Search

The crystallization of active pharmaceutical ingredients (APIs) in matrix-type transdermal patches has implications for the rate of drug absorption through the skin and patch adhesion strength. Therefore, the presence or absence and the degree of API crystallinity must be controlled to guarantee the quality of patches. In this study, the utility of laboratory-level X-ray diffractometers for the detection and analysis of crystalline APIs in transdermal patches was investigated using medical patches of tulobuterol and isosorbide dinitrate. Several matrix-type patches employ a controlled drug delivery system containing intentionally crystallized API. Both benchtop and high-resolution laboratory X-ray diffractometers can detect several characteristic peaks of the APIs in these patches even if the patches are wrapped in an outer bag, although a benchtop model provides peak heights one-seventh to one-fifth that of a high-resolution instrument. An isosorbide dinitrate patch containing an unintentionally crystallized spot was wrapped in an outer bag, followed by measurements using both X-ray diffractometers. For both instruments, several isosorbide dinitrate-derived peaks were detected only at the crystallized spot, although the signal-to-noise ratio was poorer for the benchtop model. These results show that a high-resolution X-ray diffractometer is advantageous for high-detection sensitivity and offers a high degree of freedom of the measurement position on the sample. It was concluded that a laboratory-level high-resolution X-ray diffractometer can be used to examine the crystalline state of APIs in patches inside an unopened outer bag.

Undesired drug sorption on laboratory material surfaces reduces the performance of analytical methods and results in the generation of unreliable data. Hence, we characterized the sorption of drugs and evaluated the sorption extent using a linear free energy relationship (LFER) model with Abraham solvation parameters of drugs. Furthermore, to prevent sorption, the effects of additives, such as organic solvents and salts, were evaluated.The sorption of fifteen model drugs (concentration: 2 μM), with various physicochemical properties, on materials in 0.2% dimethyl sulfoxide aqueous solutions was evaluated. Drug sorption extent on the materials was determined using high-performance liquid chromatography. The obtained results were analyzed using an LFER model with Abraham solvation parameters of the drugs. The effect of additives on the sorption of itraconazole, one of the most hydrophobic drugs among those tested in this study, was investigated.Sorption was dependent on the physicochemical properties of drugs, rather than the type of materials used, and additives altered the rate of drug sorption. Equations were developed to evaluate the sorption extent (nmol) of drugs to glass and polypropylene using the Abraham solvation parameters of the drugs.LFER modeling with Abraham solvation parameters of drugs enabled us to evaluate drug sorption on materials. All the additives altered the rate of drug sorption, and some organic solvents effectively prevented sorption. The developed LFER model would be useful for assessment of the sorption properties of compounds in in vitro evaluations in drug discovery research and various other biochemical fields.

The purpose of this study was to develop a quality evaluation method for dissolving microneedle arrays (DMNAs) and determine the spatial distribution pattern of drugs in DMNAs. Raman spectroscopy mapping was used to visualize the drug distribution in DMNAs and drug-loaded polymer films as a model. Powder X-ray diffraction (PXRD) and high-pressure liquid chromatography were also performed to characterize DMNAs. Drug-loaded polymer films and DMNAs were prepared by drying the aqueous solutions spread on the plates or casting. PXRD analysis suggested the crystallization of diclofenac sodium (DCF) in several forms depending on its amount in the sodium hyaluronate (HA)-based films. The Raman spectra of HA and DCF showed characteristic and non-overlapping peaks at 1376 and 1579 cm -1 Raman shifts, respectively. The intensity of the characteristic peak of DCF in the DCF-loaded films increased linearly with the increasing drug content in the range of 4.8 to 16.7 % (DCF, w/w). Raman imaging analysis revealed a homogenous dispersion of small DCF crystals in these films. Raman imaging indicates the distribution of DCF on the surface of the DMNA needle. This work highlights the benefit of using Raman spectroscopy mapping to reveal the spatial distribution of drugs in DMNAs.

Amorphization is a powerful approach for improving the aqueous solubility and bioavailability of poorly water-soluble compounds. However, it can cause chemical and physical instability, the latter of which can lead to crystallization during storage, diminishing the solubility advantage of the amorphous state. As there is no standard method for predicting the physical stability of amorphous materials, a long-term stability study is needed in drug development. This study investigated the correlation between the physical stability of amorphous compounds and molecular mobility based on the assumption that physical stability is governed by the diffusional motion of a molecule. Model compounds were evaluated for crystallization onset time, structural relaxation time, fragility, and fictive temperature. The crystallization onset time of acetaminophen glass correlated with its relaxation time calculated from the Adam-Gibbs-Vogel equation; however, that of felodipine glass correlated with the relaxation time calculated from the Vogel-Tammann-Fulcher equation. The different crystallization tendencies of these compounds can be explained by the differences in the rate limiting steps in their crystallization processes, indicating the importance of distinguishing the critical process associated with crystallization. These findings will be useful for more accurate prediction of long-term physical stability of amorphous materials.

The objective of this study was to develop a simpler and more practical quantitative evaluation method of cold flow (CF) in transdermal drug delivery systems (TDDSs). CF was forcibly induced by loading a weight on a punched-out sample (bisoprolol and tulobuterol tapes). When the extent of CF was analyzed using the area of oozed adhesive as following a previously reported method, the CF profiles were looked different between the samples 12 mm in diameter subjected to a 0.5-kg weight and samples 24 mm in diameter subjected to a 2.0-kg weight despite an equal load per unit area (4.42 g/mm2). The width of oozed adhesive around the original sample was suggested to be an index that properly describes the relationship between the load per unit area and the extent of CF. Further, it was clarified that the average CF width over the entire circumference of the sample was the same whether the samples were round or square as long as the sample area and load were the same. We also observed a linear relationship between the CF width and the aspect ratio of oval and rectangular samples. These results indicated that the CF properties of typical TDDS products lacking CF-proof processing at the edges could be determined by testing samples cut from the product rather than the whole TDDS patch. The proposed width measuring method was simple and useful for optimizing the composition of the adhesive and for testing the quality of the product.

N-Nitrosodimethylamine (NDMA) has been detected in some drug substances and pharmaceutical products containing sartans, ranitidine and metformin, and a potential risk of NDMA contamination exists in other drug substances and their pharmaceutical products. To quantitate NDMA in various drugs having diverse physicochemical properties, a specific, sensitive, and reliable analytical method is required, in addition to methods that can be applied to a class of nitrosamines. We aimed to develop an off-line isolation method for NDMA in drug substances using SPE for quantification with LC-APCI-MS/MS. Impediments to accurate quantitation of NDMA in drug substances using LC-MS/MS and insufficient durability of the system are attributed to the extremely large amounts of active pharmaceutical ingredients (APIs) in sample solutions in comparison to the trace amount of NDMA. A reduced retention of NDMA and/or decreased separation from other substances in LC, matrix effect in MS detection, and undesirable contamination of instruments with API and other substances may be occasionally encountered, all of which consequently result in deterioration of system performance and generation of unreliable data, even in the cases where a divert valve is configured between the column and ion source of the MS instrument. To address these problems, an off-line NDMA isolation methodology from APIs exhibiting diverse physicochemical properties, namely ranitidine hydrochloride (ranitidine), metformin hydrochloride (metformin), nizatidine, valsartan, and telmisartan, was developed. The applicability of the method was confirmed by batch analysis of metformin and ranitidine. Furthermore, contrary to previous reports, NDMA was found to be stable over a wide pH range. The proposed methodology and data from this study would contribute to the control of NDMA contamination in various drugs to realize the safe delivery of pharmaceuticals to patients.

Forty-four brands of transdermal patches for twelve kinds of active pharmaceutical ingredients (APIs) are available in Japan as of April 30, 2018. Although approximately one-third of the corresponding pharmaceutical interview forms lack information on how to evaluate the adhesive properties of the patches, the peel test, probe tack test, or inclined ball tack test have generally been adopted. This means that it might be difficult to simply compare the adhesive properties among the patches because the testing methods are not unified in some cases. In this study, measurements of the adhesive properties of 38 transdermal patches of ten different APIs were performed using several unified testing methods (180° peel test, 90° peel test, self-adhesion test, and probe tack test) under unified experimental conditions. The adhesive properties were found to be quite different among the patches, even for the same API, dose, and size. For example, the ratios of the maximum to minimum measured values of tack and 180° peel strength for tulobuterol patches were 5 and 29, respectively. In the case of generic products for which the bioequivalence to a brand-name product is assured, the variation in adhesive properties can extend the range of choices for patients, which is advantageous. Providing information to medical experts on adhesive properties through, for example, pharmaceutical interview forms and package inserts, is considered to be useful for helping patients to make better choices.

A non-destructive discrimination method for crystals in solid dosage drug forms was first developed using a combination of Raman spectroscopy and X-ray micro-computed tomography (X-ray CT). Identification of the crystal form of an active pharmaceutical ingredient (API) at the appropriate pharmaceutical dosage is crucial, as the crystal form is a determinant of the quality and performance of the final formulation. To develop a non-destructive analytical methodology for the discrimination of solid API crystals in a solid dosage form, we utilized a combination of Raman spectroscopy and X-ray CT to differentiate between ranitidine crystal polymorphs (forms 1 and 2) in tablet formulations containing three excipients. The difference in electron density correlated with the true density between ranitidine polymorphs, thereby enabling the discrimination of crystal forms and visualization of their three-dimensional spatial localization inside the tablets through X-ray CT imaging. Furthermore, X-ray CT imaging revealed that the crystal particles were of varying densities, sizes, and shapes within the same batch. These findings suggest that X-ray CT is not only an imaging tool but also a unique method for quantitative physicochemical characterization to study crystal polymorphs and solid dosage forms.

Amorphous nifedipine (NFD), which has a smooth surface immediately after preparation, was shown to have structures resembling clusters of curling and branching fibers approximately 1 μm wide by atomic force microscopy (AFM) after storage at 25°C. The size of the cluster‐like structures increased with storage over time, implying crystal growth. The average elongation rate of the fibers determined by AFM at ambient room temperature was 1.1 × 10 −9 m/s, and this agreed well with the crystal growth rate of 1.6 × 10 −9 m/s determined by polarized light microscopy. The crystal growth rate of NFD in solid dispersions with 5% polyethylene glycol (PEG) was found to be 5.0 × 10 −8 m/s by AFM. Although this value was approximately the same as that obtained by polarized light microscopy, three‐dimensional information obtained by AFM for the crystallization of NFD in a solid dispersion with PEG revealed that the changes in topography were not a consequence of surface crystal growth, but rather attributable to the growth of crystals formed in the amorphous bulk. For solid dispersions with α,β‐poly(N‐5‐hydroxypentyl)‐l‐aspartamide, acceleration of NFD crystallization by tapping with an AFM probe was observed. The present study has demonstrated the feasibility and application of AFM for interpretation of surface crystallization data. © 2011 Wiley‐Liss, Inc. and the American Pharmacists Association J Pharm Sci 100:4413–4420, 2011

To clarify the contribution of drug-polymer interaction to the physical stability of amorphous solid dispersions, we studied the crystallization rates of nitrendipine (NTR) enantiomers with identical physicochemical properties in the presence of hydroxypropylmethylcellulose (HPMC), hydroxypropylmethylcellulose phthalate (HPMCP) and polyvinylpyrrolidone (PVP). The overall crystallization rate at 60°C and the nucleation rate at 50-70°C of (+)-NTR were lower than those of (-)-NTR in the presence of 10-20% HPMC or HPMCP. In contrast, similar crystallization profiles were observed for the NTR enantiomers in solid dispersions containing PVP. The similar glass transition temperatures for solid dispersions of (-)-NTR and (+)-NTR suggested that the molecular mobility of the amorphous matrix did not differ between the enantiomers. These results indicate that the interaction between the NTR enantiomers and HPMC or HPMCP is stereoselective, and that differences in the stereoselective interaction create differences in physical stability between (-)-NTR and (+)-NTR at 50-70°C. However, no difference in physical stability between the enantiomers was obvious at 40°C. Loss of the difference in physical stability between the NTR enantiomers suggests that the stereoselective interaction between NTR and the polymers may not contribute significantly to the physical stabilization of amorphous NTR at 40°C.

The relative influences of chemical activation energy and molecular mobility in determining chemical reactivity were evaluated for insulin lyophilized with α,β-poly( N -hydroxyethyl)- L -aspartamide (PHEA), and compared with that for insulin lyophilized with trehalose, which had been found to have the ability to decrease the molecular mobility of insulin at low humidity. The ratio of the observed rate constant k obs to the chemical activation energy-controlled rate constant k act ( k obs / k act ) at glass transition temperature ( T g ) was estimated to be approximately 0.6 and 0.8 at 6% RH and 12% RH, respectively, indicating that the degradation rate is significantly affected by molecular mobility at lower humidity conditions. However, these k obs / k act values at T g were larger than those for the insulin-trehalose system, and changes in the temperature-dependent slope around T g were less obvious than those for the insulin-trehalose system. Thus, the contribution of molecular mobility to the degradation rate in the insulin-PHEA system appeared to be less intense than that in the insulin-trehalose system. The subtle change in the temperature-dependent slope around T g observed in the insulin-PHEA system brought about a significant bias in shelf-life estimation when the reaction rate was extrapolated from temperatures above T g according to the Arrhenius equation.

The purpose of this study was to elucidate whether the degradation rate of insulin in lyophilized formulations is determined by matrix mobility, as reflected in glass transition temperature (Tg), or by β-relaxation, as reflected in rotating-frame spin-lattice relaxation time $${\left( {T_{{1{\text{ $ \rho $ }}}} } \right)}$$ . The storage stability of insulin lyophilized with dextran was investigated at various relative humidities (RH; 12–60%) and temperatures (40–90°C) and was compared with previously reported data for insulin lyophilized with trehalose. Insulin degradation was monitored by reverse-phase high-performance liquid chromatography. Furthermore, the $${\left( {T_{{1{\text{ $ \rho $ }}}} } \right)}$$ of the insulin carbonyl carbon in the lyophilized insulin–dextran and insulin–trehalose systems was measured at 25°C by 13C solid-state NMR, and the effect of trehalose and dextran on $${\left( {T_{{1{\text{ $ \rho $ }}}} } \right)}$$ was compared at various humidities. The degradation rate of insulin lyophilized with dextran was not significantly affected by the Tg of the matrix, even at low humidity (12% RH), in contrast to that of insulin lyophilized with trehalose. The insulin–dextran system exhibited a substantially greater degradation rate than the insulin–trehalose system at a given temperature below the Tg. The difference in degradation rate between the insulin–dextran and insulin–trehalose systems observed at 12% RH was eliminated at 43% RH. In addition, the $${\left( {T_{{1{\text{ $ \rho $ }}}} } \right)}$$ of the insulin carbonyl carbon at low humidity (12% RH) was prolonged by the addition of trehalose, but not by the addition of dextran. This difference was eliminated at 23% RH, at which point the solid remained in the glassy state. These findings suggest that the β-relaxation of insulin is inhibited by trehalose at low humidity, presumably as a result of insulin–trehalose interaction, and thus becomes a rate determinant. In contrast, dextran, whose ability to interact with insulin is thought to be less than that of trehalose, did not inhibit the β-relaxation of insulin, and thus, the chemical activational barrier (activation energy) rather than β-relaxation becomes the major rate determinant. β-Relaxation rather than matrix mobility seems to be more important in determining the stability of insulin in the glassy state in lyophilized formulations containing trehalose and dextran.

The purpose of this study is to confirm the speculation which arose in our previous study that the degradation rate of insulin lyophilized with poly(vinylpyrrolidone) is mainly governed by the chemical activational barrier rather than molecular mobility. This speculation was based on the degradation data of insulin lyophilized with poly(vinylpyrrolidone) K-30 (PVP K-30), which was obtained at temperatures well below the glass transition temperature ( T g ). In this study, the degradation rate of insulin at temperatures below and above T g was determined using PVP 10k as an excipient, instead of PVP K-30, in order to examine whether or not the temperature dependence of the degradation rate changes around T g . The relative contributions of molecular mobility and the activational barrier, calculated from the temperature- and T g -dependence of the degradation rate, indicated that the contribution of molecular mobility to the degradation rate was negligible. Furthermore, the negligible contribution of molecular mobility was confirmed by the lack of significant change observed in the temperature- and T g -dependence of the rate around T g .

The partition behavior of tropane alkaloids (TrA) between organic solvent and water was studied under various conditions. The apparent partition coefficient, log P', increases with pH with a slope of +1 in the acidic to neutral pH range, then tends to approach to a constant value in the alkaline region. Thus the partition of TrA to the organic phase is based on the neutral form of TrA. From the pH dependency of the partition and pH titration of TrA, the intrinsic partition coefficients log P0 and pKa values for each TrA were estimated. Values of pKa and log P0 for scopolamine were smaller than those for atropine or homatropine. It was considered that the presence of the epoxide ring in scopolamine molecule reduces the log P0 and pKa because of its polar and electro-inductive effect. It was also expected from the results of pH-dependent partition behavior that in the low pH region some portion of TrA cations is transferred to the organic phase by forming an ion-pair complex with an anion, in the aqueous phase. The log P' of TrA was measured in the presence of various kinds of anions (halides and alkyl sulfonates, C5-C8) at about pH 2, where almost all TrA are present as a protonated form. The relations between the log P' and the logarithmic concentration of anions is linear at low concentrations of anions as expected from the ion-pair partition equation.(ABSTRACT TRUNCATED AT 250 WORDS)

The dehydration of different commercial brands of dibasic calcium phosphate dihydrate (DCPD; CaHPO(4).2H(2)O) was examined over a range of temperatures and water vapor pressures. To determine the main factors affecting the physical stability of DCPD, the baseline characterization of DCPD and dibasic calcium phosphate anhydrate (DCPA; CaHPO(4)) was conducted by thermogravimetric analysis, differential scanning calorimetry and X-ray diffractometry. The surface area and the DCPA content (present as an impurity) depended on the commercial source of DCPD. The larger particles contained a higher concentration of DCPA and the anhydrate exhibited a concentration-dependent acceleratory effect on the dehydration of DCPD. Unlike DCPD, DCPA is physically stable and resisted hydration even when dispersed in water for over 7 months in the temperature range of 4-50 degrees C. In dosage forms containing DCPD, there is a potential for phase transformation to DCPA, while the reverse transition, that is, DCPA --> DCPD appears to be extremely unlikely. Thus, the risk of physical transformation can be minimized by using DCPA in formulations.

The purpose of this study is to compare the effects of global mobility, as reflected by glass transition temperature (T(g)) and local mobility, as reflected by rotating-frame spin-lattice relaxation time (T(1rho)) on aggregation during storage of lyophilized beta-galactosidase (beta-GA).The storage stability of beta-GA lyophilized with sucrose, trehalose or stachyose was investigated at 12% relative humidity and various temperatures (40-90 degrees C). beta-GA aggregation was monitored by size exclusion chromatography (SEC). Furthermore, the T(1rho) of the beta-GA carbonyl carbon was measured by (13)C solid-state NMR, and T(g) was measured by modulated temperature differential scanning calorimetry. Changes in protein structure during freeze drying were measured by solid-state FT-IR.The aggregation rate of beta-GA in lyophilized formulations exhibited a change in slope at around T(g), indicating the effect of molecular mobility on the aggregation rate. Although the T(g) rank order of beta-GA formulations was sucrosetrehalosestachyose, the rank order of beta-GA aggregation rate at temperatures below and above T(g) was also sucrosetrehalosestachyose, thus suggesting that beta-GA aggregation rate is not related to (T-T(g)). The local mobility of beta-GA, as determined by the T(1rho) of the beta-GA carbonyl carbon, was more markedly decreased by the addition of sucrose than by the addition of stachyose. The effect of trehalose on T(1rho) was intermediate when compared to those for sucrose and stachyose. These findings suggest that beta-GA aggregation rate is primarily related to local mobility. Significant differences in the second derivative FT-IR spectra were not observed between the excipients, and the differences in beta-GA aggregation rate observed between the excipients could not be attributed to differences in protein secondary structure.The aggregation rate of beta-GA in lyophilized formulations unexpectedly correlated with the local mobility of beta-GA, as indicated by T(1rho), rather than with (T-T(g)). Sucrose exhibited the most intense stabilizing effect due to the most intense ability to inhibit local protein mobility during storage.

To examine the relative contributions of molecular mobility and thermodynamic factor, the relationship between glass transition temperature (T(g)) and the crystallization rate was examined using amorphous dihydropyridines (nifedipine (NFD), m-nifedipine (m-NFD), nitrendipine (NTR) and nilvadipine (NLV)) with differing T(g) values. The time required for 10% crystallization, t(90), was calculated from the time course of decreases in the heat capacity change at T(g). The t(90) of NLV and NTR decreased with decreases in T(g) associated with water sorption. The t(90) versus T(g)/T plots almost overlapped for samples of differing water contents, indicating that the crystallization rate is determined by molecular mobility as indicated by T(g). In contrast, differences in the crystallization rate between these four drugs cannot be explained only by molecular mobility, since the t(90) values at a given T(g)/T were in the order: NLV>NTR>NFD approximately m-NFD. A lower rate was obtained for amorphous drugs with lower structural symmetry and more bulky functional groups, suggesting that these factors are also important. Furthermore, the crystallization rate of NTR in solid dispersions with poly(vinylpyrrolidone) (PVP) and hydroxypropyl methylcellulose (HPMC) decreased to a greater extent than expected from the increased T(g). This also suggests that factors other than molecular mobility affect the crystallization rate.

The inhibition of crystallization of amorphous acetaminophen (ACTA) by polyvinylpyrrolidone (PVP) and polyacrylic acid (PAA) was studied using amorphous solid dispersions prepared by melt quenching. Co-melting with PVP and PAA decreased the average molecular mobility, as indicated by increases in glass transition temperature and enthalpy relaxation time. The ACTA/PAA dispersion exhibited much slower crystallization than the ACTA/PVP dispersion with a similar glass transition temperature value, indicating that interaction between ACTA and polymers also contributed to the stabilizing effect of these polymers. The carboxyl group of PAA may interact with the hydroxyl group of ACTA more intensely than the carbonyl group of PVP does, resulting in the stronger stabilizing effect of PAA. Dielectric relaxation spectroscopy showed that the number of water molecules tightly binding to PVP per monomer unit was larger than that to PAA. Furthermore, a small amount of absorbed water decreased the stabilizing effect of PVP, but not that of PAA. These findings suggest that the stronger stabilizing effect of PAA is due to the stronger interaction with ACTA. The ability of PAA to decrease the molecular mobility of solid dispersion was also larger than that of PVP, as indicated by the longer enthalpy relaxation time.

A simple mixing with hyaluronan (HA), doxycycline (DC) and divalent metal cation in an aqueous solution enabled a thermoreversible water-soluble gel to form. For the cross-linking, two kinds of interactions were supposed. One was an electrostatic interaction between a positively charged group in DC and a negatively charged carboxyl function of HA, and the other was a chelation at the phenolic diketone moiety in DC. Since the gel was formed physically, the critical polymer concentration for gelation was present, and it was about 0.05% for HA with a molecular weight of 1.6x10(6). The hydrogel would be formed holding water in the HA entanglement network when DCs on HA chains made coordinate bonds through metal chelation. By changing the mixing ratio, two types of gels with different characteristics in drug release could be prepared. One was a gel with zero-order release prepared by mixing the same amount of HA and DC in equivalent. The other was a gel indicating Fickian diffusion-type release by mixing more DC than HA. Further, by controlling the absolute concentration of HA and DC, or the molecular weight of HA, some gels with desired release profiles could be prepared.

The interaction between sodium hyaluronate (HA) and 6 kinds of tetracycline (TCs) was investigated by dissolution tests, conductivity, viscosity and turbidity measurements. The 6 TCs were tetracycline (TEC), oxytetracycline (OTC), doxycycline (DOC), methacycline (METC), minocycline (MINC) and chlortetracycline (CTC). In dissolution tests, the release of TCs from the HA solution contained in the cellulose tube was measured and it was found that the 50% release time (T50) of TCs was prolonged with the increase of HA concentration. And the dependence of T50 on HA concentration was different in TCs: the release of METC was most delayed and that of OTC was least affected. By measuring the conductivity of TCs solutions, the binding of TCs to HA was suggested. In viscosity measurements of the HA solution at various pHs, the increase of viscosity was observed by the addition of METC. However in the case of OTC such an increase was not observed. The dissolution tests at various pHs were further tried and the release rate of METC depended on the pH, while that of OTC showed no dependency. Finaly, it was found that T50 of TCs from the HA solution correlates well with the literature values of apparent partition coefficients of TCs between n-octanol and aqueous buffers. Comparing T50 of 6 TCs from carboxymethylchitin (DS = 0.56) solution, it was clarified that the lipophilicity of TCs are more effective to retard the release in the case of the HA solution. From the above results, an electrostatic interaction between HA and TCs can be assumed to be almost similar among 6 TCs.(ABSTRACT TRUNCATED AT 250 WORDS)

In the dissolution test of tablets prepared by the direct compression of chitosan with 10% sodium alginate (NaAlg), the release of theophylline was extended independently of the medium pH. The drug release mechanism of the tablet was investigated by measuring the dissolution behavior of such excipients as chitosan and NaAlg. It was shown that at the lower pH the drug release was accompanied with the dissolution of chitosan, while at the neutral pH it was controlled by the gelation and dissolution of NaAlg. Further more by the oral administration of the tablet to beagle dogs, the sustained release of theophylline was also confirmed by some bioavailability parameters. Thus the chitosan tablet with 10% NaAlg was expected to be a pH-independent sustained release dosage form.