132176111 11111111111 \| - LIBRARY #1 3 1293 15011131131 11211131111134 11 mdligan state uniVOnit” This is to certify that the thesis entitled THE STORAGE STABILITY OF A MOISTURE SENSITIVE ORAL SOLID DRUG PRODUCT IN BULK AND BLISTER PACKAGES: THE EFFECT OF REPETITIVE OPENING AND CLOSING OF BULK PACKAGES presented by Leo Clifford Pires has been accepted towards fulfillment of the requirements for M. S . degree in Packaglng 01 Law. Major professor DateMarch 31, 1988 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution PLACE ll RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE ll . ll MSU Is An Affirmative ActiorVEquel Opponunlty Institmion THE STORAGE STABILITY OF A MOISTURE SENSITIVE ORAL SOLID DRUG PRODUCT IN BULK AND BLISTER PACKAGES: THE EFFECT OF REPETITIVE OPENING AND CLOSING 0P BULK PACKAGES By Leo Clifford Pires A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE School of Packaging 1988 ABSTRACT THE STORAGE STABILITY OF A MOISTURE SENSITIVE ORAL SOLID DRUG PRODUCT IN BULK AND BLISTER PACKAGES: THE EFFECT OF REPETITIVE OPENING AND CLOSING OF BULK PACKAGES. By LEO CLIFFORD PIRES The storage stability of an Ibuprofen type product was determined for 30 and 50 count high density polyethylene bulk containers following the repetitive opening and closure of the packages and the results compared to the stability of the product in prefilled unit dose blister packages. Sealed bulk containers served as a control for the studies. Storage studies were carried out at two conditions, namely: 72 ° F/SO % RH and 78 ° F/85 t RH. Bulk containers were opened at predetermined time intervals (3 days) and allowed to stand opened in the respective storage environments for 2 minutes, after which time the containers were resealed. In all cases, the moisture content of the product was determined as a measure of product quality. The blister packages were fabricated from: (1) PVC; (ii) Saran coated PVC; and (iii) Aclar/PVC, laminate structures. The results of water vapor transmission rate (WVTR) studies carried out on the respective package systems are presented. The stability data for the drug tablets packaged in the blister packages are also presented and the results compared to the product stability data obtained for the bulk containers. DEDICATION This thesis is dedicated to my family, especially to my parents, Mr. and.Mrs. Anthony C. Pires. in appreciation of their love, support and commitment to education. To my wife Lorraine and daughter Chantelle who have been very understanding and supportive, a very special thank you. ii ACKNOWLEDGEMENTS The author thanks Dr. Jack R. Ciacin for his guidance and patience towards the completion of this thesis. Dr. Ciacin contributed a great deal of his time and support to this investigation while serving as my major advisor. The author also thanks Dr. Hugh E. Lockhart for his assistance in obtaining test product and serving as a comittee member. Dr. Paul Singh for his assistance in reviewing the statistical analysis and experimental design, and also serving as a comittee member. Dr. Ian Gray for his time as a comittee member. We acknowledge the monetary gift from Allied Corporation for research in the area of Pharmaceutical Packaging, that was used to support this study. Special thanks to Mr. T Gisondi of the Allied Corporation for his time as a comittee member. iii TABLE OF CONTENTS LIST OF TABLES ...................................... LIST or FIGURES ............... 1 ...................... 1.0 INTRODUCTION ................................... 2.0 LITERATURE REVIEW .............................. 3.0 MATERIALS AND METHODS .......................... 3.1 Determination of initial moisture content .... 3.2 Moisture sorption isotherm ................... 3.3 Package water vapor transmission rates ....... 3.4 Storage stability studies .................... 4.0 RESULTS AND DISCUSSIONS ........................ 4.2 Equilibrium moisture isotherm ................ 4.3 Water vapor permeability of package .......... 4.4 Storage stability studies .................... 4.5 Statistical analysis ......................... 5.0 CONCLUSION ..................................... IV 12 13 15 18 23 27 27 28 32 49 80 85 87 9O LIST OF TABLES Table 1. 10. 11. 12. Equilibrium relative humidities for saturated salt solutions .................................. Initial moisture content of Ibuprofen tablets .. Equilibrium moisture content of Ibuprofen tablets at 72'F ................................. Equilibrium moisture content of Ibuprofen tablets at 78’F ................................. Net weight gain of packaged desiccant in 30's stored at 78°F / 85 % RH. ....................... Net weight gain of packaged desiccant in 50's stored at 78°F / 85 % RH. ....................... Net weight gain of packaged desiccant in 30's stored at 72°F / 50 8 RH ........................ Net weight gain of packaged desiccant in 50's stored at 72°F / 50 % RH ....................... Net weight gain of packaged desiccant tablets in PVC blisters (package system I) at both storage conditions ...................................... Net weight gain of packaged desiccant tablets in PVC/Aclar blisters (package system II) at both storage conditions .............................. Net weight gain of packaged desiccant tablets in PVC/Saran blisters (package system III) at both storage conditions .............................. Water vapor transmission rates & permeability constants of the three blister package systems at both conditions of storage ...................... Page 16 27 29 29 34 35 36 37 38 39 4O 48 l3. 14. 15. 16. 17. 18. Apparent water vapor transmission rates & permeability constants of the 30 & 50 count containers at both conditions of storage ........ Equilibrium moisture content (EMC) of Ibuprofen tablets in Aclar, PVC and Saran blister packages at 78°F / 85 % RH ............................... Equilibrium moisture content (ENC) of Ibuprofen tablets in Aclar, PVC and Saran blister packages at 72°F / 50 8 RH ............................... Equilibrium moisture content (EMC) of Ibuprofen tablets in 50 6 30 count containers at ' 78°F / 85 8 RH .................................. Equilibrium moisture content (EMC) of Ibuprofen tablets in 50 & 30 count containers at 72°F / 50 8 RH .................................. Summary of the statistical analysis of bulk packages (30 & 50 count) and blister packages (systems I,II,III) at both conditions of storage. VI 48 53 54 55 55 83 LIST OF FIGURES Figure l. Sorption isotherm of Ibuprofen tablets stored at 72°F ......................................... 2. Sorption isotherm of Ibuprofen tablets stored at 78°F ......................................... 3. WVTR of 30's stored at 78°F / 85 % RH ........... 4. WVTR of 50's stored at 78°F / 85 % RH ........... 5. WVTR of 30's stored at 72°F / 50 % RH ........... 6. WVTR of 50's stored at 72°F / 50 % RH ........... 7. WVTR of Aclar blisters .......................... 8. WVTR of PVC blisters ............................ 9. WVTR of Saran blisters .......................... 10. EMC of Ibuprofen tablets in 30's stored at 78°F / 85 % RH .................................. 11. EMC of Ibuprofen tablets in 50's stored at 78°F / 85 3 RH .................................. 12. EMC of Ibuprofen tablets in 30's stored at 72°F / 50 % RH .................................. 13. EMC of Ibuprofen tablets in 50's stored at 72°F / 50 3 RH .................................. 14. EMC of Ibuprofen tablets in Aclar blisters vs 50 count stored at 78°F / 85 % RH ............... 15. EMC of Ibuprofen tablets in PVC blisters vs 50 count stored at 78°F / 85 % RH .................. 16. EMC of Ibuprofen tablets in Saran blisters vs 50 count stored at 78°F / 85 % RH ............... VII Page 31 41 42 43 44 45 46 47 56 57 58 59 6O 61 62 l7. l8. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. EMC of Ibuprofen tablets in Aclar blisters vs 30 count stored at 78°F / 85 8 RH ............. EMC of Ibuprofen tablets in PVC blisters vs 30 count stored at 78°F / 85 8 RH ................ EMC of Ibuprofen tablets in Saran blisters vs 30 count stored at 78°F / 85 8 RH ............. EMC of Ibuprofen tablets in Aclar blisters vs 50 count stored at 72°F / 50 8 RH ............. EMC of Ibuprofen tablets in PVC blisters vs 50 count stored at 72°F / 50 8 RH ................ EMC of Ibuprofen tablets in Saran blisters vs 50 count stored at 72°F / 50 8 RH ............. EMC of Ibuprofen tablets in Aclar blisters vs 30 count stored at 72°F / 50 8 RH ............. EMC of Ibuprofen tablets in PVC blisters vs 30 count stored at 72°F / 50 8 RH ................ EMC of Ibuprofen tablets in Saran blisters vs 30 count stored at 72°F / 50 8 RH ............. EMC of Ibuprofen tablets in Aclar vs PVC blisters stored at 78°F / 85 8 RH ............. EMC of Ibuprofen tablets in Aclar vs Saran blisters stored at 78°F / 85 8 RH ............. EMC of Ibuprofen tablets in PVC vs Saran blisters stored at 78°F / 85 8 RH ............. EMC of Ibuprofen tablets in Aclar vs PVC blisters stored at 72°F / 50 8 RH ............. EMC of Ibuprofen tablets in Aclar vs Saran blisters stored at 72°F / 50 8 RH ............. EMC of Ibuprofen tablets in PVC vs Saran blisters stored at 72°F / 50 8 RH ............. EMC of Ibuprofen tablets in Aclar vs PVC vs Saran blisters stored at 78°F / 85 8 RH ............. EMC of Ibuprofen tablets in Aclar vs PVC vs Saran blisters stored at 72°F / 50 8 RH ............. VIII 63 64 65 66 67 68 69 7O 71 72 73 74 75 76 77 78 79 1.0 INTRODUCTION Because of the lack of data comparing the chemical stability of bulk drugs following the repetitive opening and closing of the package, to the drug packaged in prefilled unit dose blister packages, this study was proposed to establish such stability data, under well defined conditions of storage. The stability information gleaned from this study will aid in selecting the most cost effective packaging system which will adequately protect the product, as well as in assigning accurate expiration dates to the packaged drug. The efficacy of a solid oral pharmaceutical product depends on the quality of the formulation at the time the medication is administered. The formulation, in this case a tablet, can lose its intended effect between the time of manufacture and the time when it is ingested due to various factors, such as an unstable formulation. Reactions involving moisture, oxygen, light etc. can also be factors contributing to product quality loss. Preventing such undesirable effects would be dependent on the nature of the formulation and its sensitivity, the package system and the storage environment. Packaging the product in the appropriate flexible or rigid container and storing it at recommended storage conditions could, to a large extent prevent physical and chemical deterioration of the drug, and the resultant loss of therapeutic effect. In this study the effect of moisture is the predominant factor being considered. Typical spoilage mechanisms of oral solid drugs caused by the loss of moisture include rapid loss of cohesiveness between particles in the formulation, and the tablet's loss of sheen. Further, a gain in moisture may cause the tablet's to harden with age and the dissolution time of the drug product may be increased. Therefore, the intended dosage levels may not be achieved after ingestion. Packaging the drug in container/closure systems having very low moisture permeability, in general provides adequate protection to maintain product stability and prevent the loss of the intended therapeutic effect. The opening and closing of the package is a regime nearly every patient has to go through in the case of bulk containers, especially if the medication has to be taken 'over a period of time. Typically these medications are stored in an environment (i.e., a medicine chest in the bathroom), where the humidity and temperature can be abusive (i.e., higher than 78° F/ 85 8 RH.).As a result of opening and closing of the package, the water vapor that enters the headspace of the container would be sorbed by the drug product, thereby altering its quality, even though the container and the closure provide adequate protection from moisture permeation. Further, the patient may apply a different application torque to the closure each time the container is resealed, which may also alter the permeability of the container/closure system. The specific oral solid drug formulation evaluated in this study was known to be a moisture sensitive product ( each tablet contains Ibuprofen USP 200 mg). Selection of this product was based on the following criteria. 1. Knowledge of the mechanism of drug quality loss 2. The availability of a stability indicating assay procedure to monitor product quality, as a function of moisture. 3. Present or future interest in packaging the drug tablet in a unit dose blister package. The package systems studied included two high density polyethylene bulk container/closure systems of 30 and 50 count, and three blister type packages. The objectives of this study are summarized below: 1. To determine the effect of repetitive opening and closing of bulk packages on the moisture uptake of Ibuprofen tablets vis a vis the same drug in prefilled unit dose blister packages (3 package types), at identical conditions of storage. 2. To determine the effect of repetitive opening and closing of bulk packages (30 & 50 count containers) on the moisture uptake of Ibuprofen tablets vis a vis the same drug in identical unopened bulk containers at the same conditions of storage. 3. To determine the moisture content of the stored drug at predetermined time intervals as a function of package type, storage time and storage conditions. 2.0 LITERATURE REVIEW Critical interest in the influence of water activity (Aw) on food product quality and stability began about two decades ago. It was promoted by empirical observations inconsistent with a direct relationship between total moisture and product stability. In classical papers, Lea et a1 (1943) and Bryce and Pearce (1946) reported an optimum moisture content for dried milk powder. Rockland (1957) reported an optimum moisture content for shelled walnuts and related this to an optimum Aw range. These observations were in contrast to the generally accepted belief that stability increased consistently with decreasing moisture content. It was also suggested that the optimum Aw was related to the manner in which the moisture was bound by individual chemical moieties (Pauling, 1945) and that this could be delineated using a mathematical treatment of moisture sorption data. This provided the basis for the local isotherm concept and the utilization of moisture sorption data for derivation of stability isotherms (Rockland, 1957), (Saravacos, 1969). On the basis of the local isotherm concept, which delineated, at least qualitatively, moisture binding properties and hence stability, it was also proposed that changes in temperature would displace moisture sorption isotherms and result in predictable changes in quality and stability. Previously Bull (1944) and subsequently others have subscribed to the premise that the Brunauer-Emmett-Teller (BET) equation (1938) could be useful in providing a theoretical basis for differentiating moisture binding characteristics. Of classical importance was the thermodynamic treatment of moisture sorption phenomena by Mclaren and Rowen (1951). During the past 20 years there have been a number of equations proposed to describe moisture sorption isotherms and to provide theoretical justification for mosture sorption phenomena (Van den Berg and Bruin, 1981). More recently, emphasis has been placed upon the relationships between moisture sorption characteristics, i.e., physical and chemical changes, as well as microbiological influences on the stability and quality of food and other moisture sensitive products. With respect to pharmaceutical tablets exposed to excess moisture the drug can either harden or soften and then crumble, depending on the excipients and the product. Viellard et al. ( 1979 ) measured the increase in tablet hardness as a function of moisture uptake of a drug formulation in blister packages stored at 95°F, 90 8 RH. In comparing the moisture uptake by lactose tablets versus phosphate ~tab1ets they showed the lactose formulation to be more sensitive to moisture, regardless of the method of manufacture. The extent of product quality loss as a result of moisture sorption was shown in these studies to be formulation dependent. This suggests that excipients are prime factors in the sensitivity of drug products to moisture. When stored under high humidity, the USP dissolution calibrator, (disintegrating type Prednisone 50 mg, in a lactose base) becomes harder with time and when evaluated in the apparatus suitable for dissolution testing, it dissolves much more slowly. The authors suggested a permanent reordering of tablet structure and the occurrence of chemical and physical property changes other than a mere increase in moisture content. Tablet hardness can be increased by increasing the binding agent and compression forces. Unfortunately, hardness can be increased to the point at which the only release of drug will arise from the relatively low surface area of the outside of the dosage form; The ultimate dissolution of a drug depends upon two factors, namely: the disintegration of the tablet, and the deaggregation of the particles. It was recognized that unless an oral-dosage form (pill or tablet) disintegrated into small aggregates, it could not be efficiently absorbed by the body (USP XIV). Taborsky et a1. (1981) concluded that packaging and storage conditions markedly affect tablet dissolution characteristics. Tablets stored in containers that have the highest moisture permeability experience the greatest loss in rate of dissolution. One of the major concerns of the Food and Drug Administration (FDA) is that the absorption of moisture by oral dosage forms can initiate the process of deComposition. The absorption of moisture is facilitated by containers which are permeable and not airtight (Federal Register, 1978). The location in the home where the packaged drug product is stored is another factor that contributes to the integrity of both medicated and non-medicated products (Lieberman et al. 1980). For example, medicated lozenges and over-the-counter formulations (OTC) are usually stored in the medicine cabinet, a location that exposes the product to cycles of varying temperature and relative humidity. Moisture sorption by medicated lozenges and over the counter formulations can result in changes in dissolution characteristics of the product, as well as in drug decomposition because of interaction of the medicament with the candy base, excipients, flavor or acidulent (Lieberman et a1. 1980). While researchers are concerned with the bioavailability of generic versus proprietary name drugs, it is important to note that dissolution of tablets of the same lot can be altered by packaging and storage variations, and could be the pre-eminent consideration in some climates. The effect of moisture has been the primary consideration in the present investigation, hence an insight into the mechanism of mobility of the sorbed water is desirable. Two specific motions are possible: isotropic or anisotropic reorientation. Deviations from isotropic motion cause a splitting of the 2H resonance line, which is a measure of the anisotropy of the motion. The overall conclusion of studies conducted by Lechert et a1. (1980) was that different starches show different degrees of water reorientation rates. For example, the bound water in potato starch showed distinct anisotropic water motion, while that in wheat starch showed only weak anisotropic motion, and that in corn starch showed isotropic motion. The explanation given for the isotropic motion in corn starch was that the relative residence time in the ordered state is too short, which may be caused by the starch structure itself or by impurities in the starch, such as protein (Lechert et a1. 1980). The starch components in this formulation (Ibuprofen tablets) under study are corn starch and pregelatinized starch. The majority of methods for quantitation of total water, including oven and vacuum moisture determination, Karl Fischer titration and distillation methods, are destructive to the sample. For example, in oven drying the nature of the sample can be changed by an increase in temperature, and sublimation of drug or excipients can take place when this method is applied. One of the most useful techniques employed to determine the moisture content of a product is nuclear 10 magnetic resonance spectroscopy (NMR), as it provides a rapid, sensitive, direct, and most importantly a noninvasive, nondestructive determination of the quantity of water present, as well as insight into the structure and dynamic characteristics of water (Berendsen, 1975) in complex structures. In this study Karl Fischer titration was employed, as it is the method of analysis for moisture determination by the pharmaceutical industry, with special reference to Ibuprofen. The titrimetric determination of water depends on the fact that a solution of sulfur dioxide and iodine in pyridine and alcohol (Karl Fischer reagent) reacts with water stoichiometrically. The reaction has been shown to take place in two distinct steps, water being involved in the first step only. (The reaction scheme is presented below). a, . CBDLSO CHN+HO ZCHN .' +CHN Csasmz'ss 2's: 2 "'"°“ 55 . ss ”2 _. a N c535" < I e cnzon c5 5 < 0 SO 2 I O 11 The overall reaction which takes place is given by the equation: ' a CSHSNJZ + CSHSN'SOZ‘. C’H’N 4. HP 4. CH3OH ZCSHSN < +CSHSN < 1 59c") Each mole of water requires 1 mole of iodine, 1 mole of sulphur dioxide, 3 moles of pyridine and 1 mole of methanol. 3.0 MATERIALS AND METHODS Film coated Ibuprofen, analgesic tablets were selected for this study. Tablets from the same lot were used for the entire study. The active ingredient in each tablet is Ibuprofen USP 200 mg, the other ingredients in this formulation being carnauba wax, corn starch, hydroxypropylmethyl cellulose, propylene glycol, silicon dioxide, pregelatinized starch, stearic acid and titanium dioxide. Two bulk packages (ie. 30 and 50 count containers) and three blister package types, containing Ibuprofen tablets were evaluated. The bulk packages were fabricated from white high density polyethylene resin of specific gravity 0.950 - 0 .960. The closures of 24 mm and 28 mm were fabricated from polypropylene, with-pulp backed polyethylene wax coated liners. One blister package was fabricated from 7.5 mil calendered unplasticized clear polyvinyl chloride (PVC) film. The second blister package was fabricated from a laminate of 0.75 mil Aclaro / 2 mil polyethylene (PE) / 7.5 mil PVC. While the third blister package was fabricated from a laminate of Saran‘0 coated PVC (80 g PVDC / in2 PVC). The backing material for these three blister package systems 12 13 was a dead soft 1 mil aluminium foil with a vinyl heat seal coating. Aclar 9 is the registered trade mark of a chloro-fluropolymer film manufactured by Allied-Signal Corporation. Saran 0 is the registered trade name of the Dow Chemical Company for its polyvinylidene chloride polymers and copolymers. Storage Conditions: Stability studies were carried out at two storage conditions, namely 72' F/SO 8 relative humidity (RH) and 78° F/85 8 RH. The former conditions (i.e., 72° F/50 8 RH) being representative of ambient storage, while the latter storage conditions (i.e., 78 ° F/85 8 RH) are included as representative of abusive storage conditions. All the experimental aspects of this study were initiated at the same time and carried out simultaneously. 3.1 Determination of Initial Moisture Content ( IMC ). The initial moisture content was determined by the Karl Fischer titrimetric method, adopted from the Upjohn Company general procedure, which is a standard and accurate procedure for moisture determination of Ibuprofen tablets. A precision "Aquatrator" (Precision Scientific Company, Chicago, IL 60647) manual model was used as the Karl Fischer l4 conductometric moisture determination apparatus. Standardizing of the Karl Fischer reagent: Approximately 30 m1 of absolute methyl alcohol was added to a suitable beaker and placed into position on the aquameter. The above solution was titrated to end point with Karl Fischer reagent. Approximately 300 mg of sodium tartrate, neutral ACS grade, were accurately weighed and added to the above solution. This mixture was then titrated to the end point. The water equivalent of the Karl Fischer reagent, (titer value) expressed in mg of water per ml of Karl Fischer reagent was calculated by the following formula. ( Sw x 0.1566 )/ KF - We (water equivalent. mg/ml ) --- (l) where Sw weight of sodium tartrate (mg). 0.1566 - water content of sodium tartrate. KF m1 of Karl Fischer reagent used . We water equivalent of Karl Fischer reagent (mg HZO/ml reagent). Two tablets were accurately weighed on weighing paper using a Mettler AE 160 electronic balance, having up to 4 decimal place accuracy. The tablets were then added to the above solution in the beaker and titrated to the end point. Five replicates i.e., a total of 10 tablets were analyzed for IMC. 15 Moisture contents of the tablets expressed as a percent were determined by appropriate substitutions in the following equation. (KF x We) / Sw x 100 - 8 water --- (2) Where KF - m1 of Karl Fischer reagent used in this titration We - water equivalent of Karl Fischer reagent (mg H20 /ml reagent) Sw - weight of tablets in mg The average IMC expressed as g of HZO/lOO g dry weight of product was calculated from the percent moisture content obtained from equation - (2), for all five replicates. 3.2 Moisture Sorption Isotherm: Equilibrium sorption isotherms were determined at 72° F and 78° F. A gravimetric method was used for determining the moisture sorption isotherm. Since the equilibrium moisture content (EMC) is usually expressed as percent moisture on a dry weight basis, the initial moisture content was measured with a high degree of accuracy. In developing moisture sorption isotherm data, care was taken to insure the relative humidity chambers employed were maintained at constant temperature and relative humidity. 16 The moisture sorption isotherms developed in this study were determined by placing tablets of known initial moisture content in a series of six humidity chambers per temperature, maintained at two constant temperatures at 72° F and 78° F, and registering the weight change. Humidity chambers ranging from 21 to 80 8 RH were prepared and maintained at the respective temperatures of test. The desired humidities in these chambers were obtained by placing appropriate salt solutions into tightly closed 5 gallon plastic buckets. The salt solutions employed and their corresponding relative humidities are summarized in Table 1. Table l. Equilibrium relative humidities for saturated salt solutions. Saturated salt solution 8 RH at 77° F ( 25° C ) Potassium Acetate 22.7 Magnesium Chloride 33.2 Magnesium Nitrate 53.4 Sodium Nitrite 64.3 Sodium Chloride 75.8 Ammonium Sulfate 80.3 (Adopted from "Creating and maintaining humidities by salt solutions", technical bulletin No. 5, Hygrodynamics). 17 Salt solutions were prepared by adding distilled water slowly, to the chemically pure salt in a crystallization dish with constant stirring, until about half the salt crystals were dissolved. The prepared salt solutions were placed within the tightly closed chambers and allowed to equilibrate. The relative humidity within each of these chambers was monitored daily by temperature and humidity sensors (”Hygrodynamics" Newport Scientific, INC. 8246-E Sandy Court, Jessup, MD 20794-0189.) mounted directly in the lid of the buckets. This procedure was used to assure that constant relative humidity and temperature conditions were maintained. Approximately 5 g of tablets were accurately weighed into petri dishes and the tared dishes placed in the respective humidity buckets, all with their lids off. Three replicate determinations were made at each condition of temperature and relative humidity. At predetermined time intervals, the humidity buckets were opened and lids quickly placed on each petri dish. The lidded petri dishes were allowed to equilibrate at ambient temperature (i.e., 72 ° F) and weighed on a Mettler AB 160 electronic balance. This procedure was repeated until no significant weight change was observed. 18 The equilibrium moisture content (EHC) at each relative humidity was calculated using the following equation : Pf(l+IMC)~ EMC _ [ - l ] x 100 ---- (3) Pi Where EMC - equilibrium moisture content (g H20/100 g dry wt solids) I Pf - final product weight (3) Pi - initial product weight (g) IMC - initial moisture content (g HZO/g dry wt) Moisture sorption isotherms were obtained by plotting the average equilibrium moisture content of the three replicates versus relative humidity at 72° F and 78° F. 3.3 Package Water Vapor Transmission Rates (WVTR). WVTR studies of unit dose blister packages: The WVTR values for the three blister package types were determined at the following conditions, 72° F/50 8 RH and 78 ° F/85 8 RH. For each blister package system, 5 individual strips were filled with 5 activated desiccant tablets (from United Desiccants - Gates, 6845 Westfield Avenue, Pennsauken, NJ 08110-1582) per strip (i.e., 25 dessicant filled individual units) and two strips (i.e., 10 units) filled with 19 glass beads. The blister packages were then heat sealed with the backing material on a commercial sealing unit. They were weighed and stored at each test condition. Blisters containing glass beads were used as controls. At predetermined time intervals (3 days), blisters were removed from the constant temperature humidity chambers, allowed to equilibrate at ambient temperature and weighed. This procedure ’was repeated until a constant rate of transmission was attained. Average net weight gain of desiccant filled blisters was obtained by subtracting the average control weight change from the observed gain of each blister. Y - ( Tf - Ti ) - ( Cf - Ci ) ----- (4) Where Y '- net water permeated (g). Tf - final wt of each test package (g). Ti - initial wt of each test package (g). (Cf - Ci) - the average of the difference between the final and initial weights of the controls. The water vapor transmission rate (WVTR) for each blister package system was obtained from the slope of the plot, where the average net weight gain of 5 blister strips was plotted as a function of time elapsed. The average WVTR per blister 20 cavity was calculated by dividing the slope value by five (5 blister cavities per strip). The permeability constant of each individual blister cavity for all three blister package systems was then determined by the following equation. (WVTR). Ps (RH) 100 Where Pp - permeability constant of each blister package type (g H20 / day . mm Hg . Blister cavity). WVTR - water vapor transmission rate of each blister package type (g H20 / day . Blister cavity). RH - relative humidity of storage. Ps - saturated water vapor pressure at test temperature (mm Hg). A total of 42 individual blister strips (test and controls) were used for WVTR determinations of the unit dose package structures, at the two storage conditions. 21 WVTR studies of closed 30 and 50 count bulk packages: The water vapor transmission rates for the two bulk packages (i.e., 30 and 50 count containers) used in this study were determined gravimetrically by'a procedure analagous to that described for the blister packages, at both test conditions. For each bulk package system, 10 containers were filled to approximately 13 mm of the closure with desiccant (Drierite) and the containers sealed under commercial operating conditions. Two control containers were sealed with glass beads, equivalent in weight to the packages containing desiccant. The test and control packages were then weighed and stored at both conditions of test (i.e., 72° F / 50 8 RH and 78° F / 85 8 RH). At predetermined time intervals, (i.e., 3 days) the bulk packages were removed from storage, allowed to equilibrate to ambient conditions and weighed. Average net weight gain of desiccant filled bulk packages were then obtained by substitution in equation (4), and the water vapor transmission rate determined per package. A total of 48 containers were used for the above WVTR determinations for the two bulk container systems, at both storage conditions. 22 WVTR studies of the 30 and 50 count bulk packages subjected to repetitive opening and closing: To evaluate the effect of repetitive opening and closing of the bulk packages, (i.e., 30 and 50 count containers) on total moisture uptake by desiccant, the following studies were carried out. For each bulk package system, 10 containers were filled to approximately 13 mm of the closure with desiccant (Drierite) and the containers sealed under commercial operating conditions. Two control containers were sealed with an equivalent weight of glass beads. The test and control packages were weighed and stored at both conditions of test (i.e., 72° F/50 8 RH and 78° F/85 8 RH). Every 3 days, the containers were opened in the storage environment and allowed to stand opened for 2 minutes. After this time, the containers were resealed at the appropriate application torque and weighed. The process was repeated over a sixty day period, at intervals of 3 days. Throughout the course of this study, application and removal torques were monitored. Average net weight gain of desiccant filled bulk packages was determined by substitution in equation (4), and the apparent water vapor transmission rates and permeability constants were determined by a procedure analogous to that described for the blister packages, at both test conditions. 23 Apparent WVTR values obtained were compared graphically to those of the sealed bulk containers over the same period of - storage. A total of 48 containers were used for the above determination for the two bulk container systems, at both storage conditions. 3.4 Storage Stability Studies: Both bulk containers and blister packages were placed in storage at 72° F/ 50 8 RH and 78° F/ 85 8 RH. The bulk packages were cycled through a series of opening and closing regimes during the course of the stability study. At 30 and 15 day intervals, a representative number of bulk and blister package samples were removed from storage and the moisture content of the product determined. Throughout the study, sealed bulk containers served as positive controls, and the quality of the drug in the two test packages was compared to that in the sealed container, over the same period of storage. Storage stability of unit dose packages: For each blister system at each storage condition, 6 individual strips of 10 (i.e., a total of 60 individual blisters) were filled with Ibuprofen tablets, and two strips of 10 blisters (i.e., a total of 20 individual blisters) were filled with glass beads as controls. The respective blisters 24 were heat sealed with the backing material by a commercial sealing unit. They were accurately weighed on an electronic AB 160 Mettler analytical balance, and stored at the respective test conditions, (i.e., 72° F/50 8 RH and 78° F/85 8 RH) The individual blister strips were removed from the constant temperature / relative humidity chambers at 15 day intervals and allowed to equilibrate for approximately 30 minutes at ambient temperature. The blister packages were then weighed and the moisture content of the tablets determined gravimetrically. After the blister strips were weighed, they were placed back into the enviromental chambers and stored for an additional period of time, (i.e., 15 days). This procedure was repeated over the entire 180 days of storage. The moisture gain or loss of the packaged tablets for all three blister package types was determined by the following equation. Pf (1+IMC) - 1 EMC (g H20/100 g dry wt) - [ ] x 100 Pi Where Pf - final product wt (g) Pi - initial product wt (g) IHC - initial moisture content (g HZO/g dry wt of product). 25 EMC values are reported as an average of six blister strips for each package system. A total of 48 strips with 10 blister cavities per strip were used for the above determination. Storage stability studies of bulk packages (30 and 50 count): For each bulk package system, 5 containers were filled with Ibuprofen tablets and stored at the respective storage conditions. These containers were designated test containers. Concurrent with initiating the storage studies an appropriate number of sealed bulk containers (14 containers at each condition of storage) were also placed at the specified storage conditions, and served as positive controls. The moisture content of the drug in the three blister packages, as well as the moisture content of the product in the bulk containers cycled through repetitive opening and closure cycles, were compared to that in the sealed control containers over the entire period of storage (i.e., 180 days). At the end of predetermined time intervals (i.e., 3 days), the test bulk containers were opened in the respective storage enviroments and after standing open for a period of time (i.e., 2 minutes), the containers were resealed at the appropriate application torque. All of the above steps were carried out at the constant conditions of storage. The opening and closing sequence was repeated at 3 day intervals over the entire storage period of 180 days. Throughout the course of this study, the application and removal torques were monitored, the application torques for the 24 and 28 mm closures being 12 and 15 lbf-in (0.06 N.M) respectively. The average removal torques for the 30 count (24 mm) and 50 count (28 mm) containers were found to be 7 and 11 lbf-in (0.06 N.M) at 72°F/50 8 RH and 7.2 and 11.2 lbf-in (0.06 N.M) at 78°F/85 8 RH,respectively. Removal torque data is summarized in Appendix A. At 30 day intervals, 3 tablets were removed from each of the 5 test containers at the respective storage conditions, and the moisture content determined by Karl Fischer titration. This was done for the entire six months of the study Of the 14 containers serving as bulk controls, 2 containers were opened at the end of each month and the tablets analyzed by the Karl Fischer method for moisture content. The opened bulk containers were then discarded. A total of 76 containers, including test and positive controls, were used in this aspect of the study. 4.0 RESULTS AND DISCUSSIONS 4.1 Initial Moisture Content: The Karl Fischer titrimetric method was utilized for determination of the initial moisture content of the product. The initial moisture content provides a reference basis for the entire study. Therefore, the accuracy of determination is very important. Fresh Karl Fischer reagent was used for every analysis and the Karl Fischer titer value was frequently checked, as the reagent shows a gradual decrease in strength with time. The data obtained for initial moisture content of Ibuprofen tablets are presented in Table 2. Table 2. Initial Moisture Content of Ibuprofen Tablets. sample no. 8 water moisture content (g H20/100 g dry product). l. 3.22 3.33 2. 3.22 3.33 3. 3.20 3.30 4. 3.19 3.30 5. 3.19 3.30 average 3.3 i 0.016 27 28 The average initial moisture content determined for the Ibuprofen tablets was 3.3 i 0.016 g H20/100 g dry product. The initial moisture content values obtained, agreed well with the values reported by the manufacturer of this drug,(the same lot), as part of their quality control tests. The specification for moisture content in Ibuprofen tablets, aS‘ specified by the United States Pharmacopeia, is not more than 5 8 (wt/wt), USP XXI (1985). 4.2 Equilibrium Moisture Isotherm: The numerical data for the equilibrium moisture content and associated relative humidity values for the Ibuprofen tablets determined at 72° F and 78° F are summarized in Tables 3 and 4 respectively. For better illustration, the equilibrium sorption data are presented graphically in Figures 1 and 2, where the average equilibrium moisture content is plotted as a function of relative humidity, at the two temperatures of study. 29 Table 3. Equilibrium Moisture Content (EMC) of Ibuprofen tablets at 7 20F. Relative Humidity (8) EMC 0’) (g H20] 100 g dry weight product) 21.2 2.9 32.7 3.3 55.8 4.5 64.1 5.2 75.2 5.9 79.5 6.4 (a) Average relative humidity for the entire period of sorption study. (21 days) (b) Average of 3 replicate samples. Table 4. Equilibrium Moisture Content (EMC) of Ibuprofen tablets at 78°F. Relative Humidity <8) BMC 0’) (g H20] 100 g dry weight product) 23.9 . 3.1 33.2 3.6 54.2 4.6 64.6 5.2 76.9 6.2 79.9 6.5 (a) Average relative humidity for the entire period of sorption study. (21 days) (b) Average of 3 replicate samples. 3O on m Nu. uo totem 3033 535.5. to .5050». sown—om F 0.52... x a £253: 2523. 09 on ON 0' twat-3.388458%? .. SN 86 1M Mo 5 OOl/O‘H 5 ans .._ we so 8.3m 3.333 8.895. to 5658. 8:38 a 83mm x a £253: 2:23. 3 an on o. . 0m 8 F...... . bun-ann—hbb-pbh- bub-hn-bn—D-Ppnnpunt—Ph-nbpbbn—nbbunnbbU—hnb-nnnnp 31 afltaezuuitlltslaesianln .. on... Sfl on." 8d 00$ 86 .86 1M Mo 6 OOL/O°H 6 ans 32 4.3 Water vapor permeability of package: The water vapor transmission rate data for the two bulk packages subjected to opening and closing, versus the closed controls and the three blister package types at both conditions of storage, are presented in Tables 5, 6, 7, 8, 9, 10 and 11 respectively. Figures 3, 4, 5, 6, 7, 8 and 9 show the plots of net weight gain of packaged desiccant and desiccant tablets versus time elapsed for both the bulk packages and the three blister types, at both storage conditions. Water vapor transmission rates (WVTR) for the respective package systems were determined from the slope of the straight line portion of the curves (linear regression), for both conditions of test. By substitution of appropriate values into equation (5), the permeability constants (Pp) were calculated for each package system. Summarized in Tables 12 and 13 are the WVTR and Pp values for each package system at 72° F/50 8 RH and 78° F/85 8 RH. Apparent WVTR and Pp constants are described for the containers that were cycled through opening and closing regimes, since the weight gain for these packages reflects both water pick up as a result of permeability, as well as 33 water gained as a result of sorption, when the container was opened and allowed to stand for a fixed period of time, prior to reclosure. 34 Table 5. Net weight gain of packaged desiccant in 30’s stored at 78°F/ 85% RH Time 30’s open / close 30’s closed (days) (weight gain in grams) (a) (weight gain in grams) (a) 0 0.00000 0.00000 3 0.00606 0.00303 6 0.01211 0.00511 9 0.01804 0.00691 12 0.03038 0.00890 15 0.04155 0.01045 18 0.05276 0.01286 21 0.06213 0.01454 24 0.07232 0.01686 27 0.08582 0.01896 30 0.09713 0.02136 33 0.10459 0.02301 36 0.11282 0.02509 39 0.12180 0.02753 42 0.13036 0.02992 45 0.14062 0.03212 48 0.14913 0.03449 51 0.15974 0.03683 54 0.17107 0.03874 57 0.18246 0.04126 60 0.18673 0.04393 (a) Average of 10 replicate samples. 35 Table 6. Net weight gain of packaged desiccant in 50’s stored at 78°F/ 85% RH Time 50’s open / close 50’s closed (days) (weight gain in grams) (a) (weight gain in m) (a) 0 0.00000 0.00000 3 0.00713 0.00144 6 0.01896 0.00314 9 0.02383 0.00479 12 0.03555 0.00687 15 0.04763 0.00845 18 0.06020 0.01091 21 0.07047 0.01265 24 0.08252 0.01501 27 0.09478 0.01702 30 0.10876 0.01938 33 0.12105 0.02124 36 0.13323 0.02348 39 0.14843 0.02585 42 0.16030 0.02826 45 0.17162 0.03067 48 0.18529 0.03296 51 0.19875 0.03534 54 0.21420 0.03773 57 0.22862 0.04001 60 0.23271 0.04303 (a) Average of 10 replicate samples. 36 Table 7. Net weight gain of packaged desiccant in 30’s stored at 72°F/ 50% RH Time 30’s open / close 30’s closed (days) (weight gain in grams) (a) (weight gain in grams) (a) 0 0.00000 0.00000 3 0.00445 0.00309 6 0.00707 0.00560 9 0.00924 0.00829 12 0.01108 0.01065 15 0.01287 0.01267 18 0.01501 0.01548 21 0.01646 0.01726 24 0.01802 0.01878 27 0.02006 0.02103 30 0.02159 0.02267 33 0.02333 0.02458 36 0.02518 0.02634 39 0.02663 0.02806 42 0.02810 0.02961 45 0.02972 0.03111 48 0.03112 0.03333 51 0.03259 0.03502 54 0.03404 0.03596 57 0.03566 0.03796 60 0.03687 0.03974 (a) Average of 10 replicate samples. 37 Table 8. Net weight gain of packaged desiccant in 50’s stored at 72°F/ 50% RH Time 50’s open / close 50’s closed (days) (weight gain in grams) (a) (weight gain in grams) (a) 0 0.00000 0.00000 3 0.00238 0.00180 6 0.00456 0.00315 9 0.00645 0.00455 12 0.00834 0.00560 15 0.01010 0.00684 18 0.01205 0.00828 21 0.01423 0.00967 24 0.01538 0.00979 27 0.01755 0.01117 30 0.01910 0.01223 33 0.02095 0.01346 36 0.02260 0.01455 39 0.02445 0.01574 42 0.02640 0.01666 45 0.02799 0.01766 48 0.02973 0.01930 51 0.03143 0.02021 54 0.03288 0.02071 57 0.03470 0.02197 60 0.03579 0.02323 (a) Average of 10 replicate samples. 38 Table 9. Net weight gain of packaged desiccant tablets in PVC blister package at both storage conditions. Time pvc @ 78°F / 85% RH PVC @ 72°F/50% RH (days) (weight gain in grams) (a) (weight gain in grams) (a) 0 0.00000 0.00000 3 0.05620 0.01350 6 0.08534 0.02403 9 0.10377 0.03391 12 0.11730 0.04135 15 0.12784 0.04824 18 0.13890 0.05535 21 0.14630 0.06163 24 0.15423 0.06628 27 0.16091 0.07036 30 0.16663 0.07378 33 0.17111 0.07647 36 0.17564 0.07998 39 0.17972 0.08256 42 0.18258 0.08551 45 0.18546 0.08920 48 0.18735 0.09220 51 0.18916 0.09404 54 0.19047 0.09613 57 0.19166 0.09796 60 0.19329 0.09941 (a) Average of 5 blister strips. 39 Table 10. Net weight gain of packaged desiccant tablets in PVC / Aclar blister package at both storage conditions. Time Aclar@ 78°F/ 85% RH Aclar@ 72°F/50% RH (days) (weight gain in grams) (a) (weight gain in grams) (a) u: 01 won» Sereaaeeamsursaanoawo 0.00000 0.00082 0.00199 0.00327 0.00446 0.00563 0.00702 0.00833 0.00963 0.01099 0.01220 0.01339 0.01472 0.01614 0.01740 0.01878 0.02002 0.02139 0.02269 0.02379 0.02524 0.00000 0.00043 0.00083 0.00126 0.00162 0.00200 0.00240 0.00301 0.003 19 0.00347 0.00400 0.00437 0.“)483 0.00521 0.00561 0.(X)592 0.(X)628 0.00696 0.00701 0.00755 0.00791 (a) Average of 5 blister strips. 40 Table 11. Net weight gain of packaged desiccant tablets in PVC / Saran blister package at both storage conditions. Time Saran @ 78°F / 85% RH Saran @ 72°F / 50% RH (days) (weight gain in grams) (a) (weight gain in grams) (*0 0 0.00000 0.00000 3 0.00018 0.00150 6 0.00103 0.00235 9 0.00231 0.00280 12 0.00337 0.00327 15 0.00444 0.00377 18 0.00576 0111427 21 0.00695 0.00490 24 0.00823 0.00513 27 0.00941 0.00563 30 0.01077 0.00582 33 0.01174 0.00618 36 0.01302 0.(X)664 39 0.01435 0.00700 42 0.01560 0.00756 45 0.01695 0.00801 48 0.01821 0.00848 51 0.01937 0.00880 54 0.02058 0.00913 57 0.02180 0.00935 60 0.02306 0.00982 (a) Average of 5 blister strips. 41 mm x mm\ abs so 88% pan to E; .n 88m «>00 e_ 05:. 86a 865 8.0a — gfih - 8.00s - 860w 6111 u! peugos aimsgow 8 a t. I —8 -s 1.. 1-8' #8 --8 a 1. a l—g . . _O E E u N .. 8% _._. g R e i . 8‘3 a - H i s“: leUlFlUIIIIUIITTT—[FllllfiTllTTt‘lj—TITIT—IIUITII 8.88.88. 8.88.8 ggsgs sass N '- 125.00 6111 u! paugos amislow 8 :5 me in Days 11 wer of 50's Stored at 78°F/85 7: RH Figure 4. 8 1 --3 1. e9 ~s -3 *3 1-8 p. ~92 ii In 1_ i8 1.. fir: " 13 L. 11 _. 515 1' an 8firrllUFIITTFéUIlislilfigIIIWélUTjgjlilé'TTIgUTUIs i i i i 3 a i i i i . 6w U1 peugos sinisgow WVTR of 30's Stored ot 72‘F/50 2 RH Figure 5. 3. IIITITUIU1IIIUIIUIIIITNTTIFTTIITTIIIr'lTTTIIIIUTT 888888888 a‘éi‘iags s =1 6m 11; peugoe sinisiow Time in Days WVTR of 50's Stored at 72'F/50 3 RH Figure 6. 45 5?. I 88 II a 1: ‘1 ll 1 ii 250.00 WIIIITITTTUTTITTWWNTIIIIIIPTTTFTIWITTTTITTTU 8 8 8 8 8 8 8 8 8 a 3 a a .s. a s i o Bur u; peugoo sinisgow Time in Days WVTR of Aclar Blisters 18 Figure 7. 46 8 —8 1. ~51 -s L. 1-8 —8 —:. —s H I" .- s is g... as as . _ fl . _. £2 . . I! Tfljl‘jilliIUIIIIUIWIlT—Tr"III'UITIIUUUlrITIITTllf-Q 8 8 8 8 g 8 8 8 8 8 8 11 32:4 N a 813 a :1 6111 u! paupo emisgow Time in Days WVTR of PVC Blisters Figure 8. 47 250.” - 225.00 200.00 1 25.00 75.00 3. 3. § .8. 6w 1.1; peuios emisgow 1 75.00 25.00 “lime in Days WVTR of Saran Blisters Figure 9. 48 Table 12. Water Vapor Transmission Rates & Permeability constants of the three blister packages at both conditions of storage. Blister WVTR Permeability Constant Package g 1120/ day.b1ister cavity g H20 / day.mmHg.blister cavity Type 72°F/50% RH 78°F/85% RH 72°F 78°F pvc 57.2 x 10'5 28.4 x 10 '4 5.7 x 10 '5 1.2 x 104 Aclar 26.2 x 10“5 84.9 x 10 4 2.6 x 10 ‘5 4.1 x 10 ‘6 Saran 31.0 x 10 *5 79.0 x 10‘6 3.1 x 10 '5 3.8 x 10 '5 Table 13. Apparent Water Vapor Transmission Rates & Permeability constants of the 30 & 50 count containers at both conditions of storage. Apparent WVTR Apparent Permeability Constant Bulk g H20 / daypackage g H20 / day.mmngackage Package T . y” 72°F/50% RH 78°F/85% RH 72 F 78°F 30’s closed 67.2 x 10'5 69.4 x 10 ‘5 6.7 x 10 '5 3.3 x 10 '5 30’s open 60.8 x 10 ‘5 32.8 x 10'4 6.0 x 10 ‘5 1.6 x 10'4 50’s closed 38.2 x 10'5 68.8 x 10'5 3.8 x 10 '5 3.3 x 10'5 50’s open 60.9 x 10‘5 39.2 x 10 '4 6.0 x 10 '5 1.9 x 10'4 49 4.4 Storage Stability Studies: The equilibrium moisture content (EMC) of Ibuprofen tablets in all five package types was determined at intervals of 15 and 30 days, for the entire period of study (i.e., 180 days). The equilibrium moisture content values obtained are summarized in Tables 14, 15, 16 and 17 respectively. The impact of the storage environment on the moisture content of the Ibuprofen tablets is clearly seen from these data. For the bulk containers, the gain in moisture was relatively low at ambient storage conditions (i.e., 72° F/50 8 RH). However, at the slightly accelerated storage conditions (i.e., 78° F/85 8 RH), the moisture permeating through the container / closure system plus the moisture entering the package head space when the containers were opened and closed, caused a significant increase in equilibrium moisture content of the Ibuprofen tablets. The moisture content determinations of Ibuprofen tablets in the bulk containers (30 and 50 count) were considered to be accurate when compared to values obtained for the same samples analyzed by a similar procedure at another test facility. The drug tablets in the three blister package systems was also analyzed for moisture content at the end of 180 days by the Karl Fischer method, and were found to give good agreement with moisture content values determined by the gravimetric procedure. 50 The average EMC (gms of H20/100 gms dry wt of product) of Ibuprofen tablets, in Aclar, PVC and Saran blisters as determined by the gravimetric method, was found to be 4.65, 7.66, and 4.89 respectively, at the end of 180 days of storage at 78° F/85 8 RH. The EMC values, as determined by the Karl Fischer method, were 4.60, 7.60 and 4.81 respectively. The effect of repetitive opening and closing on the moisture content of the product in the 30 and 50 count bulk containers at both conditions of storage are plotted in Figures 10, 11, 12 and 13 respectively. Superimposed on these plots is the moisture uptake data for the unopened controls, stored under similar conditions. The relative performance of the 50 count and 30 count containers versus the Aclar, PVC and Saran blister packages is shown graphically in Figures 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24 and 25 respectively, where the EMC of the Ibuprofen tablets is plotted as a function of storage time for the respective package systems, at both conditions of test. To simulate end use conditions, the storage data for the bulk packages subjected to the opening and closing cycle were used in the above comparisons. 51 To compare the relative performance of each blister type package at the two conditions of storage (i.e., 72° F/50 8 RH and 78° F/85 8 RH) the storage stability data for the various blister package paired combinations were plotted and the results presented in Figures 26, 27, 28, 29, 30 and 31 respectively. The comparative performance of the three blister package systems with Ibuprofen tablets at 78° F/85 8 RH and 72° F /50 8 RH, as a function of storage time, is plotted in Figures 32 and 33. When the moisture content of Ibuprofen tablets in the three blister package types was compared as a function of time and storage conditions, the moisture content of the tablets in PVC blisters exceeded the maximum USP recommended levels of moisture content within 15 days at 78° F/85 8 RH, while the tablets in Aclar and Saran blister package types did not gain moisture in excess of the maximum recommended levels at the end of 180 days of storage, at these conditions. Further, the product in Aclar blisters appeared to have gained less moisture than that in Saran blisters, at 78° F/85 8 RH. The relative performance of the 50 and 30 count bulk containers subjected to repetitive opening and closing, 52 versus the Aclar, PVC and Saran blister type packages at 78° F /85 8 RH showed the moisture content of Ibuprofen tablets to exceed the maximum USP recommended levels at the end of 180 days, while the moisture content of the same product in Aclar and Saran blisters at the end of 180 days had not yet exceeded the maximum acceptable moisture level (USP XXI 1985). Similar comparisons of moisture content of Ibuprofen tablets in bulk containers (30 and 50 count containers), subjected to repetitive opening and closing, versus the unopened controls and the moisture content of Ibuprofen tablets in the three blister package systems showed no significant difference at ambient storage conditions.(i.e., 72° F/50 8 RH). On visual examination the Ibuprofen tablets did not exhibit any physical changes. 53 Table 14. Equilibrium Moisture Content (EMC) of Ibuprofen tablets in Aclar, PVC, and Saran blister packages at 78 °F/ 85% RH. (a) Time Aclar PVC Saran (days) EMC 0’) EMC 0’) EMC 0’) 0 3.30 3.30 3.30 15 3.49 5.65 3.68 30 3.64 6.25 3.86 45 3.78 6.62 4.02 60 3.90 6.95 4.17 75 4.02 7.13 4.29 90 4.13 7.21 4.39 105 4.22 7.27 4.49 120 4.32 7.34 4.57 135 4.41 7.43 4.66 150 4.49 7.49 4.74 165 4.58 7.61 4.82 180 4.65 7.66 4.89 (a) Average of 6 blister strips. (b) Grams of H20] 100g dry weight of product. 54 Table 15. Equilibrium Moisture Content (EMC) of Ibuprofen tablets in Aclar , PVC, and Saran blister packages at 72 0F / 50% RH. (11) Time Aclar PVC Saran (days) EMC 0’) EMC 0’) EMC 0’) 0 3.30 3.30 3.30 15 3.32 3.50 3.33 30 3.34 3.52 3.35 45 3.36 3.57 3.37 60 3.38 3.52 3.38 75 3.39 ' 3.51 3.40 90 3.41 3.50 3.41 105 3.42 3.52 3.42 120 . 3.43 3.52 3.43 135 3.44 3.49 3.44 150 3.46 3.50 3.45 165 3.47 3.50 3.46 180 3.48 3.50 3.47 (a) Average of 6 blister strips. (b) Grams of H20/ 100g dry weight of product. 55 Table 16. Equilibrium Moisture Content (EMC) of Ibuprofen tablets in 50 a 30 count containers at 78°F] 85% RH. (c) Time 50’s 50’s 30’s 30’s open I close Closed open I close Closed (days) EMC(a) EMC(b) EMC(a) EMC(b) 0 3.30 3.30 3.30 3.30 30 3.81 3.71 3.73 3.68 60 3.95 3.79 3.90 3.70 90 4.29 4.10 4.29 4.05 120 450 4.00 4.47 4.12 150 4.90 4.20 4.80 4.13 180 5.53 4.44 5.36 4.13 Table 17. Equilibrium Moisture Content (EMC) of Ibuprofen tabletsinsoasoeountooutainetsat72°PI50%RH(c) Time 50’: 50’s 30’s 30’: open I close Closed open I close Closed (days) EMC(I) EMC(b) EMC(a) EMC(b) 0 3.30 3.30 3.30 3.30 30 3.54 3.45 3.45 3.45 60 3.55 3.47 3.50 3.47 90 3.54 3.48 3.50 3.48 120 3.55 3.50 3.49 3.48 150 3.57 3.50 3.50 3.48 180 3.57 3.52 3.50 3.49 (a) Average of 5 replicate samples. (b) Average of 2 replicate samples. (c) Grams of H20] 100 g dry weight of product. 56 180 120 an TTTIITITerTTTTIlTTjIllTrlTi—IIIITITIITTTTINT 8 8 8 8 § 8 3. 8 8 8 r4 14 6 d to e <- n n 111 M0 5 001/0‘H 5 0113 Time in Days EMC of Ibuprofen, Tablets in 30's Stored at 78 F/85 2 RH Figure 10 57 3 e3 —§ -8 _8 ii Rte 88 R .5 L8 55 an UTTTIITIIIIIIIIT1IIIIIIITITjTTI[IIIIITTTTIWT E ii i i i 5. i 3 5‘. 8 1M 100 5 001/0‘H 5 0113 Time in Days EMC of lbuproferggablets in 50's Stored at 78 /85 2 RH Figure 1 1 58 100 Jo Jo Time in Days EMC of lbuprofemgablets in 30's Stored at 72 /50 2 RH t-8 1-8 Figure 12 FITTIUIUTITITTWITTTIITITrlITIUIIITIIIITTIITTT S 8 5' 8 as 33513 1M 440 5 OOL/O°H 5 0113 3 '0’ § 3 .5 59 .3. 11 ti .1... 111-: if "It 3% 1': as F3 ii an ITIIUTIWTTIITIijlITITTIITITITIITHITTTTTIéTT 3 § 3 3 .81 i i i a 5 1M 440 6 001/0‘H 5 0713 Time in Days EMC of lbu blets in 50's 0 2 RH #75 refs pat 7 Stored Figure 13 60 100 o'o Time in Days 50 count stored at 78 F/85 2 RH EMC of Ibuprofen Tablets in.Aclar Blisters vs Figure 14 TTIUITITTIII—TTIITTTTIIITIIITUIIIUITrIIIIIITIT 8 3 8 8 8 8 8 8 8 8 r~° t4 6 d 16 in w- 9 vi n 111 Ma 5 OOL/O°H 5 ONE! 61 mm x mm}: as team o>a goo 5 652. on _ wk “—0 “COLON—m “—5400 on _ 38.32 5863. to 05 mp 83E 1-8 range-st 88.8 a inaugural-Ia... - and 5.333. 8 iii M Ma 5 00 t/o‘H 6 0W3 8 a I Til TNT—r TTII IFTT III! fiil TjIT 111 h T l l l l l l .- 8 8 h 62 ID ITIIIIIITTIITTT—TIITIIIITITITrUTlIIITIWTTIIIIUI 8 S 8 8 8 a ‘3 5 5 8 .t .. .- a 1M 150 5 001/0‘H 5 ONE) 5. Time in Days Saran Blisters vs °,.- /85 7: RH In rofen Tablets t stored at 78 50 copun EMC of lbu Figure 1 6 o 0.6 n :33 m> EH?” momk$mm www.mfl. wuvwaaawm 02m 5 95mm «>00 5 2.5... 63 U'ITleUVUIlUTYIIIITIIIYIIIIIITIIIIIITjTTITf § 3 § § § E- 3 5 M No 5 OOl/O‘H 5 am i5 8. B m n 0 So m :38 m> CHEW oguamhmwwzflh aching. mm 02m 2 230E 960 c_ Etc. 4. a... 9.. on. innate... inn-ho:- [ITTITIfi 3. ¢ IIWIFTjiTTUITTTIIIIIU[IUIIIITTI 5898.. Bill...- 8 a 3. '0 § § § § 3M Mo 5 OOl/O‘H 5 am 8, 5 § 6S T T111 UTTI 1111 1111 1111 1T1] 1111 1111 111T l l I l I l l l T 8 S 8 8 8 S 8 8 8 g N Is 6 6 I5 Iri *- «6 vi 3M 100 5 OOL/O‘H 6 am Time in Days isters vs Saran Bl /85 3: RH 1.- rofen Tablets in t stored at 78 p can EMC of lbu 30 c Figure 19 66 l 120 F8 Ill 1'11Ill[TWIII111II1ITTT11IIIjITITITIIIITIfTT 3 8 3 3 8 3 3. 3. 3. 5’2 is h 6 o In I: v «r n n 1M Ma 5 OOL/O‘H 5 am Time in Days 50 count stored at 72 F/50 2 RH EMC of Ibuprofen Tablets in,Aclar Blisters vs Figure 20 67 :m N on\.._ we so 85% e58 on Enemies... Sac-.8 a Euonbfis £3.82... ._ m> 9.32m o>a .E 3038 c3963. ”.0 9.5 K 059E 960 E acct. V cm. ow. on 0.” l_\n +rl lulfil Hui W ”a, n IIIII[‘1111ITIIIIIIIIIIIIITI[1111111[1111111 :3 O ‘2 n o "2 * O ‘9 v- o '7 In C ‘9 In 0 n. «a E O "2 f\ 1M Ma 5 OOL/O‘H 5 am 68 [-9 1m R 8b.? so “62on .58 om gm 58m 5 322: 5353. so 05 m> mama ON— _ *1 null mxao c_ 6E2. i [T mm 8%: inanbfis $898 a Enoobfisnlsliam .. 1M 150 5 OOL/O‘H 5 am 69 :1 N on n. NH “a 683m uEsau m> muoum=m La_o\m .E 3238 5353. am 05 em 2%: 960 E mEP _ _ [F k fr incubate inc-.8 n zznonbfisplil 2c ._ X and in LH Jul 1.” l O '3 In a '2 a: O "2 B O ‘9 [x IIIIIIIIIIUTTITTTII11 M Ma 5 OOL/O‘H 5 am O ’ Im N on awn ya 623m €300 0 m> 83m=m caumm. E 3032. c.8952”: “no 02m mN 0..ng 96o E mEF TV 71 _ _ _ _ X3.” llml EH lrll T. Enonbfis 88.8 a Enonbustig ._ o '1 * o “i * 8 in TM Ma 5 OOL/O‘H 5 am 0 ‘7! In § 8 h r1111IIIIIIITIII]1111]1111l1T111111111r1l1r1 § :m N mm\.._.mn so 2:3» 2335 o>n_ m) 6.2 5 3033 5333. so 02m 280 c_ 0E2. on — ON — ow 00 on _ p b _ _ z: a no}. 2. a mm 059... plan—Iota Enflbfiiii .. 3M Ma 5 OOL/O‘H 5 am E N 3&4 n so .022» 2325 cauam m> La_o< . 3032. 535.5. *0 02m mm 0.59... 280 c_ 6E5. on. on. on. 8 8 on _ _ _ _ _ 73 Eunoxufliputl EnnoDESplil Elana Ball 2M 150 5 OOL/O‘H 5 am 71+ a a ., 4.3 a -§ o > a O I ‘8 .5 o 5.5 l— *8 ii an §§ RR fl --8 II Tér1h1111 [$1111é11TléU rrrg1111§T11111 11 léi F1 - :4 § 6 «5 us as 6 § vi 5; 3M Ma 5 OOL/O‘H 5 ans 5'? PVC vs Saran tored at 78 /85 2 RH rofen Tablets Blister: s EMC of lbu Figure 28 L8 IIITITIIIIIIIFIIIIIIllrlfilllFlIIIIIIIIJTTIIT [x 4M 150 5 OOL/O‘H 5 am 3.30 Time in Days EMC of Ibuprofen Tabletsoin Aclar vs PVC Blisters stored at 72 F /50 7. RH Figure 29 :m "x. on\... am so 8.2” 93.35 76 cauom m> 8.04. c._ 3038. c3053. *0 0.5 on 930E mxao E oEP SF. ow. ow. o_m mm on a and r I lflll lfll ill F tenantfilplilgm a Ewenhustplili .. 111111Wr111fi1111111I1Tllll111l111rl11111111l 3. n o '1 * o ‘9 <- O '1 In 5"; E 2, B 3. 5 W No 5 OOL/O‘H 5 am 77 7zr/soxm nr/saxm at at Hahn i; IIrllI'III1I11rIIfiIITIIIIITIIfijIIITIrIIIIIII 8. 8 8. 8 8. S 8. 8 8. h h D b In In 1' V n M Ma 5 OOL/O‘H 5 am 35- Tlme in Days EMC of Ibuprofen Tablets Jn PVC vs Saran Blisters stored at 72 F /50 7. RH Figure 31 78 :m x autumn S 3.3» euém Exam u> 0?. n> .6. < 5 3030... :89. an. *o 26 NM 230E 260 E oEF on. o. 8 8 on . h R r _ h and T we." 1 r Ian... 1 ulooé n “land Woo... was T load a n muons funniest-uili...» 4 n inns-2o pails... u Yseas frantic-ill! .. 1 1M Ma 5 OOL/O‘H 5 am IN N on .._ an so nee” m.» m. caeam m> o>a m> L994 5 3038. con—warm“: h_o 02m nn 050E 980 E oEc. _ 79 F5 ~53 5% 8 »—s \[a Easxufii-Iifigm incubus... £8.83; :muonbussalflg Ifl‘ 1'1ITIIIIITTI1TII1I1TIII11]'111r1rfi11illl11 § 3. n 5- O '9 1' O .0 Id 0 Q In a '2 a) E O '2 [x 1M Ma 5 OOL/O‘H 5 am 80 4.5 Statistical Analysis: To test the equality of moisture uptake for the five different package types, a one - tailed test was used, with the standard t - statistic (Bhattacharyya 1977). Presented in Table 18, is a summary of the statistical analysis of moisture uptake data, collected on day 180 for Ibuprofen tablets packaged in the bulk and blister packages, at both conditions of storage. The effect of repetitive opening and closing on the moisture content of the product in the 30 and 50 count bulk containers at 78‘ F/85 % RH, versus the product in the unopened bulk containers (controls) at the same conditions of storage, showed a statistically significant difference at the 99.5 % confidence level. The EMC of the Ibuprofen tablets in the 30 and 50 count bulk containers subjected to the opening/closing cycle, exceeded the maximum USP recommended levels, while the moisture content of the tablets in the unopened controls were still below the maximum USP recommended levels, at the end of 180 days of storage. At the ambient storage conditions, i.e., 72° F/SO % RH, there was no significant difference in the EMC of Ibuprofen tablets subjected to the same treatment as above, for the 30 and 50 count containers. 81 The EMC of Ibuprofen tablets stored at 78° F/85 % RH in the PVC blister package versus the EMC of the product in the Aclar and Saran blister packages showed a significant difference, at a confidence level greater than 99.5 %. At the ambient storage conditions however there was no significant difference, when the same blister package systems were compared. The EMC of Ibuprofen tablets at 78° F/85 % RH, in the 50's and 30's bulk containers, subjected to the opening/closing regime versus the same product in the Aclar and Saran type blister packages, showed a significant difference at the confidence level greater than 99.5 %. The converse is true in the case of the EMC of Ibuprofen tablets in the PVC blister package versus the EMC of the product in the 30 and 50 count bulk containers subjected to the same opening/closing cycle, which showed a significant difference at the confidence level greater than 99.5 %. At the ambient storage conditions of 72° F/SO % RH the EMC of the Ibuprofen tablets in the 50 count bulk containers subjected to the opening/closing cycle versus the product in Aclar blister type packages showed a significant difference at the confidence levels between 95 % and 97.5 %. The EMC of Ibuprofen tablets in the 50 count containers subjected to the same treatment, when compared to the product in the Saran blister type packages, showed a significant difference at the 82 confidence levels between 97.5 % and 99.5 %. In the case of the product in 50's open versus product in PVC blister type packages, there was no significant difference at the 95 % confidence level. In comparing the EMC values of Ibuprofen tablets in 30 count bulk containers at ambient storage conditions subjected to the opening/ closing cycle, versus the product in Aclar, Saran and PVC blister type packages, there was no statistically significant difference. 83 Table 18. Summary of the statistical analysis of bulk packages (30 and 50 count) and blister packages (PVC, Aclar, Saran) at both conditions of storage. Package Types being Storage "t"-values, degree Conclusion Confidence compared Conditions of freedom Level 1. 30’s open vs. 30’s closed 78F / 85% RH 21.012 5 Significant Greater than Difference 995% 2. 50’s open vs. 50’: closed 78F/ 85% RH 6.64 5 Significant Greater Difference than 99.5% 3. 30’s open vs. 30’: closed 72F] 50% RH 0.101 5 No Significant _ Difference 4. 50’: open vs. 50’: closed 72F] 50% RH 0.044 10 No Significant __ Difi'erence 5. PVC blisters vs. Aclar 78F/ 85% RH 30.56 10 Significant Greater than blisters Difference 995% 6. Saran blisters vs. Aclar 78F] 85% RH 2.883 10 Significant Between 99% blisters Difference and 99.5% 7. PVC blisters vs. Saran 78F] 85% RH 24.152 10 Significant Greater than blisters Difference 99.5% 8. PVC blisters vs. Ads: 72?] 50% RH 0.44 10 No Significant __ blisters Difference 9. Saran blisters vs. Aclar 72F] 50% RH 0.990 10 No Significant _ blisters Difference 10. PVC blisters vs. Saran 72F / 50% RH 0.617 10 No Significant _ blisters Difference 11. 50’: open vs. Aclar 78F] 85% RH 8.905 9 Significant Greater than blim's Difference 99.5% 12. 50's opal vs. Saran 78F] 85% RH 5.393 9 Significant Greater than blisus Difference 995% Table 18. (cont’d) Package Types being Storage ”t"-valuea. degree Conclusion Confidence compared Conditions of freedom Level 13. PVC blisters vs. 50’s 78F/ 85% RH 16.089 9 Significant Greater than open Difierence 99.5% 14. 30’s open vs. Aclar 78F] 85% RH 13.721 9 Significant Greater than blisters Difference 99.5% 15. 30’s open vs. Saran 78F] 85% RH 5.661 9 Significant Greater than blisters Difference 99.5% 16. PVC blisters vs. 30’: 78F] 85% RH 22.154 9 Significant Greaterthan open Difi'erence 99.5% 17. 50’: open vs. Aclar 72F / 50% RH 2.227 9 Significant Between 95% blisters Difference and 97.5% 18. 50’s open vs. Saran 72F / 50% RH 2.437 9 Significant Between 97 5% blisters Difference and 99.5% 19. 50's open vs. PVC 72F / 50% RH 0.907 9 No Significant _ blisters - Difference 20. 30’s open vs. Aclar 72F l 50% RH 0.545 9 No Significant _ blisters Difference 21. 30's open vs. Saran 72F/ 50% RH 0.779 9 No Significant _ blisters Difference 22. 30's open vs. PVC 72F] 50% RH 0.04 9 No Significant __ blisters Difference 5.0 CONCLUSION The effect of repetitive opening and closing of the 50 and 30 count bulk containers showed a significant gain in equilibrium moisture content of the Ibuprofen tablets at the abusive storage conditions (i.e., 78' F/85 % RH) when compared to the unopened containers (controls) of the product. At the ambient storage conditions (i.e., 72 deg F/SO % RH) the effect of repetitive opening and closing showed no significant difference in the EMC of the Ibuprofen tablets, when compared to the closed controls. At 78° F/85 % RH Ibuprofen tablets in bulk containers (30 and 50 count) subjected to repetitive opening and closing, showed a significant increase in their equilibrium moisture content when compared to the same product in Aclar and Saran type blister packages. In the case of the PVC blister package, the EMC of Ibuprofen tablets increased significantly within fifteen days, when compared to the same product in the bulk containers. At ambient storage conditions (i.e., 72’ F/SO % RH) there was no significant difference between the EMC of Ibuprofen tablets packaged in the bulk containers versus tablets packaged in PVC, Aclar and Saran blister packages respectively. At the abusive storage conditions (i.e. 78° F/85 % RH) the moisture content of Ibuprofen tablets increased with time, 85 86 the greatest increase being in the PVC blister package, followed by the bulk 30 & 50 count containers subjected to repetitive opening and closing. Product in the Aclar blister package showed the least gain in moisture content at these storage conditions. At the ambient storage conditions (i.e.,72° F/SO % RH) there was no marked difference in the gain in the moisture content of Ibuprofen tablets in the bulk and the three blister packages. From this investigation it follows that for bulk containers which are subjected to repetitive opening and closing, and unit dose blisters fabricated from a material such as PVC, moisture vapor can enter the head space of the container and or permeate through the wall of the bulk containers and through the blister structures and affect the drug product, when the packages are stored at abusive conditions. Appendix A 88 Application torque and average removal torque for the 30 and 50 count bulk containers containing Ibuprofen tablets, subjected to repetitive opening and closing for the entire storage period at 72°F / 50% RH. 30 count Application Average Removal containers (a) Torque Torque (lbf-in) (lbf-in) (5) Sample 1 12 6.5 Sample 2 12 7.2 Sample 3 12 7.5 Sample 4 12 6.6 Sample 5 12 7.0 I = 6.96 s.d. = 0.4159 50 count Application Average Removal containers (a) Torque Torque (lbf-in) (lbf-in) (5) Sample 1 15 10.5 Sample 2 15 11.0 Sample 3 15 11.5 Sample 4 15 11.3 Sample 5 15 10.6 I = 10.98 s.d. = 0.4324 (a) Sample numbers refer to individual bulk containers which were subjected to the repetitive opening and closing cycle over the entire 180 days storage period. (b) Removal torque values represent the average value for the individual bulk containers and are the sum of 60 removal torque values. 89 Application torque and average removal torque for the 30 and 50 count bulk containers containing Ibuprofen tablets, subjected to repetitive opening and closing for the entire storage period at 78°F] 85% RH. 30 count Application Average Removal containers (a) Torque Torque (lbf-in) (lbf-in) (b) Sample 1 12 6.8 Sample 2 12 7.5 Sample 3 12 7.5 Sample 4 12 7.0 Sample 5 12 7 .0 :- = 7.16 s.d. = 0.3209 50 count Application Average Removal containers (:1) Torque 1'0qu (lbf-in) (lbf-in) (5) Sample 1 15 10.7 Sample 2 15 11.0 Sample 3 15 11.5 Sample 4 15 11.5 Sample 5 15 11.5 I = 11.24 s.d. = 0.3715 REFERENCES Berendsen, H.J.C. ”Specific Interactions of Water with Biopolymers" , New York , Franks F. Plenum Press , 1975. Bhattacharyya, G.K. and Johnson, R.A. "Statistical Concepts and Methods" , New York , John Wiley and Sons. INC., 1977. Brunauer, S. , Emmett, P.H. , and Teller, E. ”Adsorption of Cases in Multimolecular Layers” , Journal of the American Chemical Society, 60,1938. Bull, H.B. "Adsorption of Water Vapor by Proteins” , Journal of the American Chemical Society, 66,1944. Connors, K.A. , Amidon, C.L. and Kennon, L. "Chemical Stability of Pharmaceuticals” , New York., John Wiley and Sons , INC., 1979. Cooper, J. ”Plastic Containers for Pharmaceutical Testing and Control" , Geneva , World Health Organization , 1974. FDA, “Human and Veterinary Drugs, Good Manufacturing Practices and Proposed Exemptions for Certain OTC Products", Federal Register , 43,1978. Lea, C.H. , Moran, T. and Smith, J.A.B. "The Gas Packing and Storage of Milk Powder" , Journal of Dairy Research , 13,1943. Lechert, H. et a1. 'NMR - Study of Water in Some Starches and Vegetables”, Journal Food Proc. Pres, 3,1979. Lieberman, H.A. and Lachman, L. "Pharmaceutical Dosage Forms: Tablets” , New York, Marcel Dekker, INC., 1980. Mclaren, A.D. and Rowen, J.W. "Sorption of Water Vapor by Proteins and Polymers: A Review" , Journal of Polymer Science, 7,1951. Mitchell, J. and Smith, D.M. 'Aquametry: A Treatise on Methods for the Determination of Water" , New York, John Wiley and Sons. INC., 1980. Pauling, L. ”The Adsorption of Water by Proteins" , Journal of the American Chemical Society, 67,1945. 9O 91 Rockland, L.B. "A New Treatment of Hygroscopic Equilibria: Application to Walnuts (Juglans Regia) and other Foods" , Food Research, 22, 1957. Rockland, L.B. and Beuchat, L.R. "Water Activity: Theory and Applications to Food" , New York, Marcel Dekker, INC., 1987. Saravacos, G.D. "Sorption and Diffusion of Water in Dry Soybeans“ , Food Technology, 23,1969. Taborsky - Urdinola, C.J. , Gray, V.A. and Grady, L.T. "Effects of Packaging and Storage on the Dissolution of Model Prednisone Tablets” , American Journal of Hospital Pharm. , 38,1981. United States Pharmacopeial Convention, INC., United States PharmacOpeia, Washington D.C. (U.S.P. XIV), November, 1950. United States Pharmacopeial Convention, INC., United States Pharmacopeia, Rockville Maryland. (U.S.P. XXI), January 1985. Van den Berg, C. and Bruin, S. ”In Water Activity: Influences on Food Quality” , Edited by L.B. Rookland and G.F. Stewart, New York, P.L. Academic Press, 1981. Viellart, M. et a1. ”Moisture Transfer Tests in Blister Package Testing” , Drug Dev. Ind. Pharm. , 5,1979. Wang, M.J. ”Prediction of the Moisture Uptake by a Packaged Moisture Sensitive Pharmaceutical Product Stored Under Fluctuating Temperature and Humidity Environments” , M.S. Thesis, Michigan State University, 1985. "I111111111111115