside pH-controlled matrix tablets: An EPR imaging study Stefanie Siepe a,c,⁎, Werner Herrmann b , Hans-Hubert Borchert b , Barbara Lueckel a , Andrea Kramer a , Angelika Ries a , Robert Gurny c a Novartis Pharma AG, Technical R and D, CH-4002 Basel, Switzerland Institute of Pharmacy, Free University of Berlin, Kelchstrasse 31, 12169 Berlin, Germany University of Geneva, School of Pharmaceutical Sciences, Ecole de Pharmacie, Geneva-Lausanne (EPGL), CH-1211 Geneva 4, Switzerland b

c

Received 5 September 2005; accepted 22 December 2005 Available online 14 February 2006

Abstract Incorporation of pH modifiers is a commonly used strategy to enhance the dissolution rate of weakly basic drugs from sustained release solid dosage forms. Electron paramagnetic resonance imaging (EPRI) was applied to spatially monitor pHM and the rotational correlation time (τR), a parameter which is closely related to the surrounding microviscosity inside HPMC (hydroxypropylmethylcellulose) matrix tablets. Fumaric, citric, and succinic acid were employed as pH modifiers. 4-(methylamino)-2-ethyl-5,5-dimethyl-4-pyridine-2-yl-2,5-dihydro-1Himidazole-1-oxyl (MEP) was used as spin label. Fumaric and citric acid reduced the pHM to equal extents in the initial phase. With the progress of hydration, the more soluble citric acid diffused out from the tablet resulting in an increase in pHM, originating at the outer layers. In contrast, fumaric acid maintained a constantly reduced pHM inside the entire tablet. Due to its lower acidic strength, succinic acid did not reduce the pHM as effectively as the other pH modifiers used. The more water-soluble acids stimulated the water penetration into the matrix system, thereby rapidly decreasing τR. Once the matrix tablets were hydrated, the included pH modifiers influenced τR insignificantly. EPRI, a novel approach for monitoring pHM and τR non-invasively and spatially resolved, was used successfully for the optimization of an pHcontrolled formulation. © 2006 Elsevier B.V. All rights reserved. Keywords: EPR imaging; Microenvironmental pH; Microviscosity; HPMC; Controlled release

1. Introduction In pharmaceutical technology the incorporation of pH modifiers is a commonly used strategy to alter the microenvironmental pH (pHM) within, and in the close vicinity of the solid dosage form. An optimized pH can be used to modulate the release rate of drug compounds exhibiting pH-dependent solubility [1] and to overcome stability issues of pH-sensitive drug compounds [2,3]. Several researchers have successfully enhanced the release of weakly basic drug compounds from swellable tablets using hydrophilic polymers by incorporating pH modifiers, such as succinic, fumaric, or adipic acid [4–7].

⁎ Corresponding author. Tel.: +41 61 324 6169; fax: +41 61 324 4075. E-mail address: [email protected] (S. Siepe). 0168-3659/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.jconrel.2005.12.021

They act principally by reducing the pHM, and thereby enhancing the drug solubility and dissolution. To date, only limited data concerning the extent and duration of the pH modifying effect on the microenvironment are available, and the application of this technology has been based on empirical experiences. Streubel et al. [4] optically assessed the pH M inside matrix tablets by incorporating the pH-sensitive indicator methylorange, however, exact quantification was missing. The assessment of the pHM by potentiometry using micro-pH electrodes is destructive and lacks spatial resolution [8]. Cope et al. [8] determined the pHM in hydrated immediate release pellets containing fumaric and tartaric acid by using confocal laser scanning microscopy (CLSM), a high resolution and noninvasive technique to monitor pHM continuously and spatially resolved. Spectroscopic techniques such as nuclear

S. Siepe et al. / Journal of Controlled Release 112 (2006) 72–78 Table 1 Physicochemical properties of some pH modifiers

2.2. Methods

Acid

pKa1

Solubility a (pH 6.8) [mg/ml]

Solubility (0.1N HCl) [mg/ml]

Fumaric acid Citric acid Succinic acid

3.0 3.1 4.2

10.0 651.9 72.5

4.5 608.8 66.6

a

73

Phosphate buffer.

magnetic resonance (NMR) and electron paramagnetic resonance (EPR) have also been shown to be promising tools for evaluating microenvironmental conditions [9–11]. EPR imaging (EPRI) allows obtaining unique spatial information concerning the immediate vicinity of a drug compound to which the drug is exposed before release. This in turn allows, the pHM, micropolarity, and microviscosity to be determined. The EPR technique is based on the spectral sensitivity towards the surroundings. Protonation of pH-sensitive spin labels leads to a change in the spin density of the nitroxide group [12]. Consequently, the hyperfine splitting constant and the g factor are changed. However, besides modulating the pHM, pH modifiers influence the matrix system in several other ways. Pillay and Fassihi [13] demonstrated that the addition of electrolytes altered the swelling behavior and resulted in textural variations in the swollen matrix. Varma et al. [14] confirmed decreased swelling in presence of fumaric acid as pH modifier. We assume that compositional changes of the matrix tablets influence water penetration kinetics from the edges to the center of the tablets based on hydroxypropylmethylcellulose (HPMC), thus changing the microviscosity inside the gel layer. The principal goal of this study was to further understand pH-controlled HPMC-based matrix systems. We applied EPR imaging (EPRI) to investigate the impact of incorporated pH modifiers on the pHM and the rotational correlation time (τR) inside HPMC-based matrix tablets. Different types of pH modifiers, i.e., fumaric (FA), succinic (SA), and citric acid (CA), were selected on the basis of their acidic strength and aqueous solubility (Table 1). 2. Materials and methods 2.1. Materials The spin label 4-(methylamino)-2-ethyl-5,5-dimethyl-4-pyridine-2-yl-2,5-dihydro-1H-imidazol-1-oxyl (MEP) was purchased from Magnettech GmbH, Berlin, Germany. Dipyridamole (DP) was obtained from Chemgo Organica AG, Basel, Switzerland. Methocel K100LV and Methocel K4M (HPMC) were obtained from Dow Chemical Company, Michigan, USA. Fumaric (FA) and succinic (SA) acids were supplied by Fluka, Switzerland, and citric acid (CA) was purchased from Roche Vitamins, Europe AG, Switzerland. Milled lactose by Meggle J.A., Reitmehring, Germany, magnesium stearate from FACI SRL, Carasco, Italy, Aerosil 200 from Cabot Rheinfelden GmbH, Germany, and sodium dodecyl sulfate (SDS) from Fluka, Switzerland were used as received.

2.2.1. Preparation of matrix tablets Matrix tablets were prepared by wet granulation (granulation fluid: 90% ethanol, 10% water) and comprised 15% Methocel K100LV, 15% Methocel K4M, 10% DP, milled lactose, and 20% pH modifier, respectively. The outer phase consisted of 1.0% magnesium stearate and 1.5% Aerosil 200 as lubricant and glidant. After drying, granules were passed through an 800 μm sieved and fed manually into the die of a single-punch tabletting machine with 10 mm, flat faced punches (EK0, Korsch, Berlin, Germany). The weight of each tablet was 250 ± 5 mg and the hardness of tablets was constantly adjusted to 70 ± 5 N. 2.2.2. Preparation of matrix tablets for EPR imaging experiments The nitroxyl radical MEP was dissolved directly in the granulation fluid (90% ethanol, 10% water), thus achieving a final concentration of 5 mmol/kg granules. Tablets were manufactured as described above. 2.2.3. Dissolution study Matrix tablets were incubated in phosphate buffer pH 6.8 (containing SDS 0.1%) at 37 °C using USP 1 apparatus (A7, Sotax, Switzerland) with a rotation speed of 100 rpm. At predetermined time intervals, samples were withdrawn, filtered through 0.45 μm membranes and DP was analyzed spectrophotometrically at a wavelength of 410 nm (5L4-SP1, Lambda 20, Perkin Elmer). 2.2.4. Sample incubation for EPR imaging We limited the axial swelling of the HPMC-based matrix tablets by sealing the curved surface with silicon rubber. At predetermined time intervals matrix tablets were withdrawn from the incubation medium, handled without mechanical damage, placed on a Teflon plate, and EPR spectra were recorded. For the pH calibration measurements matrix tablets containing the same formulation, but without drug and pH modifier (replaced by milled lactose), were prepared — thus, guaranteeing similar conditions for the spin label's surrounding. The calibration samples were incubated for 2 h in different buffer solutions in a pH range from 1.2 to 7.9 (sodium citrate– HCl buffer and Sorensen phosphate buffer [15]) using the described conditions. 2.2.5. Experimental setting for EPRI EPR measurements were carried out using an ERS 230 (ZWG, Berlin-Adlershof, Germany), equipped with a 1.5 GHz L-band bridge and tomography extension (both Magnettech GmbH, Berlin, Germany), in a reentrant cavity. The cavity as described in Refs. [16,17] was applied. The bore (20 mm diameter) of the cavity is large enough to allow the insertion of a Teflon plate (12 × 12 mm width) with the sample sitting on top of it. We used the following typical parameters: modulation frequency 100 kHz, microwave power 25 mW, modulation amplitude 0.25 mT, scan width 8 mT, maximum gradient 3.3 T/m, scan time per projection 20 s, points per projection

74

S. Siepe et al. / Journal of Controlled Release 112 (2006) 72–78

512, 51 projections/8 missing projections, image reconstruction (filtered backprojection) giving an image matrix of 256 ⁎ 256 points. 2.2.6. Determination of microviscosity MEP was dissolved in water and mixed with different amounts of glycerol to achieve a final concentration of 2.5 mmol/l. cw EPR spectra of these solutions, placed in Eppendorf vessels, were recorded. The rotational correlation times (τR) of these solutions were calculated by applying an algorithm as described in Ref. [18]. The Budil algorithm for determining the microviscosity has been used assuming no orientation of the spin probe. The respective computer program is based on non-linear least square analysis using a modified Levenberg–Marquart algorithm to fit the experimental EPR spectra. EPR tomograms of the hydrated matrix tablets were split into single spectra and the first derivative of the spectra was utilized for calculating τR. The experimental setup, the approach to the fitting, and the evaluation of the results follows the description given in Ref. [19]. 3. Results and discussion 3.1. Measurements for pH calibration After incubating for 2 h, the control tablets showed the typical three-line EPR spectra of almost equal height and symmetry which can be ascribed to complete mobility of the spin label. Protonation of the 14N atom in position 3 (Fig. 1) resulted in pH-related changes in the spin density of the nitroxide group [20]. The hyperfine splitting constant aN of the spin label MEP was strongly dependent on the pH of the surrounding medium. We used a sigmoidal Boltzmann fit to the values of the hyperfine splitting constant to establish the calibration curve (Fig. 1) and, thereby, obtained a pKa of 4.13 (r2 = 0.9995).

Fig. 1. pH-dependence of the hyperfine splitting constant aN of the pH-sensitive nitroxide spin label MEP measured by EPR imaging (lines: sigmoidal Boltzmann fit).

Due to the limited number of points per spectrum (256 points) and, hence, the even smaller number of points representing the spectral changes between the protonated and deprotonated form of the spin label, only ten discrete values (10 points) of the pH values in the given range were possible. Moreover, as it can be seen from the course of the calibration curve, the difference between these values depended on the slope of the calibration curve, i.e., we obtained a higher resolution in the central part, whereas a change of one point represents a larger pH change towards both ends of the calibration curve. 3.2. Determination of pHM inside pH-controlled matrix tablets Most EPR studies in pharmaceutics have been performed using X-band EPR spectrometers (9–10 GHz) [10,11]. However, these experiments were limited to small samples (b 1 mm) and to early stages of hydration when the water content was limited since high non-resonant dielectric loss is caused by water in the samples [21,22]. Lowering the frequency (L-band) decreases the sensitivity, but increases the penetration depth of microwaves into water containing specimen, and most important, reduces the dielectric losses. The loss in sensitivity can be compensated by increasing the sample size. HPMC is a hydrophilic polymer with high swellability and creates a gel layer with large numbers of water-filled pores. Consequently, we applied L-band EPR spectroscopy (1.5 GHz) to assess, spatially and non-destructively, the effect of the pH modifiers on the pHM inside hydrated HPMC-based matrix tablets. The spin label, MEP, allowed accurate determination of pH values in the range from pH 3 to 5.5. As expected, the pH of control tablets, without pH modifiers, adapted to the pH of the surrounding dissolution medium, pH N 5.5 (data not shown). These pH values could not be quantified as MEP was not suitable for determining pH values above pH 5.5. Fig. 2 shows typical pHM profiles inside matrix tablets containing FA or CA (acid concentrations: 20% w / w) after incubating for 1, 2, and 3 h (phosphate buffer pH 6.8). After hydrating for 60 min water did not penetrate to the center of the tablet, therefore pHM could not be determined in the core regions. The incorporation of CA caused a faster decrease of the pHM values as compared to FA; this can be explained by the rapid dissolution of CA inside the matrix system (Fig. 2a). During the course of incubation, as dissolution medium penetrated into the tablet, more FA dissolved inside the gelling system, and pHM decreased. FA and CA established a similar pHM inside the matrices after 2 h of incubation (Fig. 2b). This suggests that pH modifiers with comparable acidic strength altered pHM to similar extents. In the case of FA, the pH of the entire hydrated tablet remained below 3.5 even after 3 h (Fig. 2c). Because of its low solubility, FA remained undissolved in the matrix tablet and replenished acid lost by diffusion in the outer layers of the tablet. Hence, a low and constant pHM was sustained during the investigated incubation period. In contrast, CA diffused out more rapidly as compared to FA, particularly at the margins, resulting in increased pHM values. On further incubation, the pHM would gradually increase, finally reaching

S. Siepe et al. / Journal of Controlled Release 112 (2006) 72–78

75

EPRI enabled pH measurements inside HPMC-based matrix tablets up to an incubation time of 3–4 h depending on the type of pH modifier used. Within this time-frame we obtained acceptable signal-to-noise ratios. Our results are consistent with the work of Cope et al. [8] investigating immediate release pellets containing FA and tartaric acid by CLSM. This group demonstrated that FA maintained a low and constant pH over a longer period of time as compared to tartaric acid, a result which was attributed to its low aqueous solubility. In summary, pH modifiers possessing similar acidic strength resulted in a similar reduction of the pHM. Furthermore, acids with low aqueous solubility maintain a low pHM inside the hydrated solid dosage form over an extended dissolution period. The incorporation of FA (pKa1 = 3.0) had a more pronounced effect on pHM as compared to SA (pKa1 = 4.2) due to the higher acidic strength of the former. Matrix tablets containing SA revealed higher pHM values in the core (0.4 pH units) and even more increased ones at the edges (0.7 pH units) after 3 h of incubation as compared to the matrix tablets containing FA (Fig. 3). Again, the increase of the pHM in the outer parts of the matrix tablet can be attributed to leaching out of the acid. The tablets containing SA exhibited higher pHM values inside the complete matrices during the entire incubation (data not shown). In summary, the difference in the extent and duration of pH modulation depended on the physicochemical properties of the included pH modifiers, i.e., the acidic strength and the aqueous solubility. The low pKa1 values and poor watersolubility of FA led to a significant and extended effect on pHM modification. 3.3. Impact of pH modifiers on the release of dipyridamole DP displays a distinct pH-dependent solubility in the pH environments along the gastrointestinal tract, i.e., its solubility is 29.92 mg/ml at pH 1.2 and 0.005 mg/ml [23] at pH 7. In other words, DP exhibits a 6000-fold solubility difference in this pH range. Only 15.6% dipyridamole dissolved at pH 6.8 after

Fig. 2. pH profile within HPMC-based matrix tablets containing (●) FA and (○) CA (20% w / w) after (a) 60 min, (b) 120 min, and (c) 180 min (phosphate buffer pH 6.8, SDS 0.1%).

the pH of the surrounding medium. As MEP also diffused from the matrix system, the concentration of MEP inside the matrix tablets was reduced in the course of incubation, and consequently the resolution of EPRI was reduced. Therefore,

Fig. 3. pH gradients within HPMC-based matrix tablets containing (●) FA and (△) SA (20% w / w) (incubated for 3 h in phosphate buffer pH 6.8, SDS 0.1%).

76

S. Siepe et al. / Journal of Controlled Release 112 (2006) 72–78

6 h due to its poor solubility in higher pH media, whereas dissolution was complete (95.2%) after 4 h at pH 2 (data not shown). The incorporation of acidic pH modifiers significantly enhanced the drug release by creating a more acidic microenvironment, thereby resulting in enhanced solubility and, consequently, increased dissolution rates. FA enhanced drug release to the greatest extent, followed by CA and SA (acid content: 20% w / w) (Fig. 4). The difference in the observed dissolution rates can be evidently explained by the pHM profile as investigated by EPRI. The maintenance of a constant and low acidic microenvironment by FA created the most favorable conditions for DP release. In contrast, SA, the weakest acid, exhibited the least pronounced effect on drug release. Although in presence of CA a fast increase of pHM was expected beginning at the edges and afterwards dominating the entire tablet, the difference in drug release as compared to FA was small. Espinoza et al. [24] postulated that besides the pHM effect, pore-formation derived from rapidly dissolving pH modulators may additionally influence drug release by altering the gel layer properties. This could be a possible explanation for the relatively enhanced drug release in presence of CA. 3.4. Determination of microviscosity The influence of viscosity on the tumbling behavior of MEP was investigated by dissolving MEP in different water/glycerol mixtures (Fig. 5). Paramagnetic species tumble freely in low viscous media, and thereby simplified highly symmetric spectra with three narrow lines are obtained (Fig. 5a). Increasing the glycerol content raised the viscosity and significantly decreased the molecular tumbling rate of MEP. Since the anisotropic interactions are only partially averaged out, we observed line broadening and reduced signal amplitude (Fig. 5c,d). Spectral anisotropy increases as the viscosity of the surrounding increases; the most striking expression of this change is a decrease of the signal amplitude of low and high field peaks. The rotational correlation time (τR) is dependent on the viscosity of the microenvironment and the molecular size of

Fig. 4. Effect of incorporated pH modifiers (20% w / w) on dipyridamole release (phosphate buffer pH 6.8, SDS 0.1%), (n = 3).

Fig. 5. L-band EPR spectra of MEP dissolved in (a) water; (b) 50 / 50 (v / v) glycerol / water; (c) 60 / 40 (v / v) glycerol / water; (d) 80 / 20 (v / v) glycerol / water; (e) 92 / 8 (v / v) glycerol / water; (f) 96 / 4 (v / v) glycerol / water.

the spin label, represents the spin label's mobility in a specific system [25]. Accordingly, the higher the τR the lower the viscosity in the immediate vicinity of the spin label. The changes of τR inside HPMC-based matrix tablets were assessed by employing Budil's algorithm [18]. Fig. 6 illustrates the τR profiles inside acid-free matrix tablets after different incubation times. Upon exposure to aqueous media, water penetrated into the matrix tablet, and hydrated HPMC polymer particles, which began to swell, coalesce, and form a viscous gel from the outer boundary to the center. Therefore, water penetration determined the kinetics of gel layer formation of the HPMC polymer. As MEP is readily water-soluble, the spin label was expected to dissolve immediately in the infiltrated water resulting in altered EPR spectra. Subsequent to formation of the gel layer and dissolution of the paramagnetic species, the mobility of the spin label was increased and was dependent on the microviscosity of the HPMC layer only. This was particularly evident at the margins of the tablet as τR instantaneously

Fig. 6. τR gradients within HPMC-based matrix tablets without pH control (phosphate buffer pH 6.8, SDS 0.1%) after (●) 30 min, (□) 60 min, (○) 120 min, (▾) 180 min, and ( ) 240 min incubation. Comparison of the EPR spectra between the edge and the center of the tablets after 30 and 240 min.

▪

S. Siepe et al. / Journal of Controlled Release 112 (2006) 72–78

decreased to a constant low level. At early incubation times of 30 and 60 min the immobilization of MEP progressively increased to the center of the tablet. After an incubation of 4 h, τR was constantly decreased inside the entire tablet, therefore,

77

we assumed that full hydration of the matrix tablet was achieved. Additionally, the altered microviscosity was visually assessed by investigating the spectral changes (Fig. 6). After 30 min of incubation, the edges of the matrix tablets illustrated complete mobility of the spin label, as shown by the highly symmetric three-line EPR spectra. Proceeding to the center, the tumbling rate decreased, and consequently the EPR spectrum was broadened with decreased signal amplitude. After further hydration, as more water entered the core of the tablet matrix, the spectrum of MEP expressed high mobility in the center comparable to the edges of the tablet. 3.5. Impact of pH modifiers on microviscosity To evaluate the impact of the incorporated pH modifiers on the microviscosity, the τR profile was assessed inside matrix tablets comprising the selected pH modifiers at different incubation times (Fig. 7). HPMC is a non-ionic matrix forming polymer and has strong affinity towards water. Its hydration and swelling kinetics are not influenced by the pH of the microenvironment [14]. Due to fast hydration, τR decreased rapidly and independently of the nature of the included pH modifiers at the outer boundaries. The τR values at the outer layers of the tablet were similar to the mobility of the spin label in pure water, indicating complete tumbling of the spin label. This can be explained by the rapid dissolution of the spin label once the water-filled pores inside the swollen tablet matrix have been created. At early incubation times, pH modifiers significantly influenced the τR profile of the matrix tablets particularly in the central parts. We observed notably increased τR values in the presence of FA (Fig. 7a). The reduced mobility of MEP was attributed to the low aqueous solubility of FA as compared to the other acids, thus, resulting in a retardation of the water penetration towards the center. In contrast, matrix tablets with more soluble acids, like CA and SA, exhibited more reduced τR values. Acids with a high water-solubility stimulated water diffusivity towards the tablet's center with further hydration; the mobility of MEP was increased in the inner parts of the tablet (Fig. 7b and c). However, tablets comprising FA still exhibit enhanced τR values. That means that even after 3 h of incubation the spin label was not completely tumbling in the central parts. This study confirms that the enhanced release of weakly basic drugs by incorporated pH modifiers occurs mainly through modulation of the pHM. Alteration of the water penetration, observed in the EPRI study, reduced gel thickness, a delay of the initial gel formation in presence of FA [14], and pore-formation by weak acids as discussed by Espinoza et al. [24] only slightly affect the drug release. FA achieved the most pronounced effect on reducing the pHM and consequently resulted in a distinctive increased in the drug release.

Fig. 7. Effect of incorporated pH modifiers and hydration time on the distribution of τR inside HPMC-based matrix tablets after (●) FA, (△) SA, (○) CA, (▴) without acid (phosphate buffer pH 6.8, SDS 0.1%), (a) 30 min, (b) 120 min, and (c) 180 min.

4. Conclusions L-band electron paramagnetic resonance imaging (EPRI) is a valuable technique for spatial and non-invasive monitor

78

S. Siepe et al. / Journal of Controlled Release 112 (2006) 72–78

monitoring of the microenvironmental pH (pHM) and the rotational correlation time (τR) related to the microviscosity. We applied this technique to assess these parameters inside hydrated pH-controlled matrix tablets, based on hydroxypropyl methylcellulose. Incorporation of 4-(methylamino)-2-ethyl-5,5dimethyl-4-pyridine-2-yl-2,5-dihydro-1H-imidazol-1-oxyl (MEP) as a spin label enabled measurement of the pHM within a pH range of 3 to 5. The maintenance of a low pHM depended on the physicochemical properties of the incorporated pH modifiers, and was favored by high acidic strength and low aqueous solubility. Consequently, fumaric acid resulted in a constant acidic microenvironment within the investigated period. The high water-solubility of citric acid rapidly decreased the pHM at the start of incubation, but this effect was reduced especially at the margins of the tablet due to the rapid dissolution of the acid from the matrix system. More water-soluble pH modifiers such as citric acid stimulated the water transport into the matrix and resulted in a rapid lowering of τR. Therefore the solubility of the acids is a crucial factor in the kinetics of water penetration. Once the gel matrix was hydrated the mobility of the spin label was similar to that in water and thus independent of the type of acid. EPRI helps to understand the effect of pH modifiers on the stage of hydration and the microenvironmental pH inside matrix tablets, thus aiding selection of the optimal pH modifying system. References [1] V.M. Rao, K. Engh, Y.H. Qiu, Design of pH-independent controlled release matrix tablets for acidic drugs, Int. J. Pharm. 252 (2003) 81–86. [2] M.B. Kabadi, R.V. Vivilecchia. Stabilized pharmaceutical compositions comprising an HMG-COA reductase inhibitor compound, U.S. Patent 5,356,896, Oct. 18, 1994. [3] F. Nikfar, A.T. Serajuddin, N. Jerzewiski, N. Jain. Pharmaceutical compositions having good dissolution properties, U.S. Patent 5,506,248, Apr. 9, 1996. [4] A. Streubel, J. Siepmann, A. Dashevsky, R. Bodmeier, pH-independent release of a weakly basic drug from water-insoluble and -soluble matrix tablets, J. Control. Release 67 (2000) 101–110. [5] S. Nie, W. Pan, X. Li, X. Wu, The effect of citric acid added to hydroxypropyl methylcellulose (HPMC) matrix tablets on the release profile of vinpocetine, Drug Dev. Ind. Pharm. 30 (2004) 627–635. [6] B.A. Burnside, R.K. Chang, X. Guo. Sustained release pharmaceutical dosage forms with minimized pH dependent dissolution profiles, U.S. Patent 6,287,599, Sep. 11, 2001. [7] K.E. Gabr, Effect of organic acids on the release patterns of weakly basic drugs from inert sustained release matrix tablets, Eur. J. Pharm. Biopharm. 38 (1992) 199–202. [8] S.J. Cope, S. Hibberd, J. Whetstone, R.J. MacRae, C.D. Melia, Measurement and mapping of pH in hydrating pharmaceutical pellets using confocal laser scanning microscopy, Pharm. Res. 19 (2002) 1554–1563.

[9] P.A. Burke, Determination of internal pH in PLGA microspheres using 31P NMR spectroscopy, Proc. Int. Symp. Control. Release Bioact. Mater. 23 (1996) 133–134. [10] S. Capancioni, K. Schwach-Abdellaoui, W. Herrmann, H. Brosig, R. Gurny, In vitro monitoring of poly(ortho ester) degradation by electron paramagnetic resonance imaging, Macromolecules 36 (2003) 6135–6141. [11] K. Mäder, S. Nitschke, R. Stösser, H.H. Borchert, A. Domb, Nondestructive and localized assessment of acidic microenvironments inside biodegradable polyanhydrides by spectral spatial electron paramagnetic resonance imaging, Polymer 38 (1997) 4785–4794. [12] V.V. Khramtsov, L.M. Weiner, I.A. Grigoriev, L.B. Volodarsky, Proton exchange in stable nitroxyl radicals. EPR study of the pH of aqueous solutions, Chem. Phys. Lett. 91 (1982) 69. [13] V. Pillay, R. Fassihi, Electrolyte-induced compositional heterogeneity: a novel approach for rate-controlled oral drug delivery, J. Pharm. Sci. 88 (1999) 1140–1148. [14] M.V.S. Varma, A.M. Kaushal, S. Garg, Influence of micro-environmental pH on the gel layer behavior and release of a basic drug from various hydrophilic matrices, J. Control. Release 103 (2005) 499–510. [15] H.M. Rauen, Biochemisches Taschenbuch, Zweiter Teil, Springer Verlag, Berlin, Germany, 1964, pp. 90–106. [16] M. Chzhan, M. Shteynbuk, P. Kuppusamy, J.L. Zweier, An optimized lband ceramic resonator for EPR imaging of biological samples, J. Magn. Reson. 105 (1993) 49–53. [17] M. Chzhan, P. Kuppusamy, J.L. Zweier, Development of an electronically tunable l-band resonator for EPR spectroscopy and imaging of biological samples, J. Magn. Reson. 108 (1995) 67–72. [18] D.E. Budil, S. Lee, S. Saxena, J. Freed, Nonlinear-least-square analysis of slow-motion EPR spectra in one and two dimensions using a modified Levenberg–Marquardt algorithm, J. Magn. Reson. 120 (1996) 155–189. [19] W. Herrmann, R. Stösser, K.P. Moll, H.H. Borchert, Rates of rotational diffusion and Heisenberg spin exchange as obtained from cw ESR and ESR tomographic experiments on model systems and human skin, Appl. Magn. Reson. 28 (2005) 85–106. [20] K. Mäder, Pharmazeutische Anwendungen der Elektronenspinresonanzspektroskopie (ESR), PZ Prisma 3 (1998) 202. [21] I. Katzhendler, K. Mäder, M. Friedman, Correlation between drug release kinetics from proteineous matrix and matrix structure: EPR and NMR study, J. Pharm. Sci. 89 (2000) 365–381. [22] I. Katzhendler, K. Mäder, R. Azoury, M. Friedman, Investigating the structure and properties of hydrated hydroxypropyl methylcellulose and egg albumin matrices containing carbamazepine: EPR and NMR study, Pharm. Res. 17 (2000) 1299–1308. [23] N. Kohri, N. Miyata, M. Takahashi, H. Endo, K. Iseki, K. Miyazaki, S. Takechi, A. Nomura, Evaluation of pH-independent sustained-release granules of dipyridamole by using gastric-acidity-controlled rabbits and human subjects, Int. J. Pharm. 81 (1992) 49–58. [24] R. Espinoza, E. Hong, L. Villafuerte, Influence of admixed citric acid on the release profile of pelanserin hydrochloride from HPMC matrix tablets, Int. J. Pharm. 201 (2000) 165–173. [25] H.-J. Galla, Elektronenspinresonanz, Thieme Verlag, Stuttgart, New York, 1988, p. 89.