Mechanisms of drug release in citrate buffered HPMC matrices
Introduction
Hydrophilic matrices based on hypromellose or hydroxypropyl methylcellulose (HPMC) are widely used within the pharmaceutical industry (Melia, 1991, Li et al., 2005). Drug solubility is a key determinant of release in these dosage forms and, when drug solubility is pH-dependent, the changing pH environment in the gastro-intestinal tract can give rise to changes in drug solubility and a change in the drug release mechanism (Badawy and Hussain, 2007). The most common examples are weakly basic drugs, which have a high solubility in the stomach but are poorly soluble at the higher pH of the duodenum. In an HPMC matrix tablet, this would result in a switch from diffusion to erosion-controlled release at around the time of gastric emptying and, as a consequence, we can envisage that drug bioavailability may become highly dependent on patient variables and the fed state. A common approach to this problem is to maintain the drug in its soluble form by incorporating pH-modifying excipients in the matrix.
The release of weakly basic drugs is commonly improved by the inclusion of weak acids or acidic polymers (Gabr, 1992, Thoma and Ziegler, 1998, Streubel et al., 2000, Espinoza et al., 2000, Varma et al., 2005, Kranz et al., 2005, Siepe et al., 2006). In contrast, neutral or high pH buffering excipients for HPMC matrices containing weak acid drugs have received relatively little attention. Phosphates, citrates, carbonates, magnesium oxide/hydroxide and Eudragit E have all been proposed as buffering agents for extended release dosage forms (Doherty and York, 1989, Akiyama et al., 1994, Rao et al., 2003, Li and Schwendeman, 2005, Riis et al., 2007) but few of these examples have been used in hydrophilic matrix systems.
Critically, an issue yet to be fully explored is the influence of these pH modifying agents on the drug release mechanism. For example, buffers are ionic in nature, and dissolved ions can modulate polymer: water interactions in non-ionic hydrophobically modified cellulose ethers such as HPMC. This can be conveniently monitored by the temperature dependency of the sol: gel phase transition (Touitou and Donbrow, 1982). Highly ionic electrolytes depress the temperature of this transition by competing for water molecules in the polymer hydration sheath. This increases molecular dehydration and results in the clustering of hydrophobically substituted regions and the eventual formation of a 3D network stabilized by hydrophobic interactions (Sarkar, 1979; Doelker, 1993; Haque and Morris, 1993). The observed effect is gel formation or turbid precipitation of the polymer (“salting out”). The potency of different electrolytes to do this follows a Hofmeister-like series, and the effects are greatest with multivalent anions (Nakano et al., 1999). This effect can be valuable. The suppression of polymer swelling and solubility has been used to aid polymer dispersion (Yuasa et al., 1997) and ions high in the Hofmeister series have been used to inhibit agglomeration during HPMC film coating (Nakano et al., 1999). This can be attributed to a reduction in film tackiness resulting from the inhibition of polymer solubility. Sodium citrate was found to be a particularly potent agent in inhibiting agglomeration, but it also resulted in more brittle, porous films with a low tensile strength (Nakano and Yuasa, 2001). This suggests HPMC molecules are adopting a more compact globular conformation in the presence of trivalent citrate, as might be expected if water interactions were reduced and intra-molecular hydrophobic interactions were promoted. The reduction in polymer solubility may also result in a some polymer being unable to contribute to the polymer film network.
Multivalent ions are known to influence drug release in HPMC matrices (Alderman, 1984, Touitou and Donbrow, 1982, Mitchell et al., 1990). Lapidus and Lordi (1966) first reported that the presence of electrolytes in the dissolution medium could influence drug release and Fagan et al. (1989) showed how disintegration of HPMC, HPC and HEC matrices can occur as the chloride or phosphate concentration in the medium approached the cloud point of the polymer. Mitchell et al. (1990) demonstrated that salt-induced changes can markedly affect HPMC viscosity and reported how higher sodium phosphate concentrations in the test medium resulted in a rapid ‘burst’ of drug release. Hodsdon et al. (1993) showed how increasing phosphate ion concentration decreased the disintegration time of HPMC matrices. Incorporation of ionic salts into HPMC matrices has also been used to influence drug release kinetics although only speculative reasons for the underlying mechanism have been advanced. For example, sodium dihydrogen phosphate (NaH2PO4) has been used to modulate the release of acetaminophen in HPMC matrices (Cao et al., 2005) whilst sodium carbonate and pentasodium tripolyphosphate modulate the release of metoprolol tartrate in polyethylene oxide matrices (Pillay and Fassihi, 2000). Alderman (1984) has reported the highly adverse effects on extended release when high valency salts such as aluminium sulphate (Al2(SO4)3), sodium sulphate (Na2SO4) and sodium carbonate (Na2CO3) are incorporated into a HPMC hydrophilic matrix. Changes in release rate may also occur when incorporated salts influence drug solubility through in situ salt formation or common-ion effects, for example when sodium chloride is incorporated into diclofenac matrices (Sheu et al., 1992).
Overall, the literature suggests that incorporated ionic substances have the potential to substantially alter drug release kinetics in HPMC matrices, and that multivalent buffers might have potent Hofmeister effects on the interaction of HPMC with water. As a result, in addition to increasing drug solubility, a buffering agent may also influence polymer swelling, dissolution and the structure of the hydrated polymer network, which are key factors in HPMC matrix gel layer retardation of drug release. The increasing use in recent years of hydrophilic matrix technology to develop floating gastro-retentive dosage forms has made buffering against prolonged exposure to acidic conditions an increasingly important area, and using citrate buffers in the trivalent state would maximize buffering capacity and be an obvious choice. A higher bioavailability of weakly acidic drugs such as NSAIDS has been demonstrated in the presence of alkalizing agents, and citrate buffers can be used to improve drug solubility in these circumstances (Fuder et al., 1997, Neuvonen, 1991). In this work, we investigate how incorporated trivalent sodium citrate influences the behaviour of HPMC matrices containing the weak acid NSAID drug felbinac, with respect to the underlying drug release mechanisms.
Section snippets
Materials
Methocel E4M (Hypromellose USP 2910) and K4M (Hypromellose USP 2208) HPMC CR premium EP were kind gifts from Colorcon Ltd. (Orpington, Kent). Felbinac (4-biphenylacetic acid) was obtained from Sigma (Poole, Dorset). Monobasic, dibasic and tribasic sodium citrate dihydrate were obtained from Acros Organics (New Jersey, USA). Tris(hydroxymethyl)aminomethane was obtained from Sigma–Aldrich (Poole, Dorset). Universal pH indicator was obtained from Fisher Scientific (Leicestershire, UK). Compression
The effect of citrate ion valency on the cloud point temperature of HPMC solutions
Fig. 1 shows the effect of buffer concentration on the cloud point temperature (CPT) of HPMC solutions containing different citrate salts. Whilst monovalent ions had little influence, multivalent citrate salts lowered CPT to an extent approximately proportional to the charge on the anion. The tribasic citrate salt had the most potent effect on cloud point (HPMC 2910 Methocel E4M ΔCPT = −186.05 °C M−1, HPMC 2208 Methocel K4M ΔCPT = −208.1 °C M−1). This is typical of Hofmeister behaviour, in which a
Conclusions
The inclusion of sodium citrate in HPMC matrix tablets accelerated the release of the weak acid model drug felbinac, but the mechanism for this effect was clearly more complex than simple improvement of drug solubility. The effect was more profound in HPMC 2910 matrices where faster drug release was accompanied by a loss of extended release characteristics. In HPMC 2208 matrices there was no loss of extended release control. The increased immediate ‘burst’ release observed in the HPMC 2910
Acknowledgments
We gratefully acknowledge Bristol Myers-Squibb for funding this research. We also extend our gratitude to Professor Martin Snowden for suggesting the concept of an osmotic driving force, to Hywel Williams for assistance with osmotic theory and to Christine Grainger-Boultby for assistance with photography. This work forms part of a PhD thesis undertaken by Sarah Kujawinski and sponsored by Bristol-Myers Squibb.
References (42)
- et al.
pH-independent controlled-release microspheres using polyglycerol esters of fatty-acids
J. Pharm. Sci.
(1994) - et al.
Microenvironmental pH modulation in solid dosage forms
J. Pharm. Sci.
(2007) - et al.
Microstructural imaging of early gel layer formation in HPMC matrices
J. Pharm. Sci.
(2006) - et al.
Formulation, release characteristics and bioavailability of novel monolithic hydroxypropylmethylcellulose matrix tablets containing acetaminophen
J. Control. Rel.
(2005) - et al.
Microenvironmental pH control of drug dissolution
Int. J. Pharm.
(1989) - et al.
Influence of admired citric acid on the release profile of pelanserin hydrochloride from HPMC matrix tablets
Int. J. Pharm.
(2000) - et al.
Thermogelation of methylcellulose. Part 1. Molecular structures and processes
Carbohydr. Polym.
(1993) - et al.
Development of a multi particulate extended release formulation for ZK 811 752, a weakly basic drug
Int. J. Pharm.
(2005) - et al.
Mapping neutral microclimate pH in PLGA microspheres
J. Control. Rel.
(2005) - et al.
The influence of additives on the cloud point, disintegration and dissolution of hydroxypropylmethylcellulose gels and matrix tablets
Int. J. Pharm.
(1990)