Analytical Methods
Cite this: DOI: 10.1039/c0ay00434k [Link]/methods
Dynamic Article Links <
View Online
PAPER
The GI dissolution method: a low volume, in vitro apparatus for assessing the dissolution/precipitation behaviour of an active pharmaceutical ingredient under biorelevant conditions
Tom Gravestock, Karl Box, John Comer,* Elizabeth Frake, Sam Judge and Rebeca Ruiz
Received 7th July 2010, Accepted 8th December 2010 DOI: 10.1039/c0ay00434k
Downloaded on 24 January 2011 Published on [Link] | doi:10.1039/C0AY00434K
This paper describes a low volume, in vitro apparatus for investigating the dissolution and precipitation behaviour of active pharmaceutical ingredients (APIs) under a wide range of experimental conditions and dissolution media. The apparatus has automated and dynamic pH control, allowing the simulated passage of drugs through the gastrointestinal tract (GIT). Experiments can be performed in the presence of biorelevant media and excipients, providing information related to the predicted behaviour of APIs under physiological conditions. The technique is described in detail and results are presented for a number of neutral, basic, acidic and ampholytic drug compounds.
Introduction
Dissolution testing16 plays an increasingly important role in the formulation, development and regulatory approval of pharmaceutical products.711 For many drug molecules, dissolution testing shows good in vivoin vitro correlation and dissolution data are often accepted by regulatory bodies in lieu of clinical studies.1218 This can greatly reduce the time-to-market for new drug products. Dissolution testing is also used to assess the product stability and batch-to-batch reproducibility, making it a valuable tool for the product development and quality control. Continual development of dissolution testing is needed as scientic and regulatory requirements change. The solubility and dissolution behaviour of ionizable compounds are strongly dependent on the pH environment. Many APIs have ionization constants which lie within the pH range of the GIT (gastrointestinal tract).19 This can result in complex dissolution and precipitation behaviour as the API moves through the GIT, which is not adequately modelled by experiments at xed pH. Some weakly basic drugs, for example dipyridamole, are soluble at gastric pH and have the ability to remain in supersaturated solution at intestinal pH, where they are otherwise highly insoluble, for some time before precipitating. This illustrates the need for dynamic pH control. It is also essential to be able to perform experiments in the presence of excipients and biorelevant media, which can both have a signicant impact on the supersaturation behaviour.20 Traditional dissolution methods have high sample weight requirements, typically 150700 mg.21 Consequently, the use of dissolution testing is presently limited to compounds in the later stages of development where larger sample quantities are
Sirius Analytical Instruments Ltd., Forest Row, East Sussex, RH18 5DW, UK. E-mail: [Link]@[Link]
available. There is growing interest in the development of low-volume, API-sparing techniques to allow the characterisation of compounds earlier in their development cycles. Miniaturised dissolution techniques using ca. 5 mg of API have been described previously and good correlation with traditional methods has been obtained.22 However, these methods have thus far been limited to xed pH. This paper presents the novel GI dissolution method, a low volume in vitro technique with dynamic pH control for investigating the behaviour of pharmaceutical substances throughout the GIT. Dissolution is monitored in situ by UV absorbance spectroscopy, allowing a high level of automation. The method is easily adaptable and experiments can be carried out in the presence of excipients. Disk intrinsic dissolution rates (DIDRs) obtained at xed pH show excellent agreement with published values. Dynamic pH experiments allow dissolution and precipitation events over the physiological pH range to be investigated. Results are presented for a diverse range of drug molecules, illustrating the applicability of the technique and its implications for use in the pharmaceutical industry.
Experimental
Molar extinction coefcients and pKas Molar extinction coefcients and pKas of all of the compounds investigated were determined in advance by UV-metric titration using the SiriusT3 or GLpKa (Sirius Analytical Instruments, East Sussex, UK). The SiriusT3 and its predecessor GLpKa are automated titration systems specically designed for the measurement of various physiochemical properties, for example pKa, log P and solubility. The UV-metric method allowed the determination of the molar extinction coefcients for both the neutral and ionised forms of a particular compound from a single
Anal. Methods
This journal is The Royal Society of Chemistry 2011
View Online
experiment. Samples were typically prepared as 5 mM stock solutions in DMSO and titrated between pH 2 and pH 12 in 1.5 mL of 0.15 M aqueous KCl. Sample concentrations were optimized in order to obtain a peak UV absorbance of approximately 1 absorbance unit. Where possible, titrations were carried out under aqueous conditions; however, it was necessary to investigate some compounds in the presence of methanol cosolvent due to their poor aqueous solubilities. Tablet production Tablets of 3 mm diameter comprising 540 mg of pure API were created inside a tablet disc using a manual hydraulic tablet press (Specac Ltd, Orpington, UK). The tablet disc is made from stainless steel (grade 316) and the compression rod (3 mm diameter) is certied up to a compression weight of 0.5 tonnes. Tablets were typically produced under a weight of 0.4 tonnes (approximately 900 lb) applied for 120 s. All tablets were visually examined to ensure their surfaces were smooth and free of visible defects. The tablet discs were placed in tablet disc holders and held in situ by an O-ring seal, as shown in Fig. 1. This prevented exposure of the reverse side of the tablet to the dissolution medium, giving a total exposed surface area of 0.07 cm2. The tablet disc holder was then inserted into an appropriate-sized vial. The tablet disc holder raises the tablet approximately 7 mm above the base of the vial, allowing the introduction of up to 2 mL of liquid to the vial without wetting the surface of the tablet prior to the start of the dissolution experiment. Initial calibration and disk intrinsic dissolution rates Dissolution rates are strongly dependent on the nature of the dissolution medium and on experimental conditions such as stirring speed.23 Initial calibration experiments were carried out using 30 mg tablets of pure salicylic acid in a 50 mM sodium phosphate solution at pH 7.4. Salicylic acid is an accepted reference standard for the calibration of the existing dissolution apparatus.23,24 All experiments were carried out at room temperature in 15 mL of the relevant dissolution medium. Reproducibility was assessed by repeating the experiments on a number of SiriusT3 and GLpKa instruments.
To compare the performance of the apparatus with other miniaturized dissolution systems, the disk intrinsic dissolution rates of atenolol, haloperidol, hydrochlorothiazide, ketoprofen, naproxen, piroxicam and propranolol were subsequently determined in 0.1 M HCl, 0.2 M acetate buffer at pH 4.5 and 0.2 M phosphate buffer at pH 6.8. Labetalol was investigated in 0.1 M HCl. The dissolution of ketoprofen was also investigated in FeSSIF and FaSSIF, to illustrate the applicability of the technique to biorelevant media. The simulated intestinal uids were prepared from Phares SIF powders ([Link] The FaSSIF contained 3 mM of sodium taurocholate and 0.75 mM of lecithin adjusted to pH 6.44. The FeSSIF contained 15 mM of sodium taurocholate and 3.75 mM of lecithin at pH 4.92. GI dissolution A concentrated buffer solution, typically 1.5 mL of a 0.125 M phosphate/acetate buffer at pH 1.5, is introduced into the sample vial below the surface of the tablet. The instrument then automatically adds the dissolution medium, typically 13.5 mL of 0.15 M aqueous KCl solution. The buffer ensures that the start pH of the experiment is close to pH 2, modeling gastric conditions, without the need for an initial pH adjustment. This allows data collection to be started as soon as the dissolution medium has been added. The buffer solution also serves to prevent changes in the pH of the medium due to dissolution of the sample. The medium is stirred at a constant rate throughout the GI dissolution experiment. UV-visible absorption spectra are recorded via a bre optic dip probe with a diode array spectrometer at xed intervals for a specied period. Spectra are typically recorded every 30 seconds for 30 minutes. Examples of UV spectra are displayed in Fig. 2a, which show the increase in the concentration of the drug in solution as the tablet dissolves and the UV absorbance increases. Low sample concentrations can be detected (<1 mM) giving a wide dynamic range to the measurement of dissolution rates, 1 109 to 1 103 g s1 cm2. When the specied time at the initial pH has elapsed, a titrant (e.g. 0.5 M KOH solution) is automatically dispensed to adjust to the next pH and UV spectra are again recorded for a specied period. This process is repeated for all the pH zones investigated. The GI dissolution experiment normally consists of four zones
Downloaded on 24 January 2011 Published on [Link] | doi:10.1039/C0AY00434K
Fig. 1 Apparatus required for the GI dissolution experiment.
Anal. Methods
This journal is The Royal Society of Chemistry 2011
View Online
Downloaded on 24 January 2011 Published on [Link] | doi:10.1039/C0AY00434K
Fig. 2 (a) UV absorption spectra collected at pH 1.77 during a GI dissolution experiment of salicylic acid.(b) Molar absorptivities of salicylic acid (dashed line) and salicylate anion (solid line), obtained from an experiment to measure salicylic acid pKa value using a solution with a known concentration of salicylic acid.
(pH 1.9, pH 3.8, pH 5.3 and pH 7.2) so that dissolution of the drug compound is monitored over the entire pH range typically encountered in the GIT. The desired pH zones, frequency of point collection and duration of time in each zone can be easily modied to suit the needs of the experimenter. A range of basic, acidic, ampholytic and neutral compounds were investigated to illustrate the applicability of this technique.
a term allowing for a temporal offset. Results are calculated using a renement process in which S, kd and t0 are varied in order to minimize the root mean square deviation between the modeled concentrations and the measured concentrations. The dissolution rate (g min1) is given by the product kdS, i.e. the dissolution rate at t0, when the concentration of X in solution is zero. The dissolution rate can be normalized for the surface area to give the intrinsic dissolution rate (IDR) with units of g s1 cm2. The half-life for dissolution, td1/2, can be expressed by td 1=2 ln2 kd (2)
Results and discussion
Calculation of dissolution rates The concentration of sample in solution at each time point is determined from the spectroscopic data by applying the Beer Lambert law, using previously determined molar extinction coefcients. Fig. 2b shows the molar extinction coefcient prole for salicylic acid. Analytical wavelength ranges are selected by the software; any data corresponding to the saturation of the UV light source are automatically excluded from the calculation. The concentration data are converted into absolute sample quantities and used to generate a graph showing the sample quantity in solution vs. time. The software then attempts to t a rst order exponential eqn (1) to the data to obtain the dissolution rate and an extrapolated solubility value. Xt S 1 ekd tt0 (1) Here [X]t is the weight in grams of drug X in solution at experiment time t (min); S is the extrapolated solubility (g) of the drug; kd is the rate constant for dissolution (min1); and t0 (min) is
This journal is The Royal Society of Chemistry 2011
In some cases, data obtained when samples precipitate can be analysed with eqn (3) to yield an initial precipitation rate and extrapolated solubility value. Xt C0 ekp tt0 Sf (3)
Here [X]t is the weight in grams of drug X in solution at experiment time t (min); C0 is the initial concentration of the drug in solution before the precipitation event (g); kp is the rate constant for precipitation (min1); t0 (min) is a term allowing for a temporal offset; and Sf is the extrapolated solubility of the drug after the precipitation event is complete. In cases where the initial concentration C0 is well dened and the precipitation data can be approximated by rst-order kinetics, the precipitation rate (g min1) is determined as the product kpC0, i.e. the precipitation rate at t0 when precipitate rst appears in the supersaturated solution. The half-life for precipitation, tp1/2, can be determined by substitution of kp for kd in eqn (2). It should be noted that the
Anal. Methods
View Online
accurate treatment of dissolution data obtained under turbid conditions relies on a wavelength-independent contribution to the absorbance from particulate scattering of light. This is not always observed, particularly when the suspension consists of particles of variable size. Calibration and disk intrinsic dissolution rates at xed pH Initial calibration experiments were performed on a GLpKa instrument using salicylic acid as a reference standard. The dissolution of a tablet of salicylic acid in a 50 mM aqueous solution of NaH2PO4$H2O adjusted to pH 7.4 was measured at room temperature in three identical experiments. Each tablet contained approximately 30 mg of pure salicylic acid. The average dissolution rate was determined as 86.1 1.4 mg min1 and demonstrated good repeatability. The disk intrinsic dissolution rate is thus 2.03 0.03 105 g s1 cm2, which is in reasonable agreement with the USP (1.84 105 g s1 cm2 in a 50 mM phosphate buffer at pH 7.4 and 37 C, at a 100 rpm paddle speed).25 Repeatability and reproducibility were investigated for the SiriusT3 instrument. Initially, two sets of dissolution experiments (each consisting of three experiments) were performed on the same instrument at stirrer speeds of 4800 rpm and 2700 rpm. As with the GLpKa, good repeatability was observed and the dissolution rate was found to be dependent on the stirrer speed, with results of 116.8 2.6 mg min1 and 71.2 1.9 mg min1 determined at 4800 rpm and 2700 rpm respectively. At 2700 rpm the average dissolution rate was determined as 1.68 105 g s1 cm2, which is in reasonable agreement with the USP. To investigate the reproducibility, single dissolution experiments were performed at a 2700 rpm stirrer speed on four SiriusT3 instruments. A mean result of 1.5 0.2 105 g s1 cm2 was obtained, showing a reasonable level of reproducibility. Disk intrinsic dissolution rates (DIDRs) for a selection of compounds at pH 1.2 (0.1 M HCl), pH 4.5 (0.2 M acetate buffer) and pH 6.8 (0.2 M phosphate buffer) at room temperature are shown in Table 1. All measurements were carried out on the SiriusT3. Tablets typically consisted of ca. 5 mg of pure compound and all experiments were carried out in 15 mL of the relevant medium. DIDRs in the table were measured in order to compare with values reported in the literature.22,26 No literature DIDRs were reported for atenolol and haloperidol at pH 1.2, or
for labetalol HCl at pH 4.5 and 6.8, so these DIDRs were not measured in this study. Excellent correlation with literature values was obtained, with a coefcient of determination (r2) of 0.98. It must be stressed that the GI dissolution apparatus is not intended to replace the existing large volume dissolution apparatus for the measurement of intrinsic dissolution rates. The primary purpose of the apparatus is to investigate the dissolution and precipitation behaviour of APIs under experimental conditions that are relevant to the GIT using low sample weights and small volumes. However, it is useful that the measured intrinsic dissolution rates compare well with published values. Ketoprofen at a single pH in FaSSIF and FeSSIF Fig. 3 shows the dissolution proles of ketoprofen obtained in FaSSIF (fasted state simulated intestinal uid) and FeSSIF (fed state simulated intestinal uid).27 Ketoprofen is an acidic molecule with a pKa of 4.2. It is predominantly ionised at pH above 4.2 and becomes increasingly soluble as the pH increases. The dissolution experiments were performed in 15 mL of the dissolution medium at 25.7 C and pH 6.44 (FaSSIF) and at 26.3 C and pH 4.92 (FeSSIF). Dissolution rates of 31.6 mg min1 (FaSSIF) and 8.4 mg min1 (FeSSIF) were calculated. Note that
Downloaded on 24 January 2011 Published on [Link] | doi:10.1039/C0AY00434K
Fig. 3 Ketoprofen dissolution proles obtained in FaSSIF and FeSSIF.
Table 1 Disk intrinsic dissolution rates (DIDRs) of acidic (A), basic (B), and ampholytic compounds (A and B), at pH 1.2 (0.1 M HCl), pH 4.5 (0.2 M acetate buffer) and pH 6.8 (0.2 M phosphate buffer). Tablet weights were in the range 4.9321.94 mg. All experiments were performed at room temperature (21.225.1 C) Disk intrinsic dissolution rate (DIDR)/mg min1 cm2 Compound Ketoprofen Naproxen Hydrochlorothiazide Atenolol Haloperidol Propranolol HCl Labetalol HCl Piroxicam
a
pKa 5.08 (A) 4.32 (A) 8.66 (A), 9.88 (A) 9.43 (B) 8.42 (B) 9.54 (B) 7.19 (A), 7.84 (B) 1.89(B), 5.34 (A)
pH 1.2 15.1 n/aa 207.7 7908.2 980.4 19.2
pH 4.5 56.6 16.3 96.4 6649.1 427.2 19013.7 4.6
pH 6.8 1939.6 314.1 120.5 3775.9 n/aa 10451.9 80.6
UV change not sufcient to allow determination of the dissolution rate.
Anal. Methods
This journal is The Royal Society of Chemistry 2011
View Online
the dissolution rate of ketoprofen in FeSSIF is lower than in FaSSIF despite FaSSIFs lower taurocholate and lecithin contents, largely because more of the ketoprofen is in the soluble ionised form at the higher pH of FaSSIF. Salicylic acid GI dissolution Salicylic acid is a weak acid with a pKa of 2.8. The GI dissolution prole of salicylic acid is displayed in Fig. 4. In this experiment, the pellet was made from 34 mg of salicylic acid; 1.5 mL of mixed phosphate/acetate buffer (0.125 M) was introduced beneath the pellet and 13.5 mL of 0.15 M KCl was introduced automatically to give an initial pH of 1.77. By applying eqn (1), the dissolution rates at pH 1.8, pH 3.7, pH 5.2 and pH 7.1 were found to be 26.2 mg min1, 39.2 mg min1, 51.1 mg min1 and 64.8 mg min1 respectively. The dissolution rate of salicylic acid is approximately 2.5 times greater at pH 7.1 than that at pH 1.8. An increase in the dissolution rate with pH is to be expected as salicylic acid is in fully anionic form at pH 7.1 but in predominantly neutral form at pH 1.8. According to the HendersonHasselbalch equation,28 the solubility of salicylic acid is several orders of magnitude higher at pH 7.1 than at pH 1.8. A similar increase in the dissolution rate might therefore be expected. This, however, is not the case. The diffusion of the drug compound across the aqueous boundary layer (ABL) at the tablet surface is thought to be the ratedetermining step in the dissolution process.26 This establishes a microclimate at the surface of the tablet, wherein the local concentration of the drug is considerably higher than in the bulk medium, assuming that sink conditions are in effect. For ionisable drugs, this has the consequence of creating a localised pH environment at the tablet surface. In this example, the high concentration of salicylic acid in the diffusion layer lowers the local pH at the tablet surface, lowering the observed dissolution rate considerably, relative to what might be expected based on its equilibrium solubility. The magnitude of the difference between the interfacial pH and the bulk solution, and consequently the impact on the dissolution rate, is strongly dependent on the solubility of the drug and its ionization state.26
GI dissolution experiments with dipyridamole Fig. 5a shows the GI dissolution prole of dipyridamole in 0.15 M aqueous KCl. Dipyridamole has two basic pKas, 6.2 and approximately 0.8. The whole tablet rapidly dissolves in the rst zone at pH 1.8, with an initial rate of 329 mg min1. In the second zone, at pH 3.8, no change to the concentration of the drug in solution is observed. Precipitation of dipyridamole is observed in both the third (pH 5.2) and fourth (pH 7.2) pH zones. At the end of the experiment the concentration of dipyridamole in solution is 7.9 mg mL1 which approaches the measured intrinsic solubility of dipyridamole, 3.6 mg mL1.20 In Fig. 5b, eqn (3) is tted to the dipyridamole precipitation data. The data obtained at pH 5.2 during precipitation of the sample from solution do not approximate well to rst-order kinetics. This is not surprising as the precipitation rate will depend on a number of dynamic factors such as particle size and surface area. In such cases, tting of the precipitation data cannot provide reliable quantitative results. The data obtained at pH 7.2 approximate well to rstorder kinetics. In this zone the precipitate has already been present in solution for 30 minutes and the physicochemical properties of the particles are likely to be well dened. Eqn (3) should therefore provide accurate quantitative information about the precipitation event. In the renement of the pH 7.2 data, all parameters were varied with the exception of C0 which was constrained to the nal dipyridamole concentration at the end of the pH 5.2 zone. It is a requirement that either C0 or t0 is constrained to an appropriate value in the tting procedure, since these two parameters are highly interdependent. The following results were obtained: kp 0.453 min1 (tp1/2 1.53 min); t0 97.4 min; Sf 0.139 mg. These results correspond to an initial precipitation rate (kpC0) of 602 mg min1 and an extrapolated solubility of 8.5 mg mL1. The dipyridamole GI dissolution prole obtained in 0.15 M aqueous KCl containing 1.5% v/v Triton X-100 surfactant is shown in Fig. 6. The whole tablet dissolves in the rst pH zone (pH 2.0), at a similar rate (250 mg min1) to that of the aqueous experiment. No precipitation is observed as the pH is increased through the subsequent three zones, highlighting the stabilizing effect of Triton X-100 on supersaturated solutions of dipyridamole. This is consistent with the results of earlier studies,20 which indicated that some formulation excipients have the ability to stabilize supersaturated solutions, and illustrates the importance of dynamic pH control to dissolution experiments.
Downloaded on 24 January 2011 Published on [Link] | doi:10.1039/C0AY00434K
GI dissolution experiments with well known drugs Table 2 shows the GI dissolution results obtained for a number of well known drugs. All experiments were performed on the same GLpKa instrument and under similar experimental conditions. All samples were measured in 0.15 M aqueous KCl in the presence of a mixed acetate/phosphate buffer. Neutral compounds. The dissolution rates of the neutral compounds chloramphenicol and dexamethasone were found to remain constant across the four pH zones. This is to be expected because these molecules remain in the same ionisation state throughout the experiments.
Anal. Methods
Fig. 4 GI dissolution prole for salicylic acid in 0.15 M aqueous KCl at 25.4 C.
This journal is The Royal Society of Chemistry 2011
View Online
Fig. 5 (a) Dipyridamole GI dissolution prole in 0.15 M aqueous KCl at 26.7 C.(b) Analysis using eqn (3) of the dipyridamole precipitation data obtained at pH 5.2 and pH 7.2 in the GI dissolution experiment shown in Fig. 7a.
Downloaded on 24 January 2011 Published on [Link] | doi:10.1039/C0AY00434K
Fig. 6 Dipyridamole GI dissolution prole in 0.15 M aqueous KCl containing 1.5% v/v Triton X-100 at 26.0 C.
Basic compounds. For the basic compounds carvedilol, chlorpromazine HCl, clopidogrel bisulfate, dipyridamole, haloperidol, maprotiline and propranolol several types of behaviour were observed. In general, dissolution rates were found to decrease with increasing pH and precipitation was commonly observed in the latter pH zones of the GI dissolution experiments. This is expected because basic drugs become more neutral, and consequently less soluble, as pH is increased. Haloperidol, maprotiline and propranolol continued to dissolve in the higher pH zones in 0.15 M KCl, though propranolol dissolved more slowly at higher pH. For chlorpromazine HCl, clopidogrel bisulfate and dipyridamole, precipitation was observed in one or more pH zones in 0.15 M aqueous KCl dissolution media. This illustrates the importance of an experiment that simulates the changing pH conditions of the GIT, which can be of great relevance for basic drugs where supersaturation effects can occur. Acidic compounds. Dissolution rates of the acidic compounds carprofen, ibuprofen, tolmetin and warfarin were generally observed to increase with increasing pH. This is expected because
Anal. Methods
acidic compounds become more ionised as the pH increases. Warfarin appeared to behave anomalously, dissolving at a slightly higher rate at pH 1.9 than at pH 3.8 and 5.3; this observation may be due to experimental uncertainty. Perhaps the most interesting results for the acidic compounds were obtained in the GI dissolution experiments of the sodium salts of ibuprofen and tolmetin in 0.15 M aqueous KCl. In the case of ibuprofen (Fig. 7), the free acid initially dissolves slowly in the neutral form before the dissolution rate increases at high pH where ibuprofen becomes increasingly anionic. The sodium salt, which might be considered to be considerably more soluble than the free acid at low pH, rapidly dissolves at pH 1.9. Once in solution, the anionic ibuprofen released from the salt converts to the less soluble neutral form, which precipitates from solution. Precipitation of the neutral ibuprofen occurs while the salt is still dissolving, resulting in the complex behaviour shown in Fig. 7. Later, the precipitate rapidly dissolves at pH 5.2 and pH 7.0, presumably due to the large surface area available for dissolution. These results suggest that the amount of an acidic drug available for absorption in the GIT can in some cases be increased by administering the drug in the salt form, thus allowing it to dissolve at low pH where it would otherwise be insoluble, forming a supersaturated solution. Ampholytic compounds. As with other ionizable compounds, the pKas of ampholytes are likely to have a signicant impact on their dissolution behaviour under aqueous conditions. Luminol has a basic pKa of 1.4 and an acidic pKa of 6.2. It was found to dissolve at an initial rate of 0.70 mg min1 at pH 1.9, where a signicant proportion of the compound was in the cationic form. The dissolution rate was lower at pH 3.8, where the compound was predominantly neutral. The dissolution rate then increased at higher pH, due to the presence of increasing amounts of the anionic form of the molecule. Piroxicam, which has similar pKas, displays similar behaviour.
Conclusions
An automated in vitro method has been described for measuring the dissolution and precipitation rates of pharmaceutical substances, using low sample weights and low volumes in a variety of dissolution media. The apparatus is designed to
This journal is The Royal Society of Chemistry 2011
View Online
Table 2 Dissolution rates of neutral, acidic (A), basic (B), and ampholytic compounds (A and B), obtained from GI dissolution experiments in 0.15 M aqueous KCl. Tablet weights were in the range of 6.324.3 mg and all experiments were conducted at room temperature (25.027.8 C). The four pH values correspond to the average pH values of the four zones of the GI dissolution experiments. Dissolution was monitored for 30 minutes at each pH zone, with the collection of UV absorbance spectra every 30 seconds Absolute dissolution rate/mg min1 Compound Chloramphenicol Dexamethasone Carprofen Ibuprofen Ibuprofen Na Tolmetin Tolmetin Na Warfarin Carvedilol Chlorpromazine HCl Clopidogrel bisulfate Dipyridamole Haloperidol Maprotiline HCl Propranolol HCl Luminol Piroxicam pKa Neutral Neutral 4.25 (A) 4.35 (A) 4.35 (A) 3.50 (A) 3.50 (A) 4.94 (A) 8.0 (B) 9.24 (B) 4.74 (B) 0.8 (B), 6.2 (B) 8.42 (B) 10.33 (B) 9.54 (B) 1.4 (B), 6.2 (A) 1.9 (B), 5.3 (A) pH 1.9 46.4 1.00 0.07 2.9 n/ab 0.46 n/ab 1.3 9.8 1105 1282 329.0 3.0 16.0 41.4 0.70 0.71 pH 3.8 48.9 0.77 0.64 3.6 n/ab 1.5 n/ab 1.1 8.5 n/ac n/ab n/ac 3.1 13.8 16.3 0.47 0.45 pH 5.3 52.8 0.85 0.95 5.4 1972 9.6 163.1 1.2 8.9 n/ac n/ab n/ab 6.7 12.4 14.1 2.1 0.86 pH 7.2 52.0 0.77 11.8 23.0 3666 23.2 973.2 2.9 n/ab n/ab n/ab n/ab n/aa 19.2 17.4 3.2 4.5
Downloaded on 24 January 2011 Published on [Link] | doi:10.1039/C0AY00434K
a Dissolution result could not be obtained due to insufcient change in the sample concentration with time. b Dissolution result could not be obtained due to sample precipitation. c Dissolution result could not be obtained due to dissolution of the entire tablet.
formulation strategies. Information regarding supersaturation and precipitation can be obtained, providing data for comparison with in vivo drug behaviour that cannot be obtained by traditional dissolution methods.
Acknowledgements
The authors would like to thank Roger Allen, Martin Stuart and Richard Woodward from Sirius Analytical Ltd., and John Potter from Specac Ltd., for their technical support and assistance. Special thanks go to Manuel Sanchez-Felix from Eli Lilly & Co. for helpful comments during the preparation of this manuscript.
References
1 C. B. Pheatt, J. Qual. Tech., 1980, 12, 158164. 2 Dissolution Testing of Solid Oral Dosage Forms, British Pharmacopoeia, The Stationery Ofce, London, 2002, pp. A419 A421. 3 U. V. Banakar and M. C. Makoid, Drug Dissolution and Bioavailability, Technomic Publishing Company, Lancaster, PA, 1994. 4 Dissolution, United States Pharmacopeia 23, Rockville, MD, United States Pharmacopeial Convention, 1995a, pp. 17911793. 5 In Vitro and In Vivo Evaluation of Dosage Forms, United States Pharmacopeia 23, United States Pharmacopeial Convention, Rockville, MD, 1995b, pp. 19241929. 6 J. E. Loske, Drug Inf. J., 1993, 27, 463468. 7 V. A. Gray, Dissolution Technol., 2002, 9(1), 14. 8 R. L. Williams and T. S. Foster, Dissolution Technol., 2004, 11(3), 67. 9 J. B. Dressman, Dissolution Technol., 2004, 11(3), 89. 10 V. P. Shah, Dissolution Technol., 2004, 11(3), 11. 11 V. A. Gray, Dissolution Technol., 2004, 11(3), 1314. 12 Extended Release Oral Dosage Forms; Development, Evaluation and Application of In Vitro/In Vivo Correlations, Guidance for Industry, Food and Drug Administration, Rockville, MD, 1997. 13 H. J. Malinowski, in Advances in Experimental Medicine and Biology, In VitroIn Vivo Correlations, ed. D. Young, J. Devane and J. Butler, Plenum, New York, 1997, vol. 423, pp. 261268.
Fig. 7 GI dissolution proles of ibuprofen free acid (lower prole) and ibuprofen sodium salt (upper prole) obtained in 0.15 M aqueous KCl at 27 C.
measure the dissolution of API tablets but could also be used to investigate the dissolution of other drug formulations, such as suspensions of powders. Whilst the apparatus does not conform to the USP, disk intrinsic dissolution rates show excellent correlation (r2 0.98) with published results obtained using similar miniaturized systems. Measured dissolution rates show good repeatability (CV z 2%) and are suitable for looking at batch-to-batch reproducibility. The dissolution apparatus has automated pH control and can be used in the novel GI dissolution experiment to simulate passage of an API through the pH zones of the GIT. Formulation excipients can be added to the dissolution medium to investigate dissolution under a wide range of conditions relevant to physiological conditions and
This journal is The Royal Society of Chemistry 2011
Anal. Methods
View Online
14 Immediate Release Solid Oral Dosage Forms. Scale-up and Postapproval Changes: Chemistry, Manufacturing, and Controls; In Vitro Dissolution Testing and In Vivo Bioequivalence Documentation, Guidance for industry, Food and Drug Administration, Rockville, MD, 1995. 15 Modied Release Solid Oral Dosage Forms. Scale-up and Postapproval Changes: Chemistry, Manufacturing, and Controls; In Vitro Dissolution Testing and In Vivo Bioequivalence Documentation, Guidance for Industry, Food and Drug Administration, Rockville, MD, 1997. 16 J. P. Skelly and G. F. Shiu, Eur. J. Drug Metab. Pharmacokinet., 1993, 18, 121129. 17 Note for Guidance on Quality of Modied Release Products: A: Oral Dosage Forms B: Transdermal Dosage Forms Section I (Quality), European Agency for the Evaluation of Medicinal Products, London, 1999. 18 J. P. Skelly, G. L. Amidon, W. H. Barr, L. Z. Benet, J. E. Carter, J. R. Robinson, V. P. Shah and A. Yacobi, Pharm. Res., 1990, 7, 975982.
19 R. Lobenberg and G. L. Amidon, Eur. J. Pharm. Biopharm., 2000, 50, 312. 20 K. Box, J. Comer and T. Gravestock, Chem. Biodiversity, 2009, 6, 17671788. 21 K. G. Mooney, M. A. Mintun, K. J. Himmelstein and V. J. Stella, J. Pharm. Sci., 1981, 70, 1322. 22 G. L. Amidon, A. S. Carlin, A. S. Hussain and L. X. Yu, Int. J. Pharm., 2004, 270, 221227. 23 T. X. Viegas, R. U. Curatella, L. L. VanWinkle and G. Brinker, Pharm. Technol., 2001, 25, 4453. 24 W. W. Hauck, A. J. DeStefano, W. E. Brown, E. S. Stippler, D. R. Abernethy and R. L. Williams, Dissolution Technol., 2009, 16(1), 712. 25 J. Mauger, J. Ballard, R. Brockson, S. De, V. Gray and D. Robinson, Dissolution Technol., 2003, 10(3), 615. 26 A. Avdeef and O. Tsinman, Pharm. Res., 2008, 25, 26132626. 27 M. Marques, Dissolution Technol., 2004, 11(2), 16. lgyi, Anal. 28 E. Baka, K. Box, J. Comer, K. Tak acs-Nov ak and G. Vo Chim. Acta, 2010, 673(1), 4046.
Downloaded on 24 January 2011 Published on [Link] | doi:10.1039/C0AY00434K
Anal. Methods
This journal is The Royal Society of Chemistry 2011