- Email: [email protected]
Effect of UV light exposure on hydrophilic polymers used as drug release modulators in solid dosage forms
J. DRUG DEL. SCI. TECH., 15 (2) 151-157 2005
Effect of UV light exposure on hydrophilic polymers used as drug release modulators in solid dosage forms E. Ochoa Machiste1, L. Segale1, S. Conti1, E. Fasani2, A. Albini2, U. Conte1, L. Maggi1* Department of Pharmaceutical Chemistry, University of Pavia, Via Taramelli 12, 27100 Pavia, Italy 2 Department of Organic Chemistry. University of Pavia, Via Taramelli 10, 27100 Pavia, Italy *Correspondence: [email protected]
1
The effects of UV light on hydrophilic polymers used as drug release modulators in matrix-tablets are evaluated. Three different polymers were chosen for this study: hydroxypropylmethyl cellulose (HPMC), polyethylene oxide (PEO) and polyvinyl alcohol (PVA). Diltiazem hydrocloride, a well-known photostable molecule, was chosen as model drug. When matrix tablets prepared with PEO or with HPMC of higher viscosity grade were exposed to UV irradiation, displacements of the drug release profiles were observed compared to the dissolution profile of the non-irradiated samples. In the case of PVA tablets, the exposure to UV light did not modify the dissolution profile of the matrices. The polymers that showed to be affected by UV irradiation, were then subjected to viscosity measurements. The rheological analysis indicates that viscosities of solutions prepared with irradiated PEO were much lower compared to the solution prepared with non-irradiated polymer. The same results, but less evident, were found for HPMC. Keywords: Hydroxypropylmethyl cellulose – Polyethylene oxide – Polyvinyl alcohol – UV irradiation – Matrix tablets – Dissolution test – Viscosity.
primary process of polymer degradation. However, this was due to the action of peroxides formed in the photolysis of nitrite. PEO is a non-ionic hydrophilic polymer with several industrial and biomedical applications. In pharmaceutical field, PEO is used as drug release modulator in extended-release solid formulations [5]. Although this kind of polymer should not absorb low wavelength radiation, the oxygen bridges, present in the PEO chain, make it susceptible to light degradation [6]. The photoreactions of such polymer are induced by structural defects, impurities or additives [6]. In recent years, the photooxidative degradation of PEO in presence of transition metal salts was studied [7]. The results of the research showed that even in absence of the metal salts the polymer undergoes chemical and physical changes due to UV irradiation. PVA is a partially acetylated polymer, whose properties depend on the degree of polymerisation and on the percentage of free alcohol present in its chain. In the pharmaceutical field, PVA is used in topical formulations, particularly in ophthalmic products [8], it is also used in the preparation of microparticles, nanoparticles, hydrogels and matrix tablets for controlled release formulations [9]. PVA is negligibly photosensitive. The presence of a catalyst, like titanium dioxide, is necessary to produce photodegradation of the polymer [10]. Bravar et al. [11] mentioned that even prolonged exposure of PVA to UV irradiation did not produce alterations in the molecular structure of the polymer. Although the photostability of PVA has been previously demonstrated, it is used in this work to verified its photostability at the UV irradiation conditions used in this work. To evaluate the possible modifications that UV light could produce on the polymer ability to modulate drug release rate, some matrices containing a physical mixture of a soluble model drug and one of the three polymers were prepared. The matrices were subjected to various UV light exposure times, and were
High molecular weight HPMC, PEO and PVA are hydrophilic polymers well known for their ability to swell and to form a gel in aqueous medium. For these reasons these polymers are used for the preparation of extended-released matrix tablets. It is known that most polymers are liable to photodegradation to some degree and many studies on this topic are available in the literature. However, the effect that unwanted exposition to light may have on the pharmaceutical properties of modified released oral dosage forms has received little attention. Therefore, we decided to explore the modification on hydrophilic polymers used as drug release modulators under UV-A light, similar to that present in the environment, and following the UV irradiation conditions recommended by ICH guidelines [1]. As for previous literature, HPMC is a cellulose ether widely used in food and pharmaceutical industry. HPMC has many pharmaceutical applications as coating agent, emulsifier, viscosant and drug release modulator. HPMC is the first choice for the formulation of hydrophilic matrix systems, providing a reliable mechanism for the slow release of drugs from oral solid dosage forms. In a previous work [2], it has been shown that extendedrelease matrices prepared with HPMC and irradiated with gamma rays showed alterations in the drug release profiles compared to the non-irradiated matrix, and it was demonstrated that the molecular structure of HPMC could be modified by gamma rays. In another publication, extended-release matrix formulations containing HPMC were exposed to UV light: no differences in drug release rate were found in the irradiated formulation compared to non-irradiated formulation, even after prolonged exposure time [3]. A work published in 1972 [4] demonstrated that it was possible to induce light degradation in aqueous HPMC solutions by the addition of sodium nitrite and rheological analysis indicated that chain scission was the 151
J. DRUG DEL. SCI. TECH., 15 (2) 151-157 2005
Effect of UV light exposure on hydrophilic polymers used as drug release modulators in solid dosage forms E. Ochoa Machiste, L. Segale, S. Conti, E. Fasani, A. Albini, U. Conte, L. Maggi
then evaluated by means of a dissolution test. Diltiazem hydrocloride was chosen as model drug because it is photostable in the solid state [12]. At the same time, accelerated stability tests were conducted in a dark room at 45°C and 75% HR for six months to verify whether the matrix tablets were stable in absence of light, but under stressed conditions of temperature and humidity. Moreover, the chemical stability of diltiazem HCl in the matrices exposed to stress conditions (light, heat and humidity) was verified by a specific HPLC method able to separate the possible products of degradation of diltiazem hydrochloride (13). The rheological behaviour of the polymers before and after UV irradiation was evaluated by measuring the viscosity of aqueous polymer solutions prepared with the non-irradiated and irradiated polymer powders.
polyethylene oxide molecule, an additional experiment in absence of oxygen was conducted. DTZX2 matrix tablets were kept under vacuum in a UV transparent container and irradiated in the same conditions as above for 25 days. PVA and Methocel K4M powders were not tested.
4. Accelerated stability test
The matrices were stored in closed polyethylene bottles and maintained at 40°C and 75% of relative humidity in a standard testing atmosphere chamber protected from light, for 6 months.
5. Dissolution test
Non-irradiated and irradiated matrix tablets were tested using the USP dissolution apparatus 2 (paddle) at 100 rpm [14]. The dissolution medium was distilled water, at 37°C. The amount of drug released was assessed by UV detection at 236 nm (Spectracomp 602, Advanced Products, Milan, Italy). The determinations were made in triplicate (SD ≤ 3%).
I. MATERIALS AND METHODS 1. Materials
Hydroxypropylmethylcellulose (Methocel K4M η = 4,000 cP and Methocel K100M η = 100,000 cP) (K4M and K100M) were kindly donated by Colorcon, Orpington, United Kingdom. Polyethylene oxide (Polyox WSR N60K, molecular weight = 2,000,000 and Polyox WSR 303, molecular weight = 7,000,000) (X2 and X7) were supplied by Union Carbide, Danbury, CT, United States. Polyvinyl alcohol (Erkol W40-140: degree of hydrolysis of 88.7 mol%) (PVA) was obtained from Erkol-Acetex Chimica s.r.l., Milan, Italy. Diltiazem hydrocloride (DTZ) was supplied by Profarmaco S.p.A, Milan, Italy. Acetic acid glacial and triethylamine analytical grade were supplied by Carlo Erba Reagents, Milan, Italy. Methanol and acetonitrile for chromatographic analysis were HPLC grade and obtained from Carlo Erba Reagents, Milan, Italy. Water used for solutions and buffers was HPLC grade.
6. Drug content
The DTZ content of the tablets was assessed just after production, on tablets irradiated for 25 days and on the samples after 6 months of accelerated stability test. For the determination of the drug content an HPLC method able to separate the drug from its degradation products was used. The tablets were dissolved in distilled water, HPLC grade, and diluted with TEA acetate buffer (pH 4) to final volume. The TEA acetate buffer was prepared by adding acetic acid to 0.01 M triethylamine aqueous solution up to the desired pH value. The solution was filtered off and analysed using a Hypersil BDS C18 column and 70:30 TEA acetate buffer:acetonitrile mixture as eluant, with a flux of 1 ml/min. The drug was assessed at 240 nm with a UV detector. The determinations were made in triplicate.
2. Matrix tablets preparation
7. Viscosity measurements
DTZ matrices were prepared by simply mixing 63.5% of DTZ and 36.5% of HPMC or PEO (DTZK4M, DTZK100, DTZX2, DTZX7). For the preparation of matrices containing PVA, 63.5% of polymer and 36.5% of DTZ (DTZPVA) were used. The different mixtures were compressed with a single-punch tabletting machine equipped with flat punches of 9.5 mm in diameter (Kilian, Coln, Germany) thickness of 3.15 ± 0.05 mm for HPMC and PEO matrices, and 5.30 ± 0.05 mm for PVA matrices. All DTZ matrices contained a dose of 180 mg of drug.
The amount of polymer required was dispersed in distilled water with continuous stirring, further water was added to adjust the volume and stirred until obtaining a homogeneous solution. The concentration of the K100 solution was 3% w/w. For X7 and X2 solutions the concentrations were 3 and 4% w/w, respectively. All the samples were allowed to equilibrate overnight before the rheological evaluation. The viscosity of the solutions was determined with a rotational viscometer (Viscotester VT7 R, Haake, Karlsruhe, Germany) equipped with a recirculating water bath for temperature control (23°C). Rheological measurements were done in the range of 0.1-27 s-1 values of shear rate, in triplicate (SD ≤ 2%). The solutions prepared with PEO X2 irradiated for 12 and 25 days, and the solution prepared with 25-day-irradiated X7 powder could not be measured because their viscosities were too low and out of the range of the apparatus.
3. Irradiation conditions
The samples (polymer powders or matrix tablets) were irradiated in a dark photostability cabinet fitted with 2 x 20 W phosphor-coated lamps (λ = 366 nm). The intensity of incident light was measured by means of a calibrated radiometer and found to be 0.95 ± 0.1 mWcm2. The distance between the light source and the sample was 12 cm. Irradiated matrix tablets were reversed at half time and sampled at 20 h, 4, 12 and 25 days. Samples of pure K100M, X2 and X7 powders were spread in very thin layers (about 1 mm) in large transparent containers and irradiated for 20 h, 2, 4, 12 and 25 days. To evaluate the effect of oxygen on the possible photodecomposition of the
II. RESULTS AND DISCUSSION 1. Drug content
The drug content of the matrices prepared with HPMC or PEO was tested with a specific HPLC method. No significant differences were found in the overall amount of diltiazem content 152
Effect of UV light exposure on hydrophilic polymers used as drug release modulators in solid dosage forms E. Ochoa Machiste, L. Segale, S. Conti, E. Fasani, A. Albini, U. Conte, L. Maggi
In a previous work [3], we demonstrated that the dissolution profiles of DTZK100 irradiated for 20 h and 12 days are superimposable to the dissolution profile of the non-irradiated matrices, only for the 25-day-irradiated DTZK100 a small difference in the results of the dissolution test was detectable, in fact the curve is slightly shifted compared to the profile obtained from non-irradiated matrices. The displacement of the 25-dayirradiated K100 curve does not modify significantly the overall drug dissolution trend of the matrices evaluated (Figure 2). The dissolution test performed to PVA matrices demonstrated that the devices were able to modulate the drug release and that the exposure to UV light did not modify the drug release rate from the tablets (Figure 3). These results agree with results found in literature regarding the photostability of PVA. PEOʼs (X2 and X7) are less effective in reducing the release rate of DTZ compared to HPMC matrices (at the same polymer:drug ratio) (Figures 4 and 5). Moreover, we previously demonstrated [3] that PEO matrices showed a progressive decrease of their efficiency to properly control the delivery process and a burst effect could be evidenced. This is due to the limited penetration of UV light that affects only the surface layer of the polymer, resulting in an increase of the release rate
Table I - DTZ content in matrix tablets (percentage) in different stressed conditions. Matrix tablets
Non-irradiated
25-day-UVirradiated
Stability test (40°C, 75% RH)
DTZX2 DTZX7 DTZK4M DTZK100
97.6 ± 1.2 97.7 ± 3.3 101 ± 1.4 102 ± 0.1
95.7 ± 0.8 96.0 ± 1.0 99.7 ± 2.3 101 ± 0.4
95.8 ± 0.2 96.06 ± 0.9 98.5 ± 1.3 101 ± 0.9
of the matrices (Table I), confirming the stability of diltiazem in the solid state in the different stress conditions used in this work.
2. Dissolution tests
The matrices containing DTZ and K4M or K100 as retarding polymers, show the characteristic dissolution profile of hydrophilic swellable matrices (Figures 1 and 2). Both DTZM4 and DTZK100 matrices are able to modulate the release rate of the drug over a prolonged time at a nearly constant rate. When matrices containing K4M were exposed to the different irradiation times no changes in the drug release profile could be detected.
100
100
80
80
60
60
% released
% released
J. DRUG DEL. SCI. TECH., 15 (2) 151-157 2005
40 non-irradiated DTZK4M 20-hour-irradiated 12-day-irradiated
20
40 non-irradiated DTZPVA 20-hour-irradiated
20
12-day-irradiated
25-day-irradiated
0
0 0
120
240
360 time (min)
480
600
0
720
360
480
600
100
100
80
80
60
60
40 non-irradiated DTZK100 20-hour-irradiated 12-day-irradiated 25-day-irradiated 240
360
480 600 time (min)
960 1080
non-irradiated DTZX2 20-hour-irradiated 12-day-irradiated 25-day-irradiated
0
120
840
40
20
0 0
720
Figure 3 - Release profiles of non-irradiated and irradiated matrix tablets DTZPVA.
% released
% released
240
time (min)
Figure 1 - Release profiles of non-irradiated and irradiated matrix tablets DTZK4M.
20
120
720
840
960
0
60
120
180
240
300
360
time (min)
Figure 2 - Release profiles of non-irradiated and irradiated matrix tablets DTZK100.
Figure 4 - Release profiles of non-irradiated and irradiated matrix tablets DTZX2. 153
J. DRUG DEL. SCI. TECH., 15 (2) 151-157 2005
Effect of UV light exposure on hydrophilic polymers used as drug release modulators in solid dosage forms E. Ochoa Machiste, L. Segale, S. Conti, E. Fasani, A. Albini, U. Conte, L. Maggi
during the initial minutes of the dissolution test (burst effect) (Figure 6). The burst release obtained by UV irradiation causes a shift of the dissolution profile, so that complete dissolution is anticipated by ca. 2-3 h in the DTZX2 matrices irradiated for 25 days (Figure 4). Also in the case of the DTZX7 matrices, the exposure to UV light led to a slight increase in the release rate during the first few minutes of the dissolution test (Figure 6) but, in this case, the burst was less marked and the dissolution profile was characterized by a reduced shifting, compared to the non-irradiated matrices (Figure 5). Irradiated and non-irradiated matrices containing HPMC or PVA did not show any significant differences in the percentage of drug released after 30 min (Figure 6). The results of the dissolution tests performed on DTZX2 matrices irradiated under vacuum conditions (absence of oxygen) were compared to the dissolution behaviour of the 25-day-irradiated and non-irradiated DTZX2 tablets (Figure 7). The dissolution curve of the 25-day-irradiated matrices in presence of oxygen shows a remarkable burst effect (as described above) instead, the dissolution profile of the matrices exposed to the same irradiation conditions, but in absence of oxygen, is almost superimposable to the dissolution profile of nonirradiated tablets. These results afford clear evidence of the accelerating effect of oxygen on polymer chain cleavage. In fact, matrices irradiated under vacuum conditions did not lose their efficiency in drug release control, this result is an indirect evidence of the integrity of the molecular chain of the polymer (X2). Thus, oxygen participates in the reaction leading to the photoinitiated decomposition of PEO macromolecule. The chemistry occurring is probably hydrogen abstraction by photoexcited ketones, present as impurities, from CH2 groups activated by the vicinal oxygen atom. Since such impurity is in a low quantity, hydrogen abstraction and following chain cleavage by irradiation is limited. In the presence of oxygen, peroxy radicals are formed and propagate the radical reaction leading to more extensive degradation (Scheme 1). The effect is more marked with the linear chain of PEO, particularly when starting from short chain. Though chemically similar, more complex cyclic structure of HPMC undergoes a less dramatic degradation per a single hydrogen abstraction. For PVA and HPMC a corresponding fast photodegradation has not been reported. Presumably, for HPMC the H abstraction from the α-site is favoured, the hydroxyalkyl radical formed undergoes a loss of hydrogen and initiation of chain degradation is slower than photodegradation of PEO (Scheme 2). The rapid degradation of PEO under irradiation has been previously reported. Although the polymer is transparent to UV-A, formation of a complex with oxygen [15] or the presence of ketones as impurities [16, 17] makes photo-induced
100
% released
80
60
40 non-irradiated DTZX7 20-hour-irradiated
20
12-day-irradiated
0 0
60
120
180 240 time (min)
300
360
420
Figure 5 - Release profiles of non-irradiated and irradiated matrix tablets DTZX7. 60
% of drug released after 30 min
non-irradiated 50
20 hours
40
25 days
12 days
30 20 10 0 DTZK4M
DTZK100
DTZX2
DTZX7
DTZPVA
Figure 6 - Percentage of DTZ release from the matrices after 30 min. 100
80
% released
60
40 non-irradiated DTZX2
20
25-day-irradiated 25-day-irradiated under vacuum
0 0
60
120
180 time (min)
240
300
360
Figure 7 - Release profiles of non-irradiated, 25-day-irradiated, and 25-day-irradiated under vacuum DTZX2 matrix tablets.
Scheme 2 - Proposed mechanism of photo-induced HPMC degradation.
Scheme 1 - Proposed mechanism of photo-induced PEO degradation. 154
Effect of UV light exposure on hydrophilic polymers used as drug release modulators in solid dosage forms E. Ochoa Machiste, L. Segale, S. Conti, E. Fasani, A. Albini, U. Conte, L. Maggi
J. DRUG DEL. SCI. TECH., 15 (2) 151-157 2005
hydrogen abstraction possible. In the presence of oxygen, this leads to hydroperoxides that in turn undergo light induced O-O bond cleavage generating hydroxyl and alkoxyl radical [15] that are known to decompose the chain [18]. On the contrary irradiation under anaerobic conditions is effective only when a short wavelength is used [19].
200 180 160 140
τ (Pa)
120
3. Accelerated stability test
The dissolution tests carried out on the matrices before and after storage under stressed conditions (40°C, 75% RH, for 6 months) did not show any difference in the drug release rate of the devices and for all the polymers considered. In fact, Td50% (time needed to release 50% of the drug content) remains constant after six months in forced storage conditions of temperature and humidity (Table II). These results confirm that the modifications in the dissolution profiles are due only to UV light exposure.
td50%
td50% after stability test (40°C, 75% RH)
DTZX2 DTZX7 DTZK4M DTZK100
128 ± 11 157 ± 7.9 228 ± 19 277 ± 5.5
123 ± 3.7 150 ± 2.9 230 ± 22 271 ± 11
80
non-irradiated K100 20-hour-UV irradiated 4-day-UV irradiated 12-day-UV irradiated 25-day-UV irradiated
60 40 20 0 0
5
10
. γ (1/s)
15
20
25
Figure 8 - Flow curves of different aqueous solutions (2% w/w) prepared from non-irradiated and irradiated K100 HPMC powder.
All the aqueous solutions prepared with the samples of polyethylene oxide X2 and X7, previously irradiated at different exposure times, have shown a remarkable decrease in shear stress in comparison with their respective solutions prepared with non-irradiated polymer (Figures 9 and 10). X2 solutions show evident changes in rheological behaviour. Even solutions prepared with 20-hour-irradiated samples show a marked decrease in shear stress. The lowering of the shear thinning behaviour of the solutions is proportional to the UV exposure time (Figure 9). For the solutions prepared with 12- and 25-day-irradiated X2 powder, it was not possible to determine the flow curves because the shear stress values were too low and thus out of the range of the viscometer. The rheological changes found for X2 solutions confirm the results of the dissolution test performed to X2 matrices. In fact, matrices irradiated for 12 and 25 days are no more able to modulate the drug release at the beginning of the dissolution test. In fact, the polymer present at the tablet surface has been photodegradated and thus it can not form the gel layer responsible for drug release control. On the other hand, the photodecomposition of the polymer, suggested for the decrease of the shear stress found in the sample irradiated for 20 h, is not
Table II - Time needed to release 50% of drug content (td50%) before and after stability test. Matrix tablets
100
4. Viscosity
The effect of UV irradiation on flow properties of polymer solutions was evaluated by recording rheograms of the solutions prepared with polymers samples previously irradiated at different UV exposure time. The flow curves of all solutions studied showed the characteristic pseudoplastic behaviour of hydrophilic polymers in aqueous solutions (an increase in shear stress τ and thus a decrease in apparent viscosity η, when shear rate γ increases). Neither thixotropic nor yield values of shear stress were detected experimentally. The flow curves of the solutions prepared with 20-hour-irradiated and 4-day-irradiated K100 are almost superimposable to non-irradiated K100 solution (Figure 8). This indicates that the lower UV exposure times (20 h and 4 days) does not cause evident rheological changes in the solutions and thus in the molecular structure of the polymer. These results agree with the dissolution test performed to the matrix: no differences between the drug release profiles of non-irradiated matrix and 20-hour-irradiated matrix were found, whereas by prolonging the UV light exposure of K100, more evident changes in the flow curves are detected. In fact, it is possible to observe a remarkable decrease in the shear thinning behaviour of the solutions prepared with samples of polymer exposed for 12 and 25 days to UV light. However, the photodecomposition of the polymer, suggested for the decrease of the shear stress found in the sample irradiated for 12 h, seems not to be enough to produce detectable changes in the drug release profile of the matrices exposed at the same conditions, only after 25 days of UV irradiation, a slight difference in the dissolution profile of the matrices, compared to non-irradiated tablets, can be observed (Figure 2).
80
non-irradiated X2 20-hour-irradiated 2-day-irradiated 4-day-irradiated
τ (Pa)
60
40
20
0 0
3
6
9
. γ (1/s)
12
15
18
21
Figure 9 - Flow curves of different aqueous solutions (4% w/w) prepared from non-irradiated and irradiated X2 PEO powder.
155
J. DRUG DEL. SCI. TECH., 15 (2) 151-157 2005
Effect of UV light exposure on hydrophilic polymers used as drug release modulators in solid dosage forms E. Ochoa Machiste, L. Segale, S. Conti, E. Fasani, A. Albini, U. Conte, L. Maggi
enough to produce valuable changes in the drug release profile of the matrices exposed to the same conditions (a gel layer can be formed at the tablet surface). Aqueous solutions prepared with X7-irradiated samples show also flow curves with shear stress values lower than the values found for the solution prepared with non-irradiated polymer. In this case, the difference observed among the flow curves obtained from irradiated samples is less evident; except when the polymer was irradiated for 12 days, the flow curve showed very low values (Figure 10). The solutions prepared with 25-day-irradiated X7 showed very low viscosities and thus was not possible to determine the flow curve. The photodecomposition of the polymer after 12 and 25 days of UV irradiation is confirm also for the burst effect found in X7 matrices at the same exposure time (Figure 5). The effect of UV light on the rheological characteristics of the solutions is thus more evident for PEO of lower molecular weight (X2) than PEO of higher molecular weight (X7). These results confirm the changes on dissolution profiles found for PEO matrices exposed to UV irradiation; for DTZX2 matrices the burst effect was more marked than for DTZX7. The UV exposure of powder-polymer samples probably causes chain scissions, lowering the molecular weight of the polymers and thus decreasing the viscosity of the PEO solutions. Starting from higher molecular weigh (X7 for example) the chain fragments, resulting after UV irradiation, still maintain a molecular weight high enough to produce consistent values of viscosity. This phenomenon persists for a while, but when the sample is irradiated for 12 days, the polymer fragments are so small and their aqueous solution shows a drastic decrease in shear stress. In fact, for the solution prepared with 25-day-irradiated polymer was not possible to determine the flow curve, because the shear stress values were too low and out of the detection limit of the viscometer. The flow curves of aqueous polyethylene-oxide solutions show a clear evidence of the viscosity changes produced by UV light. The effect of the UV irradiation on the apparent viscosity of the polymer solutions (γ = 1.7 s-1) are reported in Figure 11. The apparent viscosity of X2 solutions undergoes a very drastic decrease, in fact, the solution prepared with the sample irradiated for 4 days shows a viscosity of less than 1 Pa.s, while for X7, the viscosity reaches values smaller than 1 Pa.s for the solutions prepared with 12-day-irradiated sample. Although it is not possible to directly compare the behaviour of the two polymers (X2 and X7) because the experiments were conducted at different polymer concentrations, it seems evident that the UV irradiation induces a more marked decrease in the viscosity of X2 solutions rather than in X7 solutions. The viscosity of a polymer solution is proportional to the length of the molecular chain of the polymer. X2 and X7 have identical chemical identity, except for the length of the molecular chain (X7 is longer than X2) so at identical irradiation dose the same chain damage for both polymer can be assumed, but, in the case of X2 the chain fragments are smaller than X7 fragments, therefore it is possible to expect that the viscosity of the X2 solutions is lower than X7 solution viscosity at the same irradiation dose. K100 seems to be more stable to UV light, the lowering of the viscosity due to UV light is less drastic, after 25 days of UV
120 105
τ (Pa)
90 75 60
non-irradiated X7 20-hour-irradiated 2-day-irradiated 4-day-irradiated 12-day-irradiated
45 30 15 0 0
5
10
15 . γ (1/s)
20
25
30
Figure 10 - Flow curves of different aqueous solutions (3% w/w) prepared from non-irradiated and irradiated X7 PEO powder.
75 X2 (4% w/w) X7 (3% w/w)
60
η (Pa.s)
K100 (2% w/w) 45
30
15
0 0
5
10
15
20
25
exposure time (days)
Figure 11 - Apparent viscosities (γ = 1.7 s-1) of PEO and HPMC solutions as a function of exposure time.
irradiation, the viscosity of K100 solution is only about three times lower than the viscosity of the solution prepared with non-irradiated polymer. Probably the viscosity decrease of the polymer solution observed for K100 is not so marked so as to produce valuable changes in the dissolution profiles of the matrices, in fact, the rheograms obtained for all the polymer solutions examined were fitted to a power law equation obtaining correlation coefficients (r) > 0.99 and good reproducibility in triplicate (Table III). Power law: τ = Kγn, where τ is the shear stress, γ the shear rate, n the flow index, and K the consistency index. Power law flow index (n) indicates the “pseudoplasticity grade” of a fluid: for n < 1, the flow is pseudoplastic, when n = 1, the flow pattern becomes Newtonian, and n > 1 means dilatant flows. Power law flow indexes were examined as a function of the irradiation time. For all polymers the increase of the UVexposure time induces a nearly linear increase of flow indexes (Figure 12) and thus a nearly linear decrease in pseudoplastic or thinning behaviour. The results indicate an inversely proportional relationship between the UV irradiation dose and the pseudoplastic characteristic of the solutions, therefore, an increase of UV exposure time causes a failure of shear thinning behaviour 156
Effect of UV light exposure on hydrophilic polymers used as drug release modulators in solid dosage forms E. Ochoa Machiste, L. Segale, S. Conti, E. Fasani, A. Albini, U. Conte, L. Maggi
J. DRUG DEL. SCI. TECH., 15 (2) 151-157 2005
Table III - Correlation parameters obtained from fitting flow curves to power law equation. Consistency index K ± SD
Flow index n ± SD
Correlation
Exposure
X2
X7
K100
X2
X7
K100
X2
X7
K100
0 20 h 2 days 4 days 12 days 25 days
38.5±3.4 17.3±2.9 6.08±0.9 1.03±0.5 -
98.3±0.5 50.7±11 28.0±2.3 21.4±2.3 0.40±0.06 -
49.1±0.02 52.0±0.7 43.6±0.1 31.4±1.2 16.4±1.8
0.57±0.006 0.66±0.03 0.75±0.02 0.86±0.03 -
0.26±0.02 0.41±0.05 0.43±0.006 0.47±0.02 0.89±0.04 -
0.53±0.0004 0.52±0.007 0.55±0.007 0.56±0.004 0.65±0.02
0.9981 0.9967 0.9995 0.9989 -
0.9927 0.9958 0.9948 0.9974 0.9954 -
0.9963 0.9964 0.9968 0.9963 0.9985 -
3. 1,0
4.
flow index (n)
0,8
5.
0,6
0,4
0,2
X2 (4% w/w)
6.
X7 (3% w/w)
7.
K100 (2% w/w) 0,0 0
5
10
15
20
25
8.
exposure time (days)
Figure 12 - Flow indexes (n) of PEO and HPMC solutions as a function of exposure time.
9.
of the polymer solution. This behaviour is more evident for PEO molecules than for HPMC, in fact, the solution prepared with 25-day-irradiated K100 shows a slight increase of flow index comparing to the flow index of the solution prepared with non-irradiated sample. The changes in rheological behaviour found for polymer solutions confirm that UV light may alter the chemical integrity of the functional excipients used to modulate the drug release rate in matrix tablets. The alterations of the drug release profile observed in matrix tablets depend on the exposure time and on the kind of polymer used. In fact, PEO seems to be more UV sensitive compared to HPMC macromolecule. Moreover, experiments conducted in absence of oxygen showed that PEO degradation is photoinitiated by the oxygen. PVA proved to be photostable at the experimental conditions used in this work. The chemical changes, induced by UV light in the polymers used as drug release modulators, should always be evaluated together to the photostability of the drug and the behaviour of the overall formulation, since any changes in the dissolution behaviour can cause the loss of the desired therapeutic effect.
10. 11. 12. 13. 14. 15. 16.
17.
18.
REFERENCES 1. 2.
19.
International Conference of Harmonization (ICH). - Guideliners for the Photostability testing of New Drug Substances and Products. - Federal Register, 62, 27115-27122, 1997. L. MAGGI, L. SEGALE, E. OCHOA MACHISTE, A. BUTTAFAVA, A. FAUCITANO, U. CONTE. - Chemical and physical stability of Hydroxypropylmethylcellulose matrices containing Diltiazem Hydrocloride after gamma irradiation. - J. Pharm. Sci., 92, 131141, 2003.
L. MAGGI, E. OCHOA MACHISTE, E. FASANI, A. ALBINI, L. SEGALE, U. CONTE. - Photostability of extended-release matrix formulations. - Eur. J. Pharm. Biopharm., 55, 99-105, 2003. D. H. DAVIES, G. D. DIXON. - Photodepolymerization of hydroxypropylmethylcellulose. - J. Applied Polymer Science, 16, 2449-2459, 1972. L. MAGGI, L. SEGALE, M.L. TORRE, E. OCHOA MACHISTE, U. CONTE. - Dissolution behaviour of hydrophilic matrix tablets containing two different polyethylene oxides (PEOs) for the controlled released of a water-soluble drug. Dimensionality study. - Biomaterials, 23, 113-119, 2002. J.F. RABEK. - Photodegradation of Polymers: Mechanisms and Experimental Methods. - Chapman and Hall, London, 1994. H. KACZMAREK, A. SIONKOWSKA, A. KAMINSKA, J. KOWALONEK. - The influence of transition metal salts on photo-oxidative degradation of poly(ethylene oxide). - Polymer Degradation and Stability, 73, 437-441, 2001. Handbook of Pharmaceutical Excipients. - Pharmaceutical Press and American Pharmaceutical Association, 4th ed., 2003, pp. 491-492. R. MORITA, R. HONDA, Y. TAKAHASHI. - Development of oral controlled release system (SCRS)1. Design of SCRS and its release controlling factor. - J. Control. Rel., 63, 297-304, 2000. Y. CHEN, Z. SUN, Y. YANG, Q. KE. - Heterogeneous photocatalytic oxidation of polyvinyl alcohol in water. - J. Photochemistry and Photobiology A: Chemistry, 142, 85-89, 2001. M. BRAVAR, V. REK, R. KOSTELLA-BIFFI. - J. Polymer Sciences, Symposium 40, 1973. M.S. SULEIMAN, M.E. ABDULHAMEED, N.M. NAJIB, H.Y.MUTI. - Effect of ultraviolet radiation on the stability of Diltiazem. - Int. J. Pharm., 50, 71-73, 1989. V. ANDRISANO, P. HRELIA, R. GOTTI, A. LEONI, V. CAVRINI. - Photostability and phototoxicity studies on diltiazem. - J. Pharm. Biomed. Analysis, 25, 589-597, 2001. The Official Compendia of Standards. - United States Pharmacopoeia Convention, Inc., Rockville, 2004, pp. 2303-2304. I. REIMAN, G. HOLLATZ, T. ECKERT. - Investigations of the light-induced autoxidation of polyethylene glycols. - Archive der Pharmazie, 307, 321-327, 1974. S. SLOOP, M. LERNER, T. STEPHENS, A. TIPTON, D. PAULL, J. STENGER-SMITH. - Cross-linking poly(ethylene-oxide) and poly[oxymethylene-oligo(oxyethylene)] with ultraviolet radiation. - J. Applied Polymer Science, 53, 1563-1572, 1994. M. DOYTCHEVA, D. DOTCHEVA, R. STAMENOVA, A. ORAHOVATS, C.TSVETANOV, LEDER - Ultraviolet-induced cross-linking of solid poly(ethylene-oxide). - J. Applied Polymer Science, 64, 2299-2307, 1997. H. KACZMAREK, L. A. LINDEN, J.F. RABEK. - Reactions of hydroxyl (HO*) and hydroperoxyl (HO2*) radicals generated chemically and photochemically with poly(ethylene-oxide). - J. Polymer Science, Part A: Polymer Chemistry, 33, 879-890, 1995. T.V. IVANOVA, M.Y. MEL’NIKOV, N.V. FOK. - Photochemical reactions of radicals in polyethylene oxide at 77°K and their role in degradation processes. - Doklady Akademii Nauk SSSR [Phys. Chem.], 231, 649-652, 1976.
MANUSCRIPT Received 7 April 2004, accepted for publication 13 September 2004. 157
Effect of UV light exposure on hydrophilic polymers used as drug release modulators in solid dosage forms E. Ochoa Machiste1, L. Segale1, S. Conti1, E. Fasani2, A. Albini2, U. Conte1, L. Maggi1* Department of Pharmaceutical Chemistry, University of Pavia, Via Taramelli 12, 27100 Pavia, Italy 2 Department of Organic Chemistry. University of Pavia, Via Taramelli 10, 27100 Pavia, Italy *Correspondence: [email protected]
1
The effects of UV light on hydrophilic polymers used as drug release modulators in matrix-tablets are evaluated. Three different polymers were chosen for this study: hydroxypropylmethyl cellulose (HPMC), polyethylene oxide (PEO) and polyvinyl alcohol (PVA). Diltiazem hydrocloride, a well-known photostable molecule, was chosen as model drug. When matrix tablets prepared with PEO or with HPMC of higher viscosity grade were exposed to UV irradiation, displacements of the drug release profiles were observed compared to the dissolution profile of the non-irradiated samples. In the case of PVA tablets, the exposure to UV light did not modify the dissolution profile of the matrices. The polymers that showed to be affected by UV irradiation, were then subjected to viscosity measurements. The rheological analysis indicates that viscosities of solutions prepared with irradiated PEO were much lower compared to the solution prepared with non-irradiated polymer. The same results, but less evident, were found for HPMC. Keywords: Hydroxypropylmethyl cellulose – Polyethylene oxide – Polyvinyl alcohol – UV irradiation – Matrix tablets – Dissolution test – Viscosity.
primary process of polymer degradation. However, this was due to the action of peroxides formed in the photolysis of nitrite. PEO is a non-ionic hydrophilic polymer with several industrial and biomedical applications. In pharmaceutical field, PEO is used as drug release modulator in extended-release solid formulations [5]. Although this kind of polymer should not absorb low wavelength radiation, the oxygen bridges, present in the PEO chain, make it susceptible to light degradation [6]. The photoreactions of such polymer are induced by structural defects, impurities or additives [6]. In recent years, the photooxidative degradation of PEO in presence of transition metal salts was studied [7]. The results of the research showed that even in absence of the metal salts the polymer undergoes chemical and physical changes due to UV irradiation. PVA is a partially acetylated polymer, whose properties depend on the degree of polymerisation and on the percentage of free alcohol present in its chain. In the pharmaceutical field, PVA is used in topical formulations, particularly in ophthalmic products [8], it is also used in the preparation of microparticles, nanoparticles, hydrogels and matrix tablets for controlled release formulations [9]. PVA is negligibly photosensitive. The presence of a catalyst, like titanium dioxide, is necessary to produce photodegradation of the polymer [10]. Bravar et al. [11] mentioned that even prolonged exposure of PVA to UV irradiation did not produce alterations in the molecular structure of the polymer. Although the photostability of PVA has been previously demonstrated, it is used in this work to verified its photostability at the UV irradiation conditions used in this work. To evaluate the possible modifications that UV light could produce on the polymer ability to modulate drug release rate, some matrices containing a physical mixture of a soluble model drug and one of the three polymers were prepared. The matrices were subjected to various UV light exposure times, and were
High molecular weight HPMC, PEO and PVA are hydrophilic polymers well known for their ability to swell and to form a gel in aqueous medium. For these reasons these polymers are used for the preparation of extended-released matrix tablets. It is known that most polymers are liable to photodegradation to some degree and many studies on this topic are available in the literature. However, the effect that unwanted exposition to light may have on the pharmaceutical properties of modified released oral dosage forms has received little attention. Therefore, we decided to explore the modification on hydrophilic polymers used as drug release modulators under UV-A light, similar to that present in the environment, and following the UV irradiation conditions recommended by ICH guidelines [1]. As for previous literature, HPMC is a cellulose ether widely used in food and pharmaceutical industry. HPMC has many pharmaceutical applications as coating agent, emulsifier, viscosant and drug release modulator. HPMC is the first choice for the formulation of hydrophilic matrix systems, providing a reliable mechanism for the slow release of drugs from oral solid dosage forms. In a previous work [2], it has been shown that extendedrelease matrices prepared with HPMC and irradiated with gamma rays showed alterations in the drug release profiles compared to the non-irradiated matrix, and it was demonstrated that the molecular structure of HPMC could be modified by gamma rays. In another publication, extended-release matrix formulations containing HPMC were exposed to UV light: no differences in drug release rate were found in the irradiated formulation compared to non-irradiated formulation, even after prolonged exposure time [3]. A work published in 1972 [4] demonstrated that it was possible to induce light degradation in aqueous HPMC solutions by the addition of sodium nitrite and rheological analysis indicated that chain scission was the 151
J. DRUG DEL. SCI. TECH., 15 (2) 151-157 2005
Effect of UV light exposure on hydrophilic polymers used as drug release modulators in solid dosage forms E. Ochoa Machiste, L. Segale, S. Conti, E. Fasani, A. Albini, U. Conte, L. Maggi
then evaluated by means of a dissolution test. Diltiazem hydrocloride was chosen as model drug because it is photostable in the solid state [12]. At the same time, accelerated stability tests were conducted in a dark room at 45°C and 75% HR for six months to verify whether the matrix tablets were stable in absence of light, but under stressed conditions of temperature and humidity. Moreover, the chemical stability of diltiazem HCl in the matrices exposed to stress conditions (light, heat and humidity) was verified by a specific HPLC method able to separate the possible products of degradation of diltiazem hydrochloride (13). The rheological behaviour of the polymers before and after UV irradiation was evaluated by measuring the viscosity of aqueous polymer solutions prepared with the non-irradiated and irradiated polymer powders.
polyethylene oxide molecule, an additional experiment in absence of oxygen was conducted. DTZX2 matrix tablets were kept under vacuum in a UV transparent container and irradiated in the same conditions as above for 25 days. PVA and Methocel K4M powders were not tested.
4. Accelerated stability test
The matrices were stored in closed polyethylene bottles and maintained at 40°C and 75% of relative humidity in a standard testing atmosphere chamber protected from light, for 6 months.
5. Dissolution test
Non-irradiated and irradiated matrix tablets were tested using the USP dissolution apparatus 2 (paddle) at 100 rpm [14]. The dissolution medium was distilled water, at 37°C. The amount of drug released was assessed by UV detection at 236 nm (Spectracomp 602, Advanced Products, Milan, Italy). The determinations were made in triplicate (SD ≤ 3%).
I. MATERIALS AND METHODS 1. Materials
Hydroxypropylmethylcellulose (Methocel K4M η = 4,000 cP and Methocel K100M η = 100,000 cP) (K4M and K100M) were kindly donated by Colorcon, Orpington, United Kingdom. Polyethylene oxide (Polyox WSR N60K, molecular weight = 2,000,000 and Polyox WSR 303, molecular weight = 7,000,000) (X2 and X7) were supplied by Union Carbide, Danbury, CT, United States. Polyvinyl alcohol (Erkol W40-140: degree of hydrolysis of 88.7 mol%) (PVA) was obtained from Erkol-Acetex Chimica s.r.l., Milan, Italy. Diltiazem hydrocloride (DTZ) was supplied by Profarmaco S.p.A, Milan, Italy. Acetic acid glacial and triethylamine analytical grade were supplied by Carlo Erba Reagents, Milan, Italy. Methanol and acetonitrile for chromatographic analysis were HPLC grade and obtained from Carlo Erba Reagents, Milan, Italy. Water used for solutions and buffers was HPLC grade.
6. Drug content
The DTZ content of the tablets was assessed just after production, on tablets irradiated for 25 days and on the samples after 6 months of accelerated stability test. For the determination of the drug content an HPLC method able to separate the drug from its degradation products was used. The tablets were dissolved in distilled water, HPLC grade, and diluted with TEA acetate buffer (pH 4) to final volume. The TEA acetate buffer was prepared by adding acetic acid to 0.01 M triethylamine aqueous solution up to the desired pH value. The solution was filtered off and analysed using a Hypersil BDS C18 column and 70:30 TEA acetate buffer:acetonitrile mixture as eluant, with a flux of 1 ml/min. The drug was assessed at 240 nm with a UV detector. The determinations were made in triplicate.
2. Matrix tablets preparation
7. Viscosity measurements
DTZ matrices were prepared by simply mixing 63.5% of DTZ and 36.5% of HPMC or PEO (DTZK4M, DTZK100, DTZX2, DTZX7). For the preparation of matrices containing PVA, 63.5% of polymer and 36.5% of DTZ (DTZPVA) were used. The different mixtures were compressed with a single-punch tabletting machine equipped with flat punches of 9.5 mm in diameter (Kilian, Coln, Germany) thickness of 3.15 ± 0.05 mm for HPMC and PEO matrices, and 5.30 ± 0.05 mm for PVA matrices. All DTZ matrices contained a dose of 180 mg of drug.
The amount of polymer required was dispersed in distilled water with continuous stirring, further water was added to adjust the volume and stirred until obtaining a homogeneous solution. The concentration of the K100 solution was 3% w/w. For X7 and X2 solutions the concentrations were 3 and 4% w/w, respectively. All the samples were allowed to equilibrate overnight before the rheological evaluation. The viscosity of the solutions was determined with a rotational viscometer (Viscotester VT7 R, Haake, Karlsruhe, Germany) equipped with a recirculating water bath for temperature control (23°C). Rheological measurements were done in the range of 0.1-27 s-1 values of shear rate, in triplicate (SD ≤ 2%). The solutions prepared with PEO X2 irradiated for 12 and 25 days, and the solution prepared with 25-day-irradiated X7 powder could not be measured because their viscosities were too low and out of the range of the apparatus.
3. Irradiation conditions
The samples (polymer powders or matrix tablets) were irradiated in a dark photostability cabinet fitted with 2 x 20 W phosphor-coated lamps (λ = 366 nm). The intensity of incident light was measured by means of a calibrated radiometer and found to be 0.95 ± 0.1 mWcm2. The distance between the light source and the sample was 12 cm. Irradiated matrix tablets were reversed at half time and sampled at 20 h, 4, 12 and 25 days. Samples of pure K100M, X2 and X7 powders were spread in very thin layers (about 1 mm) in large transparent containers and irradiated for 20 h, 2, 4, 12 and 25 days. To evaluate the effect of oxygen on the possible photodecomposition of the
II. RESULTS AND DISCUSSION 1. Drug content
The drug content of the matrices prepared with HPMC or PEO was tested with a specific HPLC method. No significant differences were found in the overall amount of diltiazem content 152
Effect of UV light exposure on hydrophilic polymers used as drug release modulators in solid dosage forms E. Ochoa Machiste, L. Segale, S. Conti, E. Fasani, A. Albini, U. Conte, L. Maggi
In a previous work [3], we demonstrated that the dissolution profiles of DTZK100 irradiated for 20 h and 12 days are superimposable to the dissolution profile of the non-irradiated matrices, only for the 25-day-irradiated DTZK100 a small difference in the results of the dissolution test was detectable, in fact the curve is slightly shifted compared to the profile obtained from non-irradiated matrices. The displacement of the 25-dayirradiated K100 curve does not modify significantly the overall drug dissolution trend of the matrices evaluated (Figure 2). The dissolution test performed to PVA matrices demonstrated that the devices were able to modulate the drug release and that the exposure to UV light did not modify the drug release rate from the tablets (Figure 3). These results agree with results found in literature regarding the photostability of PVA. PEOʼs (X2 and X7) are less effective in reducing the release rate of DTZ compared to HPMC matrices (at the same polymer:drug ratio) (Figures 4 and 5). Moreover, we previously demonstrated [3] that PEO matrices showed a progressive decrease of their efficiency to properly control the delivery process and a burst effect could be evidenced. This is due to the limited penetration of UV light that affects only the surface layer of the polymer, resulting in an increase of the release rate
Table I - DTZ content in matrix tablets (percentage) in different stressed conditions. Matrix tablets
Non-irradiated
25-day-UVirradiated
Stability test (40°C, 75% RH)
DTZX2 DTZX7 DTZK4M DTZK100
97.6 ± 1.2 97.7 ± 3.3 101 ± 1.4 102 ± 0.1
95.7 ± 0.8 96.0 ± 1.0 99.7 ± 2.3 101 ± 0.4
95.8 ± 0.2 96.06 ± 0.9 98.5 ± 1.3 101 ± 0.9
of the matrices (Table I), confirming the stability of diltiazem in the solid state in the different stress conditions used in this work.
2. Dissolution tests
The matrices containing DTZ and K4M or K100 as retarding polymers, show the characteristic dissolution profile of hydrophilic swellable matrices (Figures 1 and 2). Both DTZM4 and DTZK100 matrices are able to modulate the release rate of the drug over a prolonged time at a nearly constant rate. When matrices containing K4M were exposed to the different irradiation times no changes in the drug release profile could be detected.
100
100
80
80
60
60
% released
% released
J. DRUG DEL. SCI. TECH., 15 (2) 151-157 2005
40 non-irradiated DTZK4M 20-hour-irradiated 12-day-irradiated
20
40 non-irradiated DTZPVA 20-hour-irradiated
20
12-day-irradiated
25-day-irradiated
0
0 0
120
240
360 time (min)
480
600
0
720
360
480
600
100
100
80
80
60
60
40 non-irradiated DTZK100 20-hour-irradiated 12-day-irradiated 25-day-irradiated 240
360
480 600 time (min)
960 1080
non-irradiated DTZX2 20-hour-irradiated 12-day-irradiated 25-day-irradiated
0
120
840
40
20
0 0
720
Figure 3 - Release profiles of non-irradiated and irradiated matrix tablets DTZPVA.
% released
% released
240
time (min)
Figure 1 - Release profiles of non-irradiated and irradiated matrix tablets DTZK4M.
20
120
720
840
960
0
60
120
180
240
300
360
time (min)
Figure 2 - Release profiles of non-irradiated and irradiated matrix tablets DTZK100.
Figure 4 - Release profiles of non-irradiated and irradiated matrix tablets DTZX2. 153
J. DRUG DEL. SCI. TECH., 15 (2) 151-157 2005
Effect of UV light exposure on hydrophilic polymers used as drug release modulators in solid dosage forms E. Ochoa Machiste, L. Segale, S. Conti, E. Fasani, A. Albini, U. Conte, L. Maggi
during the initial minutes of the dissolution test (burst effect) (Figure 6). The burst release obtained by UV irradiation causes a shift of the dissolution profile, so that complete dissolution is anticipated by ca. 2-3 h in the DTZX2 matrices irradiated for 25 days (Figure 4). Also in the case of the DTZX7 matrices, the exposure to UV light led to a slight increase in the release rate during the first few minutes of the dissolution test (Figure 6) but, in this case, the burst was less marked and the dissolution profile was characterized by a reduced shifting, compared to the non-irradiated matrices (Figure 5). Irradiated and non-irradiated matrices containing HPMC or PVA did not show any significant differences in the percentage of drug released after 30 min (Figure 6). The results of the dissolution tests performed on DTZX2 matrices irradiated under vacuum conditions (absence of oxygen) were compared to the dissolution behaviour of the 25-day-irradiated and non-irradiated DTZX2 tablets (Figure 7). The dissolution curve of the 25-day-irradiated matrices in presence of oxygen shows a remarkable burst effect (as described above) instead, the dissolution profile of the matrices exposed to the same irradiation conditions, but in absence of oxygen, is almost superimposable to the dissolution profile of nonirradiated tablets. These results afford clear evidence of the accelerating effect of oxygen on polymer chain cleavage. In fact, matrices irradiated under vacuum conditions did not lose their efficiency in drug release control, this result is an indirect evidence of the integrity of the molecular chain of the polymer (X2). Thus, oxygen participates in the reaction leading to the photoinitiated decomposition of PEO macromolecule. The chemistry occurring is probably hydrogen abstraction by photoexcited ketones, present as impurities, from CH2 groups activated by the vicinal oxygen atom. Since such impurity is in a low quantity, hydrogen abstraction and following chain cleavage by irradiation is limited. In the presence of oxygen, peroxy radicals are formed and propagate the radical reaction leading to more extensive degradation (Scheme 1). The effect is more marked with the linear chain of PEO, particularly when starting from short chain. Though chemically similar, more complex cyclic structure of HPMC undergoes a less dramatic degradation per a single hydrogen abstraction. For PVA and HPMC a corresponding fast photodegradation has not been reported. Presumably, for HPMC the H abstraction from the α-site is favoured, the hydroxyalkyl radical formed undergoes a loss of hydrogen and initiation of chain degradation is slower than photodegradation of PEO (Scheme 2). The rapid degradation of PEO under irradiation has been previously reported. Although the polymer is transparent to UV-A, formation of a complex with oxygen [15] or the presence of ketones as impurities [16, 17] makes photo-induced
100
% released
80
60
40 non-irradiated DTZX7 20-hour-irradiated
20
12-day-irradiated
0 0
60
120
180 240 time (min)
300
360
420
Figure 5 - Release profiles of non-irradiated and irradiated matrix tablets DTZX7. 60
% of drug released after 30 min
non-irradiated 50
20 hours
40
25 days
12 days
30 20 10 0 DTZK4M
DTZK100
DTZX2
DTZX7
DTZPVA
Figure 6 - Percentage of DTZ release from the matrices after 30 min. 100
80
% released
60
40 non-irradiated DTZX2
20
25-day-irradiated 25-day-irradiated under vacuum
0 0
60
120
180 time (min)
240
300
360
Figure 7 - Release profiles of non-irradiated, 25-day-irradiated, and 25-day-irradiated under vacuum DTZX2 matrix tablets.
Scheme 2 - Proposed mechanism of photo-induced HPMC degradation.
Scheme 1 - Proposed mechanism of photo-induced PEO degradation. 154
Effect of UV light exposure on hydrophilic polymers used as drug release modulators in solid dosage forms E. Ochoa Machiste, L. Segale, S. Conti, E. Fasani, A. Albini, U. Conte, L. Maggi
J. DRUG DEL. SCI. TECH., 15 (2) 151-157 2005
hydrogen abstraction possible. In the presence of oxygen, this leads to hydroperoxides that in turn undergo light induced O-O bond cleavage generating hydroxyl and alkoxyl radical [15] that are known to decompose the chain [18]. On the contrary irradiation under anaerobic conditions is effective only when a short wavelength is used [19].
200 180 160 140
τ (Pa)
120
3. Accelerated stability test
The dissolution tests carried out on the matrices before and after storage under stressed conditions (40°C, 75% RH, for 6 months) did not show any difference in the drug release rate of the devices and for all the polymers considered. In fact, Td50% (time needed to release 50% of the drug content) remains constant after six months in forced storage conditions of temperature and humidity (Table II). These results confirm that the modifications in the dissolution profiles are due only to UV light exposure.
td50%
td50% after stability test (40°C, 75% RH)
DTZX2 DTZX7 DTZK4M DTZK100
128 ± 11 157 ± 7.9 228 ± 19 277 ± 5.5
123 ± 3.7 150 ± 2.9 230 ± 22 271 ± 11
80
non-irradiated K100 20-hour-UV irradiated 4-day-UV irradiated 12-day-UV irradiated 25-day-UV irradiated
60 40 20 0 0
5
10
. γ (1/s)
15
20
25
Figure 8 - Flow curves of different aqueous solutions (2% w/w) prepared from non-irradiated and irradiated K100 HPMC powder.
All the aqueous solutions prepared with the samples of polyethylene oxide X2 and X7, previously irradiated at different exposure times, have shown a remarkable decrease in shear stress in comparison with their respective solutions prepared with non-irradiated polymer (Figures 9 and 10). X2 solutions show evident changes in rheological behaviour. Even solutions prepared with 20-hour-irradiated samples show a marked decrease in shear stress. The lowering of the shear thinning behaviour of the solutions is proportional to the UV exposure time (Figure 9). For the solutions prepared with 12- and 25-day-irradiated X2 powder, it was not possible to determine the flow curves because the shear stress values were too low and thus out of the range of the viscometer. The rheological changes found for X2 solutions confirm the results of the dissolution test performed to X2 matrices. In fact, matrices irradiated for 12 and 25 days are no more able to modulate the drug release at the beginning of the dissolution test. In fact, the polymer present at the tablet surface has been photodegradated and thus it can not form the gel layer responsible for drug release control. On the other hand, the photodecomposition of the polymer, suggested for the decrease of the shear stress found in the sample irradiated for 20 h, is not
Table II - Time needed to release 50% of drug content (td50%) before and after stability test. Matrix tablets
100
4. Viscosity
The effect of UV irradiation on flow properties of polymer solutions was evaluated by recording rheograms of the solutions prepared with polymers samples previously irradiated at different UV exposure time. The flow curves of all solutions studied showed the characteristic pseudoplastic behaviour of hydrophilic polymers in aqueous solutions (an increase in shear stress τ and thus a decrease in apparent viscosity η, when shear rate γ increases). Neither thixotropic nor yield values of shear stress were detected experimentally. The flow curves of the solutions prepared with 20-hour-irradiated and 4-day-irradiated K100 are almost superimposable to non-irradiated K100 solution (Figure 8). This indicates that the lower UV exposure times (20 h and 4 days) does not cause evident rheological changes in the solutions and thus in the molecular structure of the polymer. These results agree with the dissolution test performed to the matrix: no differences between the drug release profiles of non-irradiated matrix and 20-hour-irradiated matrix were found, whereas by prolonging the UV light exposure of K100, more evident changes in the flow curves are detected. In fact, it is possible to observe a remarkable decrease in the shear thinning behaviour of the solutions prepared with samples of polymer exposed for 12 and 25 days to UV light. However, the photodecomposition of the polymer, suggested for the decrease of the shear stress found in the sample irradiated for 12 h, seems not to be enough to produce detectable changes in the drug release profile of the matrices exposed at the same conditions, only after 25 days of UV irradiation, a slight difference in the dissolution profile of the matrices, compared to non-irradiated tablets, can be observed (Figure 2).
80
non-irradiated X2 20-hour-irradiated 2-day-irradiated 4-day-irradiated
τ (Pa)
60
40
20
0 0
3
6
9
. γ (1/s)
12
15
18
21
Figure 9 - Flow curves of different aqueous solutions (4% w/w) prepared from non-irradiated and irradiated X2 PEO powder.
155
J. DRUG DEL. SCI. TECH., 15 (2) 151-157 2005
Effect of UV light exposure on hydrophilic polymers used as drug release modulators in solid dosage forms E. Ochoa Machiste, L. Segale, S. Conti, E. Fasani, A. Albini, U. Conte, L. Maggi
enough to produce valuable changes in the drug release profile of the matrices exposed to the same conditions (a gel layer can be formed at the tablet surface). Aqueous solutions prepared with X7-irradiated samples show also flow curves with shear stress values lower than the values found for the solution prepared with non-irradiated polymer. In this case, the difference observed among the flow curves obtained from irradiated samples is less evident; except when the polymer was irradiated for 12 days, the flow curve showed very low values (Figure 10). The solutions prepared with 25-day-irradiated X7 showed very low viscosities and thus was not possible to determine the flow curve. The photodecomposition of the polymer after 12 and 25 days of UV irradiation is confirm also for the burst effect found in X7 matrices at the same exposure time (Figure 5). The effect of UV light on the rheological characteristics of the solutions is thus more evident for PEO of lower molecular weight (X2) than PEO of higher molecular weight (X7). These results confirm the changes on dissolution profiles found for PEO matrices exposed to UV irradiation; for DTZX2 matrices the burst effect was more marked than for DTZX7. The UV exposure of powder-polymer samples probably causes chain scissions, lowering the molecular weight of the polymers and thus decreasing the viscosity of the PEO solutions. Starting from higher molecular weigh (X7 for example) the chain fragments, resulting after UV irradiation, still maintain a molecular weight high enough to produce consistent values of viscosity. This phenomenon persists for a while, but when the sample is irradiated for 12 days, the polymer fragments are so small and their aqueous solution shows a drastic decrease in shear stress. In fact, for the solution prepared with 25-day-irradiated polymer was not possible to determine the flow curve, because the shear stress values were too low and out of the detection limit of the viscometer. The flow curves of aqueous polyethylene-oxide solutions show a clear evidence of the viscosity changes produced by UV light. The effect of the UV irradiation on the apparent viscosity of the polymer solutions (γ = 1.7 s-1) are reported in Figure 11. The apparent viscosity of X2 solutions undergoes a very drastic decrease, in fact, the solution prepared with the sample irradiated for 4 days shows a viscosity of less than 1 Pa.s, while for X7, the viscosity reaches values smaller than 1 Pa.s for the solutions prepared with 12-day-irradiated sample. Although it is not possible to directly compare the behaviour of the two polymers (X2 and X7) because the experiments were conducted at different polymer concentrations, it seems evident that the UV irradiation induces a more marked decrease in the viscosity of X2 solutions rather than in X7 solutions. The viscosity of a polymer solution is proportional to the length of the molecular chain of the polymer. X2 and X7 have identical chemical identity, except for the length of the molecular chain (X7 is longer than X2) so at identical irradiation dose the same chain damage for both polymer can be assumed, but, in the case of X2 the chain fragments are smaller than X7 fragments, therefore it is possible to expect that the viscosity of the X2 solutions is lower than X7 solution viscosity at the same irradiation dose. K100 seems to be more stable to UV light, the lowering of the viscosity due to UV light is less drastic, after 25 days of UV
120 105
τ (Pa)
90 75 60
non-irradiated X7 20-hour-irradiated 2-day-irradiated 4-day-irradiated 12-day-irradiated
45 30 15 0 0
5
10
15 . γ (1/s)
20
25
30
Figure 10 - Flow curves of different aqueous solutions (3% w/w) prepared from non-irradiated and irradiated X7 PEO powder.
75 X2 (4% w/w) X7 (3% w/w)
60
η (Pa.s)
K100 (2% w/w) 45
30
15
0 0
5
10
15
20
25
exposure time (days)
Figure 11 - Apparent viscosities (γ = 1.7 s-1) of PEO and HPMC solutions as a function of exposure time.
irradiation, the viscosity of K100 solution is only about three times lower than the viscosity of the solution prepared with non-irradiated polymer. Probably the viscosity decrease of the polymer solution observed for K100 is not so marked so as to produce valuable changes in the dissolution profiles of the matrices, in fact, the rheograms obtained for all the polymer solutions examined were fitted to a power law equation obtaining correlation coefficients (r) > 0.99 and good reproducibility in triplicate (Table III). Power law: τ = Kγn, where τ is the shear stress, γ the shear rate, n the flow index, and K the consistency index. Power law flow index (n) indicates the “pseudoplasticity grade” of a fluid: for n < 1, the flow is pseudoplastic, when n = 1, the flow pattern becomes Newtonian, and n > 1 means dilatant flows. Power law flow indexes were examined as a function of the irradiation time. For all polymers the increase of the UVexposure time induces a nearly linear increase of flow indexes (Figure 12) and thus a nearly linear decrease in pseudoplastic or thinning behaviour. The results indicate an inversely proportional relationship between the UV irradiation dose and the pseudoplastic characteristic of the solutions, therefore, an increase of UV exposure time causes a failure of shear thinning behaviour 156
Effect of UV light exposure on hydrophilic polymers used as drug release modulators in solid dosage forms E. Ochoa Machiste, L. Segale, S. Conti, E. Fasani, A. Albini, U. Conte, L. Maggi
J. DRUG DEL. SCI. TECH., 15 (2) 151-157 2005
Table III - Correlation parameters obtained from fitting flow curves to power law equation. Consistency index K ± SD
Flow index n ± SD
Correlation
Exposure
X2
X7
K100
X2
X7
K100
X2
X7
K100
0 20 h 2 days 4 days 12 days 25 days
38.5±3.4 17.3±2.9 6.08±0.9 1.03±0.5 -
98.3±0.5 50.7±11 28.0±2.3 21.4±2.3 0.40±0.06 -
49.1±0.02 52.0±0.7 43.6±0.1 31.4±1.2 16.4±1.8
0.57±0.006 0.66±0.03 0.75±0.02 0.86±0.03 -
0.26±0.02 0.41±0.05 0.43±0.006 0.47±0.02 0.89±0.04 -
0.53±0.0004 0.52±0.007 0.55±0.007 0.56±0.004 0.65±0.02
0.9981 0.9967 0.9995 0.9989 -
0.9927 0.9958 0.9948 0.9974 0.9954 -
0.9963 0.9964 0.9968 0.9963 0.9985 -
3. 1,0
4.
flow index (n)
0,8
5.
0,6
0,4
0,2
X2 (4% w/w)
6.
X7 (3% w/w)
7.
K100 (2% w/w) 0,0 0
5
10
15
20
25
8.
exposure time (days)
Figure 12 - Flow indexes (n) of PEO and HPMC solutions as a function of exposure time.
9.
of the polymer solution. This behaviour is more evident for PEO molecules than for HPMC, in fact, the solution prepared with 25-day-irradiated K100 shows a slight increase of flow index comparing to the flow index of the solution prepared with non-irradiated sample. The changes in rheological behaviour found for polymer solutions confirm that UV light may alter the chemical integrity of the functional excipients used to modulate the drug release rate in matrix tablets. The alterations of the drug release profile observed in matrix tablets depend on the exposure time and on the kind of polymer used. In fact, PEO seems to be more UV sensitive compared to HPMC macromolecule. Moreover, experiments conducted in absence of oxygen showed that PEO degradation is photoinitiated by the oxygen. PVA proved to be photostable at the experimental conditions used in this work. The chemical changes, induced by UV light in the polymers used as drug release modulators, should always be evaluated together to the photostability of the drug and the behaviour of the overall formulation, since any changes in the dissolution behaviour can cause the loss of the desired therapeutic effect.
10. 11. 12. 13. 14. 15. 16.
17.
18.
REFERENCES 1. 2.
19.
International Conference of Harmonization (ICH). - Guideliners for the Photostability testing of New Drug Substances and Products. - Federal Register, 62, 27115-27122, 1997. L. MAGGI, L. SEGALE, E. OCHOA MACHISTE, A. BUTTAFAVA, A. FAUCITANO, U. CONTE. - Chemical and physical stability of Hydroxypropylmethylcellulose matrices containing Diltiazem Hydrocloride after gamma irradiation. - J. Pharm. Sci., 92, 131141, 2003.
L. MAGGI, E. OCHOA MACHISTE, E. FASANI, A. ALBINI, L. SEGALE, U. CONTE. - Photostability of extended-release matrix formulations. - Eur. J. Pharm. Biopharm., 55, 99-105, 2003. D. H. DAVIES, G. D. DIXON. - Photodepolymerization of hydroxypropylmethylcellulose. - J. Applied Polymer Science, 16, 2449-2459, 1972. L. MAGGI, L. SEGALE, M.L. TORRE, E. OCHOA MACHISTE, U. CONTE. - Dissolution behaviour of hydrophilic matrix tablets containing two different polyethylene oxides (PEOs) for the controlled released of a water-soluble drug. Dimensionality study. - Biomaterials, 23, 113-119, 2002. J.F. RABEK. - Photodegradation of Polymers: Mechanisms and Experimental Methods. - Chapman and Hall, London, 1994. H. KACZMAREK, A. SIONKOWSKA, A. KAMINSKA, J. KOWALONEK. - The influence of transition metal salts on photo-oxidative degradation of poly(ethylene oxide). - Polymer Degradation and Stability, 73, 437-441, 2001. Handbook of Pharmaceutical Excipients. - Pharmaceutical Press and American Pharmaceutical Association, 4th ed., 2003, pp. 491-492. R. MORITA, R. HONDA, Y. TAKAHASHI. - Development of oral controlled release system (SCRS)1. Design of SCRS and its release controlling factor. - J. Control. Rel., 63, 297-304, 2000. Y. CHEN, Z. SUN, Y. YANG, Q. KE. - Heterogeneous photocatalytic oxidation of polyvinyl alcohol in water. - J. Photochemistry and Photobiology A: Chemistry, 142, 85-89, 2001. M. BRAVAR, V. REK, R. KOSTELLA-BIFFI. - J. Polymer Sciences, Symposium 40, 1973. M.S. SULEIMAN, M.E. ABDULHAMEED, N.M. NAJIB, H.Y.MUTI. - Effect of ultraviolet radiation on the stability of Diltiazem. - Int. J. Pharm., 50, 71-73, 1989. V. ANDRISANO, P. HRELIA, R. GOTTI, A. LEONI, V. CAVRINI. - Photostability and phototoxicity studies on diltiazem. - J. Pharm. Biomed. Analysis, 25, 589-597, 2001. The Official Compendia of Standards. - United States Pharmacopoeia Convention, Inc., Rockville, 2004, pp. 2303-2304. I. REIMAN, G. HOLLATZ, T. ECKERT. - Investigations of the light-induced autoxidation of polyethylene glycols. - Archive der Pharmazie, 307, 321-327, 1974. S. SLOOP, M. LERNER, T. STEPHENS, A. TIPTON, D. PAULL, J. STENGER-SMITH. - Cross-linking poly(ethylene-oxide) and poly[oxymethylene-oligo(oxyethylene)] with ultraviolet radiation. - J. Applied Polymer Science, 53, 1563-1572, 1994. M. DOYTCHEVA, D. DOTCHEVA, R. STAMENOVA, A. ORAHOVATS, C.TSVETANOV, LEDER - Ultraviolet-induced cross-linking of solid poly(ethylene-oxide). - J. Applied Polymer Science, 64, 2299-2307, 1997. H. KACZMAREK, L. A. LINDEN, J.F. RABEK. - Reactions of hydroxyl (HO*) and hydroperoxyl (HO2*) radicals generated chemically and photochemically with poly(ethylene-oxide). - J. Polymer Science, Part A: Polymer Chemistry, 33, 879-890, 1995. T.V. IVANOVA, M.Y. MEL’NIKOV, N.V. FOK. - Photochemical reactions of radicals in polyethylene oxide at 77°K and their role in degradation processes. - Doklady Akademii Nauk SSSR [Phys. Chem.], 231, 649-652, 1976.
MANUSCRIPT Received 7 April 2004, accepted for publication 13 September 2004. 157