UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Biopharmaceutical properties of solid dosage form Woo, Wendy Weng Wah 1968-06-23

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


831-UBC_1968_A5_5 W66.pdf [ 2.99MB ]
JSON: 831-1.0104187.json
JSON-LD: 831-1.0104187-ld.json
RDF/XML (Pretty): 831-1.0104187-rdf.xml
RDF/JSON: 831-1.0104187-rdf.json
Turtle: 831-1.0104187-turtle.txt
N-Triples: 831-1.0104187-rdf-ntriples.txt
Original Record: 831-1.0104187-source.json
Full Text

Full Text

THE BIOPHARMACEUTICAL PROPERTIES OF SOLID DOSAGE FORMS The Operating Characteristics of a Continuous Flow Dissolution Apparatus by WENDY WENG WAH WOO B. S. P., University of British Columbia, 1966 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF Master of Science in Pharmacy (M. S. P.) in the Facult}' of Pharmacy We accept this thesis as conforming to the required standard. THE UNIVERSITY OF BRITISH COLUMBIA April, 196S In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and Study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by h.i.'S representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of *7U OLA^iOi-H^ The University of British Columbia Vancouver 8, Canada Date f 7 - H - L? fr.  ii ABSTRACT A completely automatic continuous flow dissolution procedure was developed and tested. Pertinent dissolution conditions were investigated and chosen to study the dissolution characteristics of seven brands of phenylbutazone tablets. A pumping system enabled the simulated digestive fluid to flow from,the dissolution vessel into the flow cell of a recording spectrophotometer for a continuous recording of the drug concentration in the dissolution medium, which was gradually changed from an acidic medium to a basic one. From the "in vitrott data obtained by this test procedure, a T^Q^ value of 120 minutes was chosen as a limit of acceptance for the test products. The "in vivo" characteristics of six of the brands were compared with those observed for a pharmaceutically acceptable product. Of the seven test products, only four were acceptable on the basis of both the "in vitro" and the "in vivo" data. Correlation of the "in vitro" and the "in vivo" data resulted in "least squares" lines with negative slopes. M. Pernarowski, Ph. D. Supervisor iii TABLE OF CONTENTS Page I. INTRODUCTION 1 II. LITERATURE SURVEY 5 III. THE "IN VITRO" CHARACTERISTICS OF PHENYLBUTAZONE (a) The Solubility of Phenylbutazone in Buffered Solutions 28 (b) Determination of the Isosbestic Point 29 (c) Determination of Absorptivity Value 30 IV. THE OPERATING CHARACTERISTICS OF A CONTINUOUS FLOW DISSOLUTION APPARATUS (a) Description of the Apparatus 35 (b) Calibration of the Recorder 6 (c) The Effect of Stirring Rate on Dissolution 39 (d) The Effect of Pumping Rate on Dissolution 42 (e) Test Procedure 45 (f) Results 46 V. THE "IN VIVO" CHARACTERISTICS OF PHENYLBUTAZONE TABLETS 56 VI. "IN VIVO" - "IN VITRO" CORRELATION 61 iv Page VII. DISCUSSION 67 VIII. SUMMARY 70 IX. BIBLIOGRAPHY 1 LIST OF TABLES Page I. The Effect of Stirring Rate on Dissolution 41 II. The Effect of Pumping Rate on Dissolution 43 III. Dissolution Data for Seven Brands of Phenylbutazone Tablets 50 ^50% anc* ^9<yfo V"alues f°r Seven Brands of Phenylbutazone Tablets 52 V. Areas under Blood Curves for Seven Brands of Phenylbutazone Tablets 60 vi LIST OF FIGURES Page 1. The Effect of pH on the Solubility of Phenylbutazone 32 2. The Isosbestic Point of Phenylbutazone 33 3. Calibration Curve for Phenylbutazone in Simulated Intestinal Fluid at 240 mu 34 4. Diagram of Continuous Flow Dissolution •Apparatus 37 5. Calibration Curve for Recorder External to the Spectronic 505 33 6. pH Changes at a Pumping Rate of 60 ml./min. 44 7a. Dissolution Profiles for Two Brands of Phenylbutazone Tablets 47 7b. Dissolution Profiles for Two Brands of Phenylbutazone Tablets 43 7c. Dissolution Profiles for Three Brands of Phenylbutazone Tablets 49 Vll Page 8. Concentration of Phenylbutazone in Serum after Administration of Product A to Subject 1 59 9. Correlation of T^Q^ Values with "In Vivo" Data for Subject 1 63 10. Correlation of T^Q^ Values with "In Vivo" Data for Subject 1 64 11. Correlation of T^Q^ Values with "In Vivo" Data for Subject 2 65 12. Correlation of T^Q^ Values with "In Vivo" Data for Subject 2 66 viii ACKNOWLEDGEMENT The author wishes to express her sincere gratitude to Dr.. M. Pernarowski for suggesting and supervising this study. The author is indebted to Mr. R.0. Searl for his many helpful suggestions throughout the research. Thanks is also due to Mrs. J. Naylor for her technical assistance in the laboratory. I. INTRODUCTION Tablet disintegration implies that the solid dosage form has broken up into smaller particles. However, it does not necessarily mean that the active ingredient has been released from the primary particles. This was first pointed out in the early 1950's by several researchers (23, 28) and more recently by'Levy and Hayes (13) and Yen (34). By the mid 1950Ts, Chapman and his co-workers (5, 6), in separate studies on products containing riboflavin and p-amino salicylic acid, showed that physiological availability could be predicted from the disintegration times determined by the official disintegration test (30). However, their results did suggest that the more important criterion to consider in relation to availability is the release of drug from the primary drug particle. Even though the correlation between disintegration time and dissolution has recently been shown to be invalid (1, 9, 13), the disintegration time is still an indication of drug release because, if the solid dosage form remains intact, the surface area for dissolution is limited. As shown by the Noyes-Whitney Equation (13), the amount of drug dissolved is dependent on the surface area of the dissolving solid. da _' KS ( Cs - C ) ctt " where a = amount of drug dissolved in time t 2 S = surface area of the dissolving solid K = solution rate constant of the solid C1, - concentration of the dissolving solid in the medium Cs = concentration of the dissolving solid in the diffusion layer surrounding the solute particles. Dissolution is the process whereby solid particles go into solution. It can be considered as specific types of heterogeneous reactions in which a mass transfer is effected through the net result of escape and deposition of solute molecules at the solute-solvent interface. From this layer of dissolved particles, diffusion occursscarrying the dissolved solute from the interface into the main body of the dissolution medium. Because the reaction at the interface is much slower than the convection transport process, it ultimately determines the rate of dissolution (32). Even though it is generally recognised to-day that disintegration times do not necessarily bear a relation ship to the "in vivo" action of the tablet, the disintegration time test is still the only official "in vitro" check on drug release from tablets. It has been stated that this test is inadequate and should be replaced by a dissolution test. Since the early 1960's,. several workers have designed dissolution tests ranging from Levy's beaker method (13) to completely automated systems (24). 3 The great variations in the equipment used for "in  vitro" dissolution studies can be attributed to the wide differences in the physical forms of the test preparations as well as to the attempts to duplicate "in vivo" conditions. However, basically they all consist of a vessel for the dissolution medium and the test preparation, a means of agitation, a sampling system and a means of controlling the temperature of the dissolution medium. Most of the "in vitro" dissolution studies have been done in a static system, i.e., the dissolution medium remains unchanged throughout the entire dissolution run. This system may encounter a saturation problem especially with the less soluble drug preparations. Several researchers (7, 22., 33) have used a continuous flow system to carry out dissolution studies. This type of system eliminates the possibility of saturation of the medium by the drug. It may also simulate "in vivo" conditions more than the static system because under "in vivo" conditions the digestive juices are continuously being secreted and absorbed by the mucosal cells of the gastro-intestinal tract. In an attempt to investigate the potentials of this type of system, an apparatus for continuous flow was developed with,a gradual change in the pH of the medium, from that of gastric juice to intestinal juice. This system was investigated and using a standard product, pertinent 4 dissolution conditions were established for this study. Under these conditions, seven brands of commericially available phenylbutazone tablets were tested. The "im vitro" dissolution data obtained was then correlated with "in vivo" data obtained for several subjects. 5 II. LITERATURE SURVEY The official U.S.P. disintegration apparatus (30) has the basic parts needed in a dissolution test. Workers have either used the disintegration apparatus itself or modified it to carry out disintegration as well as dissolution studies. The apparatus consists of a basket rack assembly that can be raised and lowered at a constant frequency rate of between twenty-eight to thirty-two cycles per minute through a distance of not more than six cm. and not less than five cm. The volume of the dissolution medium in the thermostatically controlled vessel (35°C - 39°C) is such that at.the highest point of the upward stroke the wire mesh remains at least 2.5 cm. below the surface of the medium and descends to not less than 2.5 cm. from the bottom of the vessel on the downward stroke. The basket rack ss;? assembly consists of six vertically held open-ended tubes whose lower ends are covered by a sieve of varying mesh size. Each tube is provided with a slotted and perforated cylindrical disk of plastic. The disk just fits the tube and moves up and down with the movement of the basket as it is raised and lowered. Using this apparatus, Schroeter and his co-workers (25) determined the disintegration times and dissolution rates of seventy-six lots of compressed tablets, including 6 an anti-inflammatory steroid, a sulfonamide, an anti-diabetic agent and an aspirin-phenacetin combination. Without stopping the agitation of the U.S.P. tablet disintegration apparatus, aliquots of the dissolution medium were withdrawn at different intervals for analysis. Time for 50% of the drug to go into solution (t^o^) or the amount of drug in solution at a specific time was used as the criterion for the rate of dissolution. Disintegration times were obtained with and without the use of the plastic disks. These workers found for the steroid a highly significant linear correlation between the rate of dissolution and the average disintegration time without the use of the plastic disks. This correlation was absent when disks were used. For the sulfonamide tablets, a similar correlation was observed when sodium chloride was present as an ingredient. Both the anti-diabetic agent and the aspirin-phenacetin combination showed no significant correlation between dissolu tion rate and disintegration time. The results indicated that there was an extreme specificity in the absence or presence of a relationship between rate of dissolution and disintegration times. When a quantitative relationship existed, the "least squares" lines relating the variables were found to vary widely depending upon the particular drug as well as on the other adjuvants in the tablet. The use of plastic disks in the 7 disintegration test was observed to mask real differences between the various lots tested. Middleton and his co-workers (19), using the U.S.P. disintegration apparatus, found a significant relationship between disintegration time, dissolution rate and physiological availability of riboflavin in sugar-coated tablets. The tablets were immersed for thirty minutes in simulated gastric fluid and the remainder of the time in simulated intestinal fluid. Without stopping the agitation of the apparatus, aliquots of the simulated digestive f.Mi fluids were withdrawn at various intervals for analysis. From the results, these workers concluded that the disinteg ration time provided a valid indication of the rate at which riboflavin went into solution from the sugar-coated tablets, thus either disintegration time or dissolution rate can provide a useful estimate of the physiological availability of riboflavin from sugar-coated tablets. However, this relationship may not exist for other drugs. In their study of the dissolution rates of eighteen brands of commericially available tolbutamide tablets in gastric fluid, Brudney and his co-workers (1) found that there Was no correlation between dissolution rates and the disintegration times. Disintegration time was determined by the official method of the Food and Drug Directorate (3). Dissolution was determined by the method described by Levy and Hayes (13) with modifications. Six litres of 0.1 N.HC1 containing 0.2% NaCl was used as the dissolution medium. The stirring rate was maintained at 250 r.p.m. and the temperature On controlled at 37 C + 0.5 C. Twenty-five ml. samples of the medium were withdrawn through a nylon filter pad at five minute intervals over a sixty minute period. Following each withdrawal, twenty-five ml.' of fresh medium was added to maintain a constant volume. Lu and his co-workers (18), in their comparative , study of'twenty-six brands of commericially available tolbutamide tablets, found that there was no correlation between disintegration time and dissolution rate in gastric fluid as had been reported earlier (1), but that there was a statistically significant correlation between disinteg ration time and dissolution rate in intestinal fluid. The disintegration times were determined by using the Food and Drug Directorate method. The dissolution rates were deter mined in both gastric and intestinal fluids of pH 1.5 and 7.5 respectively. Two and a half litres of the solution was heated to 37°C in a three-litre glass jar. The tablet was placed at the bottom of the jar and the solution stirred by a blade stirrer set at one inch from the bottom of the jar at 300 r.p.m. In two of the lots, the disintegration time exceeded the Food and Drug Directorate limit of sixty minutes, whereas the other twenty-four lots disintegrated well within .this limit. Dissolution rates were calculated 9 on the basis of the amount in mg. dissolved in one hour and the time required to reach 20$ and 40$ dissolution in gastric fluid and 50$ and 90$ dissolution in intestinal fluid. The results showed a considerable variation in the dissolution rates among the different brands tested. These findings tend to reinforce the observations of Schroeter and his co-workers (25) that the existence of a relationship between dissolution and disintegration time depends on the particular drug as well as on the conditions under which the dissolution and disintegration tests are carried out. Among the various factors to be considered in the design of a dissolution test, the means and intensity of agitation are varied most and have been reported to have a significant effect on the dissolution rate. Hamlin and his co-workers (S) used three different methods to study constant surface pellets of two polymorphic forms of methylprednisolone compressed at high pressure. One form had greater water solubility and hence should have a greater initial dissolution rate in water. The first method was that of Nelson (21), a method in which free convection, diffusion coefficient and concen tration were stated to be the rate determining factors. The only modification was that instead of determining weight loss of the pellets, the fluid surrounding the steroid pellets was assayed spectrophotometrically. In the second method, the pellets were held in a polyethylene holder at 10 the center of oval four fluid ounce bottles containing 120 ml. of deionized water. The bottles were attached to the machine described by Wurble (31). After rotation for differ ent intervals of time, the fluid surrounding the pellet was assayed spectrophotometrically. The dimensions of the pellet before and after exposure to the dissolution medium were also determined. The machine was rotated at 6 r.p.m. and at 12 r.p.m. The third method was similar to the second except that the machine described by Souder and Ellenbogen (29) was used instead of the Wurble machine. The bottles were rotated at 40 r.p.m. during exposure in a'constant temperature bath held at 37°C. It was found that when the intensity of agitation or the velocity of the dissolution medium across the dissolv ing solid was of a low order of magnitude, a significant difference in dissolution rates was observed for both forms of the steroid. When the speed of rotation of the wheel in the Wurble machine was increased from 6 r.p.m. to 12 r.p.m., the rates of dissolution of the two polymorphs became equal. Rotation at 40 r.p.m. in the machine described by Souder and Ellenbogen showed no observable significant difference in the dissolution rates of the two polymorphs. In the "in vivo" studies carried out by implanting the pellets in rats, the two polymorphs showed significantly different rates of weight loss. The "in vivo" data tend to indicate that the "in vitro" test in which there was a relatively low intensity 11 of agitation correlated best with the "in vivo" results. Levy (11), in his study of the effects of agitation on dissolution, used two methods that varied greatly in agitation intensity - the beaker method (13) and the oscillating tube method. In the second method, a plexiglass cylinder with a 100-mesh stainless steel wire screen on the bottom was attached to the basic unit of the U.SI'P. disintegration apparatus and immersed in a beaker contain ing 300 ml. of 0.1 N.HC1 at 37°C With the apparatus in motion, a tablet was dropped into the cylinder and the medium sampled at intervals by a fritted-glass immersion filter tube for analysis. Two proprietary brands'of aspirin were tested. One contained alkaline additives while the other did not. With the beaker method, which utilized a very low agitation rate and permitted the solid particles to remain as an aggregate on the bottom of the beaker, the tablets containing aspirin and the alkaline additives dissolved much more rapidly than the plain aspirin tablets. This was in agreement with the "in vivo" absorption tests carried out. The dissolution half-time of one was approx imately three times that of the other. On the other hand, there was practically no difference in the dissolution rates of the two tablet formulations when tested by the oscillating tube method, which involved relatively high agitation intensities and caused the solid particles to be dispersed. 12 Levy attributed this difference to the fact that in an aggregate of disintegrated tablet solids, the alkaline components of the buffered aspirin tablet caused an increase in the pH of the micro-environment from pH 1 to about pH 5.6. This increase in pH resulted in a more rapid dissolution of the aspirin particles. This effect was absent when aspirin and the alkaline components were physically separated by intensive agitation. The method developed by Levy and Hayes (13) in their study of the dissolution rate of plain and buffered aspirin tablets has been used by Levy and other workers in dissolution studies with only minor modifications. The dissolution assembly consisted of a 400-ml. Pyrex Griffin beaker immersed in a constant temperature bath adjusted o o to 37 C + 0.1 C. A three-blade, 5-cm. diameter polyethylene stirrer was attached to an electronically controlled stirr ing motor. Two hundred and.fifty ml. of 0.1 N.HG1 was placed in the beaker and allowed, to equilibrate to 37°C and the acetysalicylic acid tablet was dropped along the side of the beaker. The polyethylene stirrer was immersed in the dissolution medium to a depth of 27 mm. and accurately centered by means of a guide. The stirring rate was 59 r.p.m. - just sufficient to keep the solution homogeneous but low enough to allow the solids of the disintegrated tablet to remain as an aggregate at the centre of the bottom of the beaker within an area of one or two sq. cm. Seven ml. samples 13 were taken at five, ten, twenty and thirty minutes by means of a fritted-glass immersion filter tube of medium porosity. Six brands of plain acetysalicylic acid tablets, one brand of the buffered drug, and one brand of calcium acetysalicy-late carbamide complex product were tested. Six tablets were subjected to the dissolution test; another six to disin tegration test in 0.1 N'.HCl by the U.S.P. method. The results showed that there was a great differ ence in the' solution rates of the various brands of plain tablets. The tablet with the calcium salt dissolved at a faster rate than all the other products tested. This was expected due to the greater solution of the salt in aqueous medium. They found that the dissolution curve did not progress arithmetically and that it would take much longer for the second half of the drug to go into solution. The rapidly dissolving products exhibited longer disinteg ration times than the more slowly dissolving products. It was concluded that the disintegration time which is often alluded to as an index of drug availability is no criterion of availability or rate of solution. When disintegration times are abnormally long, the rate of solution and absorp tion of a drug will be seriously affected since disintegration time is just the time required by. the tablet to fall apart into reasonably small particles and not the time required for the drug to go into solution.. 14 Using the method just described (13), Levy and his co-workers (17) correlated the effect of stirring rates with the intestinal absorption of three different dosage forms of aspirin. These dosage forms differed markedly in drug absorption rate as well as in the principal mechanism involved in the release of drug to the dissolution medium. One was a rapidly disintegrating tablet containing aspirin as micro-encapsulated particles, another was a rapidly-disintegrating "plain" aspirin tablet and the third was a rapidly disintegrating tablet containing aspirin and alkaline additives. Different stirring rates in a clock wise direction using the precision stirring apparatus of Levy and Tanski (15) were used with 0.1 N.HC1 as the dissol ution medium. The resulting comparison of dissolution at 60 r.p.m. with plasma levels obtained by "in vivo" studies showed that the "in vitro" dissolution from the micro-encapsulated particles was too rapid relative to the other dosage forms. Dissolution from the plain tablets was found to be very much more sensitive than from the micro-encapsulated particles which release drug by a diffusion process. At 50 r.p.m., it was found that the ratio of "in vivo" absorption rates from the plain tablets and the micro-encapsulated particles was equal to the ratio of the dissolution rates. At 30 r.p.m., the more rapidly absorbed plain tablets were found to dissolve 15 more slowly than the much more slowly absorbed micro-encap sulated particles. In this study, it was observed that a change of only 20$ in stirring rates made a difference between successful correlation or failure with "in vivo" data. Changes in the composition of the dissolution medium was subsequently reported to have similar effects. Besides stirring rate, Levy and his co-workers (16) found that certain formulation factors have an effect on the dissolution rate of the active ingredient. Using the method described (13), these workers observed that the dissolution rate of salicylic acid in compressed tablets increased with decreasing granule size but this increase was not strictly proportional to the increase in the apparent surface area of the granules. The increase in dissolution rate from the tablets containing 20 - 40 mesh and 40 - 60 mesh granules was not as great as the increase in the apparent surface area since the disintegrated tablet particles remain aggregated on the beaker bottom due to the low intensity of agitation. The much greater dissolution of the tablets with 60 - 30 mesh granules was due to the fact that these granules were small enough to be dispersed somewhat in the medium despite the low intensity of agitation, thus permitting the moving solvent to come in contact with a greater portion of the potentially available surface. An increase of starch content in the tablets from 5 - 20$ resulted in an increase 16 in the dissolution rate of salicylic acid, probably because of more rapid and thorough disintegration of the granules. Dissolution rates were also found to increase with an increase in the precompression pressure of the tablets. This may be due to the fracturing of the drug particles at the higher slugging pressure, thus yielding smaller primary particles. Fragmentation of the more highly compressed granules during subsequent tableting may also occur. Finally, the softer granules obtained at the lower precompression pressures are more likely to undergo bonding during tablet ing and thus yield larger granules. In their study of the effects of lubricants on dissolution rates of tablets, Levy and Gumtow (12) used the beaker method (13) and the rotating disk method (14). These workers found that magnesium stearate, a hydrophobic lubri cant, decreased appreciably the dissolution rate of salicylic acid tablets compressed from granules of the pure drug while sodium lauryl sulfate, a hydrophilic lubricant, had the opposite effect. This enhancing effect of sodium lauryl sulfate was even greater with tablets made from granules that contained salicylic acid and starch. Experiments with nondisintegrating disks of salicylic acid indicated that the more commonly used hydrophobic lubricants (magnesium stearate, aluminum stearate, stearic acid, talc) decreased the effective drug-solvent interfacial area and thereby decreased the rate of dissolution of the drug, while 17 water-soluble lubricants (sodium oleate, sodium lauryl sulfate) did not have this effect. The dissolution rate enhancing effect of sodium lauryl sulfate was not due to any modification of the micro-environmental pH or solubiliz ation by micelles, but rather to the better penetration of the solvent into the tablets and their component granules and the resulting greater availability of drug surface. Levy and Sahli (14), in their comparative study of the gastro-intestinal absorption of acetysalicylic acid and its aluminum salt, correlated the urinary salicylate excretion with the dissolution rate obtained by the rotat ing disk method. The compressed tablets were mounted on Plexiglas holders with paraffin wax so that only one surface of the tablet was exposed. The holder was connected to an electronically controlled precision stirring motor. A two hundred ml. quantity of the dissolution medium was placed in a 500-ml. three-neck round-bottom flask, which was immersed in a constant temperature bath adjusted to 37°C. After temperature equilibration, the tablet attached to the holder was immersed into the dissolution medium to a depth of one inch, with the stirrer turned on at 555 r.p.m. Aliquots of five or ten ml. were removed from the flask at appropriate intervals of time for analysis. A similar volume of medium was added to maintain a constant volume. The tablet was weighed before and after the dissolution run as a check on the assay. The conditions of the experiment was such that 18 the concentration of drug in the medium was maintained at a small fraction of its total solubility. In 0.1 N.HC1, the dissolution rate of acetysalicy-lic acid was found to be 65.1 mg./hr.cm. and that of aluminum acetysalicylate in terms of salicylic acid was 8.88 mg./hr.cm. Similar differences were found when alkaline medium was used. Similarly, "in vivo" studies showed that urinary salicylate excretion from subjects taking aluminum acetysalicylate was markedly less than that from subjects taking the acid. Since the absorption of salicylates is rate-limited by their dissolution rate in gastro-intestinal fluids, these workers concluded that the less rapid absorption and subsequent excretion of aluminum acetysalicylate was probably due to its slow dissolution. Recently, Searl and Pernarowski (26) evaluated'cthe disintegration times and dissolution rates of twenty-three brands of phenylbutazone tablets. The "in vivo" character istics of three brands were compared with that observed for a clinically acceptable product by a comparison of the levels of the drug in the blood after oral administration. Disinteg ration times were determined by using the Erweka tablet disintegration apparatus. The procedure is in a publication issued by the Food and Drug Directorate (4). Six tablets were placed in the apparatus and the mean disintegration times calculated. Dissolution rates were determined by an 19 apparatus similar to that of Levy and Hayes (13). Two and a half litres of simulated intestinal fluid in a three-litre glass jar was allowed to equilibrate in a constant temper ature water bath set at 37°C + 1°C. One tablet was placed in a cylindrical wire basket (2.2 cm. in diameter and 2.8 cm. in length) which was attached below the impeller of a stirr ing shaft (three-blade, Teflon-coated propeller, 5 cm. in diameter). The shaft which was connected to a Fisher Stedi-Speed stirrer was inserted to a depth of ten cm. below the surface of the liquid and rotated at exactly 100 r.p.m. in a clockwise direction. Ten ml. aliquots were taken at fifteen-minute intervals for the. first hour, at thirty-minute intervals for the second hour and hourly thereafter for seven hours. Results showed that nine of the twenty-three products had disintegration times of more than thirty minutes. Twelve products (some with disintegration times greater than thirty minutes and some with disintegration times less than thirty minutes) were selected for more extensive study. Of these twelve products, three were found to have a disinteg ration time of more than thirty minutes with a range of more than thirty minutes. From the disintegration time data, it would appear that the drug in these tablets may not he available to the patient. Dissolution studies showed that four of the twelve products had TCQ^ value of more than 20 120 minutes but most of t>he twenty-three products disinteg rated and released their phenylbutazone content to the medium quickly. A standard T^Q^ value of 120 minutes was chosen by these workers on the basis of the data available. It was not possible to correlate the "in vivo" data to that obtained "in vitro". For the tablets examined, the best correlation was still to tablet disintegration time. However, maximum disintegration times would be more meaningful than mean disintegration times. Nash and Marcus (20) used a periodic solvent exchange method to study d-amphetamine sulfate "sustained release" capsules and tripelennamine hydrochloride "sustained release" tablets. The test sample was put into a 600-ml. Buchner type funnel with a medium porosity fritted disk filter bed and gently agitated in 400 ml. of simulated gastric juice. Finer porosity filters prevent proper drainage of the fluid while coarser types leak during sampling periods. Positive air pressure was maintained between sampling periods to prevent leakage. After thirty minutes, 200 ml. was drawn off by vacuum through the filter bed into the suction flask and collected in a 250 ml. beaker. A.four-inch standard two-way stop cork fused into the bottom of the suction flask facilitated the removal of fluid samples during the course of each run without upsetting the positive pressure at the bottom of the filter bed. An additional 200 ml. portion of simulated gastric fluid was added to the funnel to replace 21 the withdrawn volume. Further samples were withdrawn at one and a half, two, three, five, seven and twenty-four hour intervals, each time with replacement of the fluid. Norby (22) developed a continuous solvent exchange system to study the release pattern of "sustained release" tablets. One unit of the drug was placed in a beaker contain ing ten ml. of dissolution medium at thermostatically controlled temperature. This solution was kept at a constant level with the reservoir of fresh medium to ensure a constant volume in the dissolution vessel. The medium from the dissolu tion vessel flowed through tubing which had a cotton filter at the open end into a collecting vessel at a rate (0-10 ml./min.) controlled by a magnetic valve. There' was an automatic mechanism controlling this valve which controlled the flow rate through the system to give a reprod ucible test. The opening and closing of the valve was controlled by an electronically driven mechanical, device. The stirrer was power driven. A screen around the propeller protected the test'preparation from coming in contact with the blades. The stirrer was moving fast enough so that the drug was moving freely and did not stick to the sides or the bottom of the dissolution vessel. Concentration of the medium was determined either by analysing a fraction of the collected medium or by a continuous recording on a spectro photometer. 22 In their study of the effect of compression pressure on dissolution, Ganderton and his co-workers (7) used two ° methods to test compressed tablets of phenindione of differ ent formulations, all containing 100 mg. crystalline lactose with 15 mg. of. potato starch dispersed within the granules, over a pressure range of 60 - 2500 Kg.cm. on a single punch machine. In the first method, the dissolution vessel was a 2-litre beaker (containing 1.5 litres of dissolution medium) immersed in a water bath. The medium was stirred by a perspex paddle 11 cm. in diameter, held 0.5 cm. above the bottom of the vessel and rotating at 56 r.p.m. The two blades of the paddle were 2.5 cm. deep and pitched at 45° to promote axial mixing. Two diametrically opposed baffles 1.3 cm. wide were fixed in the vessel. The test tablet was placed in a cube of lOOsmesh stainless steel gauze of sides 1.5 cm., rigidly suspended in the bath 4 cm. from the paddle axis and 2 cm. below the surface of the liquid. The test was performed in 0.001 N.NaOH at 37°C Five ml. samples were withdrawn through a filter tube at suitable intervals-over a period of one hour, diluted and assayed. The second method was a continuous dissolution process, using a cylin drical perspex cell 5.1 cm. in diameter. A 100-mesh, concave, stainless steel gauze was fixed across the cell and the tablet held lightly at the centre with a vertical pin. Water buffered at pH 7 was admitted through the centre of the cell 23 base and directed radially on the gauze below the tablet. When the tablet was wetted, the retaining pin was removed and the liquid issuing from the top of the cell was collected o and assayed. The test was carried out at pH 7 and 20 G to allow the direct assay of the emerging solution without further dilution. The first litre of solution was collected and assayed. The test lasted approximately eleven minutes, giving a mean liquid velocity in the cell of 0.075 cm./sec. Disintegration tests were carried out using the method described in the British Pharmacopeia 1963. From the results, it was found that the speed of disintegration progressively decreased as the compression pressure increased. The dissolution rates fell steeply at the low compression pressures and then rise to form a peak at pressures which varied with formulation from 500 to 800 Kg. cm. . The rate then decreased to give an extended high pressure region in which dissolution rate was independ ent of both pressure and formulation. Although very weak and easily penetrated by the dissolution medium, tablets produced at very low pressures did not break up extensively during the test. Little fragmentation had occurred so that particle loss and dissolution rate were low. At higher pressures, penetration still occurred quickly and stress release and the loss of small air bubbles caused much more disruption. The van der Waals forces holding the tablet together in the dry state were ineffectual in the presence of a liquid of high dielectric constant and penetration by the dissolution medium caused the tablet to break up. With further increase in pressure, rebonding of the material occurred and a stronger and denser tablet was formed, which was less easily penetrated. Particle loss and dissolution rate were, therefore, depressed. Ultimately, high strength and low penetration prevented break-up of the tablet, and dissolution occurred only from the surface of the tablet and was therfore independent of any formulation variables. At low pressure, the dissolution rate of tablets increased as the filler size or the granule size decreased. These results indicated that these factors greatly affect the size of the particles liberated by penetration and break-up. In the case of the filler, decrease in size increased the homogeneity of the original mix, opposing the formation of large agglomerates of phenindione. Decrease in granule size modified the disposition of the external starch, a factor which has been shown to greatly affect penetration and break-up of the tablet. Smaller granules would allow its more effective distribution as a hydrophilic or antibonding layer. Both these effects will disappear at high pressures of compaction when break-up is depressed and solution occurs only from the surface of the tablet. 2 A continuous flow system was designed by Woo (33) to study eleven brands of commericially available tolbut amide tablets. The dissolution vessel was a two-litre aspir ator bottle. The rubber stopper at the bottom outlet of the bottle carried a short piece of glass tubing that was connected on the inside of the dissolution vessel to a piece of rubber tubing. A piece of very fine gauze that covered the open end of the rubber tubing acted as the filter for the outflowing medium. The glass tube that proje ed out of the vessel was connected to a T-glass tube, one arm of which was joined to an air supply, while the other arm led to a 500-ml. suction flask via rubber tubing. The rubber stopper at the top of the dissolution vessel carried a thermometer and an inlet tube from a reservoir of simul ated gastric fluid. The dissolution vessel was set on a pyro-magnetstir, that kept the medium in the vessel at a temperature of 37°G + 0.5°C, and caused the magnetic stirrer in the dissolution vessel to stir the medium at varying speeds. The dissolution vessel was filled with one and a half litres of simulated gastric fluid. After the o medium had equilibrated to 37 C, a test tablet of.tolbut amide (500 mg.) was dropped into the dissolution vessel. The stirrer was turned on to give a vigorous stirring intensity. After an initial ten minutes, the clamp on the 26 tube to the suction flask was partially unscrewed to allow a flow rate of 500 ml./30 minutes, when suction was applied. At the same time, the inflow rate from the reservoir was adjusted to deliver 500 ml./30 minutes, by partially unscrew ing the clamp on the delivery tube. This ensured that the volume, in the dissolution vessel was kept constant. During the changing of the suction flasks (after each thirty-minute interval) both the inflow and outflow tubes were clamped off by a second clamp. As the fluid flowed through the fine gauze filter, particles of the test tablet collected on the filter and caused it to be clogged up, thus slowing the outflow rate. This was overcome by turning on the air supply occassionally to blow the particles off the gauze back into the dissolution medium. The dissolution test was carried on for a period of three hours, collecting in total a volume of three litres. Individual analyses were carried out on each portion of dissolution medium collected. Disintegration tests were carried out on these tolbutamide tablets using the official U.S.P. disintegration apparatus (30). The disintegration studies were carried out without the use of the plastic disks. The various brands of tolbutamide tablets seemed to follow a similar pattern of dissolution. In the initial ten minutes, most of the tablets broke up into very fine particles or flakes. The release of the active ingredient increased gradually till a maximum was reached in the first half hour. The level remained constant for about an hour and then gradually decreased due to a dilution process by the incoming solvent, thus giving a plateau-like curve. Even though the dissolution pattern was similar for the different brands, the total.amount of the active ingredient, released in three litres of solution ranged from 60 mg. to 270 mg. No correlation with the disintegration times could be found, confirming an earlier report by Brudney and his co-workers (1). However, the dissolution test was carried out on only one tablet of each product, so definite conclusions about the tablet quality could not be made. The dissolution test described was very time consuming and required constant supervision. Moreover, the filtering system was not too efficient and tend to blog up towards the end of the dissolution run and hence would be inadequate for longer dissolution runs. 28 III. THE "IN VITRO" CHARACTERISTICS OF PHENYLBUTAZONE The drug being investigated is phenylbutazone and is used for the relief of joint pain caused by rheumatoid arthritis, gout, bursitis and other related disorders. Phenylbutazone (1,2-Diphenyl-4-butyl-3,4 Pyrazolidinedione), a weak organic acid of pKa 4.4, is very slightly soluble in water (less than 0.7 mg./ml. at 25°C) but freely soluble in alcohol (50 mg./ml.), acetone, ether and ethyl acetate. Phenylbutazone powder tends to agglomerate and float on the surface of an aqueous medium. Attempts at recrystallization yielded long needles that float on the surface of aqueous solutions. Hence, Butazolidin (Geigy) tablets (which have been reported to give good clinical response (2, 27)) were chosen as a standard for this investigation. (a) The Solubility of Phenylbutazone in Buffered Solutions It is necessary, during the dissolution test, to change from an acidic medium of pH 1.2 to a pH 6.2. Since the solubility of phenylbutazone is affected by the p.H of the dissolving medium, solubility studies were carried out in buffers of different pH values. Procedure - Using the Smith, Kline & French disintegration apparatus (29), phenylbutazone powder is put into each of six'bottles containing 50 ml. of buffer (two samples of each buffer are used). The bottles are suspended in a water bath maintained at 37°C and rotated so as to provide good mixing for a period of twenty-four hours. At the end .of this period, the solutions are filtered quickly, diluted if necessary with the buffer and assayed spectro-photometrically. Results - The solubility curve (Figure 1), drawn from the data obtained, showed that the solubility of phenylbutazone is very low in buffers of low pH values. However, once the pH value reached 6, the solubility increased greatly. (b) Determination of the Isosbestic Point The isosbestic point of a solution is the wave length at which changes in the pH of the solution do not affect the absorbancy (As) reading of the solution. Since the conditions of this dissolution test involve a change from an acidic to a basic medium, the wavelength at which absorbancy readings are to be recorded must be at the isosbestic point. Procedure - Weigh accurately 100 mg. of phenyl butazone powder and dissolve^in 100 ml. of 95% ethanol. Ten ml. aliquots of this solution are diluted accurately to 1000 ml. with buffers of various pH values (ranging from pH 1.2 to pH 7.5)'. The spectrum of these solutions are recorded on a recording spectrophotometer (Spectronic 505, Bausch and Lomb). From the spectra of the solutions, 30 an isosbestic point was observed at between 238 mu and 245 mu. The absorbancy of these solutions are then read on a Beckman DU spectrophotometer from 238 mu to 245 mu with each solution blanked against the respective buffer. A plot of the absorbancy readings for the solutions showed that the wavelength that is closest to being the isosbestic- point for phenylbutazone is 240 mu (Figure 2). This point was chosen as the wavelength at which all subsequent absorbancy readings were recorded. (c) Determination of the Absorptivity Value The absorptivity (as) is a constant for a particular solute in a particular solvent at a certain wavelength. According to Beer's Law, it is the slope of the line relating the absorbancy of a solution and its concentration. As = asbc where As is the absorbancy of the solution as is the absorptivity b is the cell length in cm. c is the concentration in gm./L. Procedure -.Weigh accurately 100 mg. phenylbutazone poxtfder and dissolve in 100 ml. of 95% ethanol. Dilute aliquots of five to twenty ml. to 1000 ml. with Simulated Intestinal Fluid U.S.P. The absorbancy of each solution is read on a Beckman DU spectrophotometer at 240 mu, using Simulated Intestinal Fluid U.S.P. as a blank. 31 Results - The absorbancy readings were plotted and a straight line was obtained. The absorptivity was calculated from the slope of this line and found to be 41.5 (Figure 3). 3 2 240 243 246 0 Wavelength (mu) Figure 2. The Isosbestic Point of Phenylbutazone 34 0.80 0.60 % 0.40 0.20 As= asbc aq - As = 41.5 5 10 15 20 Concentration (mg./L.) Figure 3. Calibration Curve for Phenylbutazone in Simulated Intestinal Fluid at 240 mu 35 IV. THE OPERATING CHARACTERISTICS OF A CONTINUOUS FLOW DISSOLUTION APPARATUS (a) Description of the Apparatus The dissolution vessel (a one-litre three-neck round-bottom flask) is immersed in a constant temperature o o water bath at 37 C t 0.5 C. Dipping into the central neck of the flask is the shaft of a Fisher Stedi-Speed adjustable stirrer (Model 12) with a four-blade impeller blade (3 cm. in diameter) and a cylindrical wire basket at its bottom. The basket which is A cm. long and 2.5 cm. in diameter is made from 10-mesh stainless steel wire cloth. A glass tube which acts as the inlet tube for the dissolution medium dips through the right side neck into the dissolution vessel to a depth of 10 cm. Connected to this tube is a two-way stop cork, one arm of which is connected to a reservoir of simulated gastric juice (2 gm. NaCl + 7 ml. HCl in 1000 ml. distilled water, pH 1.2) while the other arm is joined to a reservoir of simulated intestinal juice (6.8 gm. KH2P0^ + 1.52 gm. NaOH in 1000 ml. distilled water, pH 7.5). Both reservoirs of dissolution medium are placed on heaters to keep the solution at 37°C 0.5°C An inverted sintered glass funnel (30 ml. capacity) of coarse porosity dips into the left side arm of the dissol ution vessel. It is joined to a short length of glass tube carrying a sintered tip (coarse porosity), that dips into the dissolution vessel to a depth of '5.0 cm. A combination glass electrode (not illustrated in Figure 4) leading from a recording potentiometer (Potent-iograph E336A) and a short piece of glass tube, which acts as the outlet tube, dip into the broad end of the sintered glass funnel. The outlet tube is joined by Tygon tubing to an adjustable pump (Cole Palmer A-769) from which tubing leads to a flow cell (1 cm. in length) in a recording spectro photometer (Spectronic 505, Bausch & Lomb). A recorder (Varicord Model 43) is attached to the spectrophotometer for a direct recording of the concentration of the solution passing through the flow cell. From the flow cell, the solution is delivered into a twelve-litre covered glass vessel, that acts as the collecting vessel. (b) Calibration of the Recorder  Procedure - The recorder is set on the 25 and mV range. It is zeroed with gastric juice in the sample cell and air as the blank. Solutions of known concentration of phenyl butazone are read in the sample cell of the spectrophotometer at 240 mu and the chart readings on the recorder scale are recorded. From the known concentrations and the chart readings on the recorder scale, a graph is plotted (Figure 5). gure 4. Diagram of Continuous Flow Dissolution Apparatus ISPECTRONIC 5Q5 • t DISCHARGE COLE PALMER A-769 ADOUSTABLE PUMP t FILTERING DEVICE FOR CONTINUOUS FLOW SINTERED GLASS FUNNEL (COARSE 30 mi.) SINTERED TIP (COARSE) VARICORD MODEL FISHER STEDI* SPEED STIRRER TO STIRRING SUCTION / SHAFT T£ST FUJ|D x GLASS TUBE TWO WAY STOPCOCK TEST FLUID H GLASS TUBE ONE LITER FLASK BASKET 38 150 to •H Ti crj CD ca •p u crj X! o 100 10 20 Concentration (mg. phenylbutazone/L^) Figure 5. Calibration Curve for Recorder External to the Spectronic 505 (c) The Effect of Stirring Rate on Dissolution The stirring rate has been reported by many researchers (8, 11, 17) to have a considerable effect on the dissolution rate. Levy (11) has stated that too intensive an agitation rate is to be avoided since X-ray photographs have shown that tablet particles tend to remain as an aggregate on the stomach surface. This implies that the mixing process in the stomach is very low. However, in the small intestine, the particles appeared to be very well dispersed indicating that the mixing in this region is quite vigorous. Since phenylbutazone is nearly insoluble in solu tions of pH 1.2 but very soluble in solutions of pH 7.5, it would appear that the absorption of phenylbutazone occurs mainly under the basic conditions of the small intestine. Hence, a stirring rate that causes the drug particles to be well dispersed throughout the medium may be desirable in the dissolution test under study. Consequently, stirring rates of 65, 100 and 115 r.p.m. in the forward direction were investigated using the designed dissolution apparatus and test procedure. These stirring rates were investigated with the stirrer at differ ent depths in the dissolution medium, using Butazolidin tablets as the test preparation, and at a constant pumping rate of 60 ml./minute. Each rate was tested in duplicate. 40 Results - The data obtained from the dissolution runs are tabulated in Table I. It can be seen that when the stirring is not enough to move the particles, the release rate of the drug is low even though the solution in the dissolution vessel appears to be homogeneous. The low release rate is due to the fact that the tablet particles remain as an aggregate on the dissolution vessel bottom, thus present ing a limited surface area to the dissolution medium. When the basket is 6 Cm. from the vessel bottom, the stirring is insufficient even at the high stirring rate of 115 r.p.m. At 3.5 cm. from the bottom of the vessel, the stirring is adequate except at the lox^est rate of 65 r.p.m., when about half of the tablet particles remain aggregated on the vessel bottom. The other two stirring rates caused the particles to move through the medium, the higher rate of 115 r.p.m. giving a more vigorous movement. At a depth of 1 cm. from the vessel bottom, the stirring effect is quite vigorous even at the rate of 65 r.p.m. To avoid this stirring effect, this depth was not chosen for product testing. The depth finally chosen was 3.5 cm. with a stirr ing rate of! 100 r.p.m. rather than 115 r.p.m. The former rate produced the desired stirring effect. At this speed, the particles move slowly through the medium. 41 Table I. The Effect of Stirring Rate on Dissolution Stirring rate Depth of basket Mg. phenylbutazone (forward direction) from vessel bottom in solution in 3 hours 1 cm. .87.5 mg. 65 r.p.m. 3*5 cm. 62.1 mg. 6 cm. 55.4 mg. 1 cm. 83.1 mg. 100 r.p.m. 3.5 cm. 83.5 mg. 6 cm. 75.0 mg. 115 r.p.m. 1 cm. 3.5 cm. 6 cm. 90.0 mg. 85.0 mg. 78.5 mg. 42 (d) The Effect of Pumping Rate on Dissolution Three pumping rates were studied for their effect on dissolution rate. The dissolution test was carried out with Butazolidin tablets as the test preparation and with the basket at 3.5 cm. fronr the vessel bottom stirring at a rate of 100 r.p.m. The pumping rates studied were 50 ml./min., 60 ml./min. and 70 ml./min. Rates lower than 50 ml./min. were not studied due to instrumental limitations in reading the more concentrated solutions. Results - The data obtained are tabulated in Table II. This data shows that there is no significant differ ence in the release of the drug in the three-hour period. The pH changes in the medium are affected by the pumping rate to a slight extent. Higher pumping rates gave a much faster pH change from acidic to basic pH. At the pumping rate of 60 ml./min., the sharp rise in pH from about pH of 2 to pH of 6.2 occurred within twenty-five minutes (Figure 6), while at a pumping rate of 70 ml./min., it occurred within twenty minutes. However, at 70 ml./min., the final volume of solvent is too great to be handled easily. At 50 ml./min., the pH change occurred in thirty minutes but at this rate, the solution is becoming too saturated to be read on the spectrophotometer. A pumping rate of 60 ml'./min. was therefore, finally chosen for the dissolution test. 43 Table II. The Effect of Pumping Rate on Dissolution Pumping rate Mg. of phenylbutazone in solution 1 hour 2 hours 3 hours 50 ml./min. 1. 4.0 mg, 2. 3.9 mg. 56.0 mg, 51.0 mg. 86.5 mg. 86.4 mg. 60 ml./min. 1. 9.0 mg. 2. 8.3 mg. 70 ml./min. 1. 5.9 mg. 2. 6.7 mg. 68.4 mg. 62.6 mg. 62.2 mg. 63.0 mg. 89.6 mg. 86.4 mg, 86.5 mg. 88.2 mg. 45' (e) Test Procedure With the apparatus set up as described and as shown in Figure 4, the dissolution vessel (containing one litre of simulated gastric juice) is allowed to equilibrate in a constant temperature water bath at 37°C + 0.5°C The entire system is flushed out with simulated gastric juice to remove air from the system. The recorder is zeroed with simulated gastric juice in the sample flow cell and air as the blank. A single phenylbutazone tablet (100 mg.) is put into the wire basket below the impeller blade and the whole apparatus is set infemotion. The stirring rate is 100 r.p.m. in a forward (clockwise) direction at a depth of 3.5 cm. from the bottom of the dissolution vessel. With the inlet tube from the reservoir of simulated gastric juice opened, the pump is turned on to deliver 60 ml./min. through the entire system. After the initial thirty minutes, simulated intest inal juice is allowed into the system. At this point, the potentiometer is switched on to record pH changes as the medium is gradually changed from an acidic medium to a basic one. Aliquots from the collecting vessel are taken at hourly intervals for analysis on a Beckman DU spectrophoto meter at 240 mu, with simulated intestinal juice as the blank. As the solution from the dissolution vessel passes 46 through the flow cell in the spectrophotometer set at 240 mu, the Varicord records the concentration. This gives a contin uous recording of the concentration of the solution as dissolution occurs. The dissolution run is carried on for three to six hours depending on the dissolution rate! of the indivi dual products. At the end of each run, the entire system is flushed out with 95% ethanol and water. Using this test procedure, seven brands of commer-icially available phenylbutazone tablets (100 mg.) were tested in triplicate. Four of the products are sugar-coated tablets (Products A, E, W, X) while the other three are enteric coated tablets (Products AA, CC, DD^). Each product was tested in triplicate. (f) Results An average continuous concentration plot was obtained for each product from the curves drawn out on the recorder. This plot gave the dissolution profile of the product and also allows the calculation of the total amount of drug released in any period of time (Figures 7a, b, c). The hourly analysis on the Beckman DU spectrophotometer also gave the amount of drug released at each hourly interval, thus providing a check on the amount registered by the recorder (Table III). From the data on the products obtained by the two 47 L8 LO I-H e o •H •P cd U •P c © o o o 20 CC 1 2 Time (hours) Figure 7b. Dissolution Profiles for Two Brands of Phenylbutazone Tablets 50 Table III. Dissolution Data for Seven Brands of Phenylbutazone Tablets Mg. of phenylbutazone in solution Product 1 hour 2 hours 3 hours 1. 5.2 mg. 62.5 mg. 86.5 mg. A 2. 5.2 mg. 61.9 mg. 88.0 mg. 3. 5.2 mg. 65.9 mg. 93.5 mg. 1. 1.2 mg. 52.5 mg. 86.5 mg. E 2. 1.8 mg. 50.4 mg. 81.8 mg. 3. 1.5 mg. 48.0 mg. 79.2 mg. 1. 0.8 mg. 43.2 mg. 85.3 mg. AA 2. 0.8 mg. 43.2 mg. 86.5 mg. 3. 1.2 mg. 51-5 mg. 89.6 mg. 1. 1.4 mg. 91.4 mg. 96.2 mg. CC 2. 1.7 mg. 95.7 mg. 100.0 mg. 3. 1.7 mg. 90.7 mg. 95.8 mg. 51 Table III (Continued) Mg. of phenylbutazone in solution Product 3 hours 4 hours 5 hours 6 hours 1. 7.4 mg. 23.6 mg. 55.8 mg. 75.7 mg. X 2. 16.2 mg. 55.3 mg. 78.7 mg. 93.0 mg. 3. 23.8 mg. 50.8 mg. 75.0 mg. 92.2 mg. 1. 36.0 mg. 55.2 mg. 70.6 mg. 86.3 mg. W 2. 36.2 mg. 55.0 mg. 70.3 mg. 86.0 mg. 3. 36.3 mg. 55.3 mg. 70.5 mg. 86.2 mg. 1. 5.4 mg. 27.0 mg. 94.0 mg. 101.5 mg. DDX 2. 4.5 mg. 22.5 mg. 91.7 mg. 99.0 mg. 3. 4.5 mg. 22.3 mg. 90.9 mg. 93.0 mg. 52 Table IV. T50% and T90% Values for Seven Phenylbutazone Tablets Brands of Product T$0$ value T90% value A 100 minutes 180 minutes E 120 minutes 215 minutes W 225 minutes 735 minutes. X 255 minutes 735 minutes AA 125 minutes 185 minutes CC 80 minutes 120 minutes DD-L 225 minutes 300 minutes 53 methods, time for 50$ of the drug (T5Q$) and 90$ of the drug (T^0^) to go into solution was calculated and tabulated in Table IV. The four brands of sugar-coated tablets showed little dissolution in simulated gastric fluid. However, once the medium reached a pH of 6.2, the concentration of the active ingredient increased greatly. This change in pH occurred in twenty-five minutes at the pumping rate of 60 ml./minute. Products A and E (which disintegrated within fifteen minutes of the start of the experiment) showed similar dissolution patterns. The concentration of active ingredient continued to increase gradually once the medium reached a pH of 6. A peak was reached in about one and a half hours. The curve then gradually declined due to a dilution process by the incoming solvent. Product E released slightly less active ingredient from the tablet than did Product A. The other two sugar-coated products showed differ ent dissolution profiles due to the lack of disintegration. Product W did not disintegrate at all even at the end of six hours. Dissolution by this product was very low and stayed at the same level throughout the entire dissolution run. The active ingredient seemed to dissolve out from the surface of the tablet, which was smaller but still intact 5k at the end of the dissolution run. Product W showed very little•tablet variation. Product X was found to vary from one tablet to another in the same lot. After the.tablet coating came off after three hours, large particles could be'seen coming off and,,at this point, the concentration of the active ingredient registered on the recorder was seen to increase suddenly but not greatly. The concentration level remained approximately the same for about an hour, then decreased very slightly for the rest of the time. The dissolution rate of Products ¥ and X was so low that the dissolution-run had to be carried on for six hours instead of three hours. Products A and E released about 90% of the active ingredient in three hours while it took Products W and X six hours to release the same amount of active ingredient. The three enteric coated Products AA, CC and DD-^ remained intact till the dissolution medium reached a pH of 7. Products AA and CC broke up very quickly into small particles. Product DD^ did not start to disintegrate until about three and a half hours after the start of the dissolu tion run. Product CC gave a much faster dissolution rate than Product AA, with a' concentration peak much higher than those of the other test products and, even earlier than the sugar-coated products. This was probably due to its 55 almost immediate disintegration in the alkaline medium into very fine particles. The dissolution profile for Product AA was similar to those for the sugar-coated Products A and E, except for a delay of about twenty-five minutes in its peak due to the enteric coating. The fall in the dissolution curve of Product CC was very sharp while that of Product AA was much more gradual. Product DD]_ released its active ingred ient rapidly after disintegration occurred, giving a peak second only to Product CC. Henee, it would appear that the disintegration time of the tablet had an effect on the dissolution of the tablet when the disintegration time was excessively long. A T^Q^ value of 120 minutes would seem to be a reasonable limit of acceptance of the products that could be said to be effective "in vivo". This implies that only four of the products tested, A, E, AA and CC are acceptable. This "in vitro" estimation can be supported by the "in vivo" data that showed the three Prdducts E, AA and CC were at least 75$ as effective as Product A (a product tested and found to give good clinical response). Product. AA would be a border-line case by both the "in vivo" and the "in vitro" results. 56 V. THE "IN VIVO" CHARACTERISTICS OF PHENYLBUTAZONE TABLETS Burns and his co-workers (2) claimed that due to the affinity of phenylbutazone for plasma proteins as compared to tissue proteins, a great portion of administered phenylbutazone is found in the plasma. They reported that peak plasma levels were reached in about two hours after oral administration, indicating that the drug was rapidly absorbed from the gastrointestinal tract. Plasma levels did not increase proportionately with increasing doses of the drug but tend to. reach a limiting concentration, which varied considerably among subjects. On repeated daily dosage, the drug.accumulated in the body with a progressive increase in the plasma concentration until a plateau was reached on the third or fourth day. In a comparative study of commericially available brands of phenylbutazone tablets, Searl and Pernarowski (26) determined serum levels after administration of the products; to human subjects. The seven products were administered to nine subjects as shown in Table V. Each subject was given two tablets, that is approximately 200 mg. of phenylbutazone. In general, the product was given to the subject shortly after breakfast. Three to five samples of blood were taken over a 43-hour period but only the results to 30 hours are shown. 57 Determination of Phenylbutazone in Serum Fifteen ml. of blood are withdrawn by means of a dry sterile syringe and needle from a large vein in the bend of the elbow. The blood is allowed to clot, then centrifuged and the serum is removed. Two ml. of serum are transferred to a 50-ml. glass-stoppered centrifuge tube. Five ml. of 3 N.HC1- solution and 20.0 ml. of heptane are added and the tube is shaken for thirty minutes in an auto matic shaker and centrifuged at 1200 r.p.m. for five minutes. Fifteen,ml. of heptane are withdrawn and transferred to a 2.00-ml. bottle. Five ml. of 2.5 N.NaOH solution are added and swirled gently. The solution is then shaken for five minutes and poured into a 50-ml. centrifuge tube and centrifuged at 1200 r.p.m. for three minutes. The absorbancy of the sodium hydroxide solution is read at 265 mu. The absorptivity of phenylbutazone at 265 mu in 2.5 N.NaOH is 65.2. It may be necessary to dilute the sodium hydroxide solution or to decrease the amount of serum taken for analysis if the concentration of phenylbutazone in the serum exceeds 30 mg. per litre. A blood sample taken before the administration of the drug serves as a control or "blood-blank". This sample when treated as described above, absorbs some ultraviolet radiant energy at 265 mu. Consequently, all subsequent assay values must be adjusted to compensate for this "blood-blank". 58 Results Peak plasma levels were obtained in five rather than in two hours as had been reported by Burns and his co-workers (2). A typical blood curve is shown in Figure 8. After the peak region, the serum drug level decreased quite gradually. The areas under the blood curves were determined by a planimeter and are tabulated in Table V. Using Product A as a chosen standard, only three other products, E, AA and CC appeared to be at least 75% as effective as Product A, and thus considered acceptable. Since data for subjects 3-9 was incomplete, it was not possible to use it for a comparison. However, available data from these subjects tend to indicate that response to the products follow the same trend as in subjects 1 and 2. 10 20 Time (hours) Figure 8. Concentration of Phenylbutazone in Serum after Administration u! of Product A to Subject 1 60 Table V. Areas under Blood Curves for Seven Brands of Phenylbutazone Tablets ^sProduct Sub jectNv A E W X AA CC DD]_ 1 18.57 14.8 1.82 13.3 12.42 15.33 6.55 2 20.55 15.2 4.61 7.0 15.51 22.84 11.21 3 24.1 26.4 8.9 17.0 — — — 4 — — — — 8.62 — — 5 — — — — 10.48 — — 6 — — — — 23.34 — 7 — — — — • — 11.48 — 8 — — — — — — 3.79 9 — . — — — • — — 4.65 61 VI« "IN VIVO" - "IN VITRO" CORRELATION "In vitro" results only serve to reduce the number of samples considered suitable for "in vivo" testing. However, they are significant when they can be quantitatively correlated with "in vivo" data. Even though the "in vivo" data available is not complete for all test products, an attempt was made to correlate this data with the T^Q^ and the T^Q^ values obtained by the "in vitro" test procedure. Correlation was attempted only for the two subjects with complete "in vivo" data, since data for Product A was not available for the other subjects. However, the response shown by these subjects seem to follow the same trend as that of subjects 1 and 2. It can be assumed, therefore, that the same type of correlation could be obtained for these subjects, taking into account the possible differences in the metabolism of the drug in these subjects. By the use of linear regression, the line of "least squares" drawn for each set of data for subjects 1 and 2 (Figures 9, 10, 11, 12). The points for subject 2 seem to fit better on the line of "least squares" drawn than those for subject 1. One of the reasons for the deviations from the "least squares" line could be that some of the "in vivo" blood curves were drawn from the inadequate data. Tablet variability (especially for Product X) could also have contributed to the deviations. 62 The lines of "least squares" had negative slopes indicating an inverse relationship between the "in vivo" and the "in vitro" data. The slopes of these lines ranged from -0.042, to -0.0755 with the slopes of the T^Q^ lines greater than the T^Q^ lines of both subjects. The slopes of the lines for subject 2 were greater than those of subject 1. This was expected since subject 2 showed greater response to the drug than subject 1. The difference between the slopes of the' T^Q$ line and the T^Q$ line of subject 1 was twice that of subject 2. 64 66 67 VII. DISCUSSION The dissolution procedure described herein is completely automatic. No routine analysis of the medium is required, because a complete recording of the amount of the active ingredient released from the tablet throughout the dissolution process is obtained on the recorder. Determination of the total area under the recorded curve will give the total amount of drug released. At the same time, a complete recording of the pH changes in the dissolution medium can be obtained on a recording potentiometer. This recording enables one to see how the dissolution process is affected by the changing pH of the dissolution medium. A single analysis of the collected medium will also give the total amount of drug released in a period of time but it will not give the dissolution time for a spec ified amount of drug to go into solution. Further, this final analysis does not give the dissolution profile which can be seen to vary from product to product. Even though the apparatus set-up seems to involve many components, the basic parts required for the continuous flow system are the reservoirs of dissolution medium, the dissolution vessel, the pump and the collecting vessel. The recording potentiometer and the spectrophotometer with its attached recorder are accessories that furnish more inform ation and avoid manual analysis of the medium. These 68 accessories are not necessary if one is interested in the total amount of drug released in a certain period of time. The basic parts of this dissolution apparatus are relatively easy to obtain and assemble and provide a good reproducible method of checking on product quality. The inverse relationship established between the "in vivo" and the "in vitro" data obtained by this test procedure tends to indicate that this procedure may possibly be used in the prediction of the "in vivo" availability of the active ingredient from the solid dosage form. A T^Q% value of 120 minutes as the limit of acceptance for the products seems reasonable on the basis of both the "in vivo" and the "in vitro" data. From the "least squares" lines, it would seem that products with a T^0% value of more than five hours would not be absorbed into the blood stream at all. From both the "in vivo" and the "in vitro" data, only four of the seven products tested, namely A, E, AA and CC, can be accepted as products that will be effective when administered to patients. Products W and X give such low serum levels and long T^Q^ values that one can conclude that these products are unsatisfactory and should not be administered to patients. Product DD^ with slightly higher serum levels is still unsatisfactory. The poor "in vivo" release from these products is most probably due to poor formulation, which.is revealed by their poor disintegration. characteristics. Product W and X do not disintegrate whil the disintegration time of Product DD-j_ is abnormally long 70 VIII. SUMMARY In this investigation, a completely automatic continuous flow dissolution procedure was developed and tested. Pertinent conditions of dissolution were studied and chosen to test seven brands of commericially available phenylbutazone tablets. "In vivo" studies were carried out on these products by determining serum levels after the administration of the products to nine subjects. From the "in vitro" data obtained by the test dissolution procedure, a T^Q$ value °f 120 minutes was chosen as the limit of acceptance for the test products. Of the seven test products, only four were acceptable on the basis of both the "in vivo" and the "in vitro" data. Correlation of the "in vivo" and the "in vitro" data resulted in "least squares" lines with negative slopes. This correlation indicated the possibility of using this "in vitro" dissolution procedure in the prediction of the "in vivo" availability of phenylbutazone from the solid dosage form. 71 IX. BIBLIOGRAPHY 1. Brudney, N., Stewart, D.J. and Eustace, B.T. : Canad. Med. Ass. J., 90 : 980 (1964). 2. Burns, J.J., Rose, R.K., Chenkin, T., Goldman', A., Schulert, A., and Brodie, B.B. : J. Pharmacol. Exp. Ther., 109 : 346 (1953). 3. Canada, Dept. of National Health and Welfare, Food and Drug Directorate : The determination of the disintegration time of tablets, Ottawa, 1965. 4. Canada Food and Drugs Act, and Regulations : Food and Drugs act and regulations, loose-leaf office consolidation, Sec.0.015, The Queen's Printer, Ottawa, 1954. 5. Chapman, D.G., Crisafio, R. and Campbell, J.A. : J. Pharm. Sci., 43 : 297 (1954). 6. Idem : J. Pharm. Sci., 45 : 374 (1956). 7. Ganderton, D., Hadgraft, J.W., Rispin, W.T. and Thompson, A.G. : Pharm. Acta Helv., 42 : 152 (1967). 8. Hamlin, W.E., Nelson, E., Ballard, B.E. and Wagner, J.G. : J. Pharm. Sci., 51 : 432 (1962). 9. Levy, G. : J. Pharm. Sci., 50 : 388 (1961). 10. Levy, G. : J. Pharm. Sci., 51 : 947 (1962). 11. Levy, G. : J. Pharm. Sci., 52 : 1039 (1963). 12. Levy, G. and Gumtow, R.H. : J. Pharm. Sci., 52 : 1139 (1963) . 13. Levy, G. and Hayes, B.A. : New Eng. J. Med., 262 : 1053 (I960). 14. Levy, G. and Sahli, B.A. : J. Pharm. Sci., 51 : 58 (1962). 15. Levy, G. and Tanski, W., Jr. : J. Pharm. Sci., _53 : 679 (1964) . 16. Levy, G., Antkowiak, J.M., Procknal, J.A. and White, D.C. : J. Pharm. Sci., 52 : 1047 (1963). 72 17. Levy, G., Leonards, J.'R. and Procknal, J.A. : J. Pharm. Sci., 54 : 1719 (1965). 18. Lu, F.C., Rice, W.B. and Mainville, C.W. : Ganad. Med. Ass. J., 92 : 1166 (1965). 19. Middleton, E.J., Davies, J.M. and Morrison, A.B. : J. Pharm. Sci., 53 : 1378 (1964). 20. Nash, R.A. and Marcus, A.D. : Drug Std., 28 : 1 (I960). 21. Nelson, E. : J. Am. Pharm. Ass. (Sci. Ed.), 47 : 297 (1958). — 22. Norby, A.P. : Dansk Tidsskr. Farm., 39 : 171 (1965). 23. Parrott, E.L., Wurster, D.E. and Higuchi, T. : J. Am. Pharm. Ass. (Sci. Ed.), 44 : 269 (1955). 24. Schroeter, L.G. and Wagner, J.G. : J. Pharm. Sci., 51 : 957 (1962). ~ 25. Schroeter, L.C., Tingstad, J.E., Knoechel, E.L. and Wagner, J.G. : J. Pharm. Sci., 51.: 865 (1962). 26. Searl, R.O. and Pernarowski, M. : Ganad. Med. Ass. J., 96 : 1513 (1967). 27. Stephens, C.A.L., Jr., Yeoman, E.E., Holbrook, W.P., Hill, D.F. and Goodin, W.L. : J. A. M. A., 150 : 1084 (1952) . 28. Sperandio, G.J., Evanson, R.V. and DeKay, H.G. : J. Am. Pharm. Ass. (Sci. Ed.), 37 : 71 (1948). 29. Souder, J.C. and Ellenbogen, W.C. : Drug Std., 26 : 77 (1953) . 30. United States Pharmacopeia, 16 th. rev., Mack Publishing Co., Easton, Pa., I960. 31. Wurble, M.S. : Am. J. Pharm., 102 : 318 (1930). 32. Wurster, D.E. and Taylor, P.W. : J. Pharm. Sci., 54 : 169 (1965). 33. Woo, W.W. : B. S. P. Thesis, University of British Columbia, Vancouver, 1966. 34. Yen, J.K.C. : Can. Pharm. J., 97 : 493 (1964). 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items