`
`ISSN 0001-6837
`Polish Pharmaceutical Society
`
`DRUG TECHNOLOGY
`
`THE PHYSICAL CHARACTERISTICS OF LYOPHILIZED TABLETS
`CONTAINING A MODEL DRUG IN DIFFERENT CHEMICAL FORMS
`AND CONCENTRATIONS
`
`MA£GORZATA SZNITOWSKA*, MARCIN P£ACZEK and MA£GORZATA KLUNDER
`
`Department of Pharmaceutical Technology, Medical University of GdaÒsk,
`107 Hallera Str., 80-416 GdaÒsk, Poland
`
`Abstract: Orodispersible tablets, usually prepared using freezeñdrying method, are becoming a popular drug
`formulation for patients who have difficulties swallowing solid dosage forms. The influence of drug solubility
`and concentration on the physical characteristics of lyophilized tablets composed of mannitol and gelatin was
`investigated. Phenobarbital and phenobarbital sodium were studied as model drugs. The tablets were analyzed
`for mechanical strength using a new method employing Instron, a material testing apparatus. For tablets con-
`taining phenobarbital in the form of sodium salt, better mechanical strength was demonstrated than for tablets
`prepared with water insoluble phenobarbital ñ acid form. Besides, the mechanical characteristics of the tablets
`indicate that plasticity and porosity were reduced when sodium phenobarbital was incorporated at a higher dose.
`Lyophilized tablets were not hygroscopic and only a small increase of tablet mass by 1% and 3% was observed
`after 4 weeks of storage at 40% and 60% RH, respectively. All formulations disintegrated in seconds in water,
`at a temperature of 37OC.
`
`Keywords: lyophilization; freeze-dried tablets; mannitol; gelatin; phenobarbital
`
`Oral route of drug administration is convenient
`and preferred by patients and this is why tablets and
`capsules are the most popular pharmaceutical
`dosage forms. However, sometimes patients, espe-
`cially children and elderly people, may have diffi-
`culty swallowing solid dosage forms. It is estimated
`that 50% of population is affected by this problem,
`which results in a high incidence of non-compliance
`and ineffective therapy (1). The problem can be
`solved by producing orodispersible forms (Ph. Eur.),
`which placed in the mouth, allow to rapidly disperse
`or dissolve in the saliva and then can be swallowed
`as a liquid, in the normal way. Such systems in the
`literature are called: FMTs (fast-melting tablets),
`ODT (orally disintegrating tablets) or FDDF (fast
`dissolving/disintegrating dosage forms).
`There are three major methods for manufactur-
`ing this kind of tablets: 1) freeze-drying, 2) mould-
`ing by compression or heat-moulding, 3) direct
`compression. Because of the high porosity, freeze-
`dried tablets disintegrate in oral cavity faster than
`other systems. Unfortunately, lyophilization, as a
`method of tablet preparation, has also disadvan-
`tages: lack of physical resistance, hygroscopicity of
`the product, as well as high cost of production and
`low dose of the drug which can be incorporated (2).
`
`* Corresponding author: msznito@amg.gda.pl
`
`Typical freeze-dried tablet consists of a drug
`enclosed in a water soluble matrix made of a
`hydrophilic structure-forming polymer (usually gel-
`atin) and a filler (saccharide ñ usually mannitol).
`Other ingredients of the tablets may be sweetening
`agents (aspartam), taste-masking additives and
`preservatives (3). At present, many researchers are
`studying the possibilities to utilize other matrix
`forming agents in the FDDF formulations. These
`are: maltodextrins, different kinds of gelatins, xan-
`tan gum and cellulose-derivatives (4). The type and
`amount of the excipients have a significant influence
`on the characteristics of the lyophilized tablets,
`however, the incorporated drug can also modify
`properties of tablets to a great extend.
`The aim of the study was to determine the influ-
`ence of concentration and chemical form of the drug on
`physical properties of the lyophilized tablets containing
`mannitol and gelatin. Phenobarbital and its sodium salt
`were used as model drugs. Besides, the applicability of
`a material testing machine Instron to determine the
`mechanical strength of the tablets was evaluated.
`
`EXPERIMENTAL
`
`Preparation of tablets
`Tablets were made by freeze-drying. Gelatin
`(I.G.G. Eberbach, Germany) and mannitol (POCh
`
`25
`
`Mylan Ex 1038, Page 1
`
`
`
`26
`
`MA£GORZATA SZNITOWSKA et al.
`
`Gliwice, Poland) were dissolved in purified water at
`60oC. Phenobarbital, F (donated by Galenus,
`Warsaw, Poland) and
`its sodium salt, FNa
`(Tarchomin Pharmaceutical Works, Warsaw,
`Poland) were introduced at various concentrations to
`the tablet solution with constant mixing. The com-
`positions of different formulations, both with and
`without drug, are shown in Table 1. FNa dissolved
`completely at all concentrations, but with F suspen-
`sion was formed, when the concentration exceeded
`0.1% (w/w). The pH of FNa solution was in the
`range 7.6-9.2 and pH of the F dispersions was 5.1.
`Freeze-dried tablets were prepared as follows: the
`wells of PVC blisters (20 mm in diameters) were
`filled with 2.0 ml of solution or suspension and the
`blisters were placed immediately on the pre-frozen
`shelves of the freeze-dryer (Alpha 2-4, Christ,
`Osterode, Germany). The samples were frozen at
`ñ 45oC and kept at this temperature for 60 minutes.
`Primary drying was performed for 40 h at a pressure
`of 0.08 mbar with gradual increasing the shelf tem-
`perature to 20oC. In the second drying step the shelf
`temperature was increased to 35oC for 2 hours.
`
`Visual inspection of tablets
`Morphological characterization of the freeze-
`dried tablets comprised of: shape, color, surface and
`friability analysis. Adhesion of tablets to the blister
`
`and the easiness of taking them out were also evalu-
`ated.
`
`Mechanical strength testing
`A compression testing machine Instron (Model
`5543, Instron Corporation, Canton, USA) was used
`to determine the tablets strength. Each tablet was
`placed horizontally on the lower support and com-
`pressed using a constant speed of 0.22 mm/min to
`the maximum force of 80 N. Relationship between
`the load (N) and deformation of the tablets (%) was
`studied. Tablets containing different amounts of F or
`FNa as well as tablets lyophilized in different loca-
`tions on the freeze-dryer shelf were compared. From
`each formulation 10 tablets were analyzed to evalu-
`ate reproducibility of the results.
`
`Stability of tablets at different relative humidity
`Tablets with F or FNa (n = 6) were placed in a
`hygrostat and stored at relative humidities (RH)
`45%, 60% and 90% over sulphuric acid solutions
`(48%, 38% and 17%, respectively) for 4 weeks.
`Changes in tablet mass and appearance were
`observed.
`
`Disintegration test
`For determination of disintegration time of
`tablets two different methods were employed. In the
`
`Table 1. The visual characteristics of freeze-dried tablets: placebo and containing phenobarbital (F) or phenobarbital sodium
`(FNa) as model drugs
`
`Composition of solution or suspension
`subjected to freeze-drying [%]
`gelatin mannitol
`drug
`
`Drug content
`per tablet [mg]
`
`Surface of tablets Durability of tablets Removal from
`the blisters
`
`0.5
`0.75
`1.0
`2.0
`2.0
`
`5,0
`
`3.0
`
`1.0
`
`5
`
`-
`
`0.1 FNa (solution)
`1.0 FNa (solution)
`10.0 FNa (solution)
`0.1 F (solution)
`0.5 F (suspension)
`1.0 F (suspension)
`
`-
`
`PLACEBO
`porous, burst
`porous
`porous
`porous
`porous
`TABLET WITH DRUG
`2.0
`porous, irregular
`20.0
`porous, irregular
`200
`porous, burst
`2.0
`10.0
`20.0
`
`porous, irregular
`
`very fragile
`fragile
`durable
`hard
`hard
`
`durable
`durable
`hard and fragile
`
`stable, hard
`
`-
`+/-
`+
`-
`-
`
`+
`+
`+/-
`
`+
`
`(+) the tablet can be easily taken out from the blister;
`(-) the tablet sticks to the blister and can not be removed without damage
`
`Mylan Ex 1038, Page 2
`
`
`
`The physical characteristics of lyophilized tablets containing a model drug...
`
`27
`
`Figure 1. Deformation curves of tablets prepared with phenobar-
`bital sodium, FNa (0.1% and 1.0% solution) and phenobarbital, F
`(0.1% solution); error bars ± SD (n=10).
`
`Figure 2. The effect of localization of the blisters on a freeze-dryer
`shelf (peripheral or central) on deformation curves of tablets con-
`taining phenobarbital sodium (0.1% solution).
`
`first one, a Ph. Eur. disintegration apparatus (Phar-
`ma Test, Hamburg, Germany) was used. Tablets
`were placed in cylinders on the bottom mesh, bur-
`dened with disks and immersed in water at room
`temperature or 37oC. The time required for com-
`plete disintegration of each tablet was recorded. In
`the second method, the tablets were placed in a plas-
`tic syringe funnel on a blotting paper covered with
`glass pellets. The time required for 5 mL of water to
`pass through the syringe was measured.
`
`RESULTS AND DISCUSSION
`
`In the first step of the studies the best ratio of
`gelatin to mannitol was established in order to for-
`mulate freeze-dried matrices of satisfying mechani-
`cal characteristics. In Table 1 the investigated com-
`positions of placebo tablets are shown. On the basis
`of visual inspection and evaluation of adhesion to
`the blister as well as on the integrity after removing
`from a blister formulation composed of mannitol
`and gelatin in ratio 5:1 was chosen as optimum. This
`composition of structure-forming excipients was
`used for preparation of lyophilized tablets with F
`and FNa. All tablets obtained by the lyophilisation
`process were white and porous, their shape was
`cylindrical, corresponding to the shape of the blister
`well. Tablets containing up to 20 mg of F or FNa
`(14.3% of the total mass) had satisfying appearance,
`and they did not differ visually. The increase of the
`amount of FNa to 200 mg per tablet (62% of total
`mass) resulted in increased hardness and, finally,
`tablets could not be removed from blister without
`braking (Table 1).
`
`Figure 3. Changes in weight of lyophilized tablets containing dif-
`ferent concentrations of model drugs during 4 weeks of storage at
`room temperature and 40% and 60% RH (error bars ± SD, n=6): F
`ñ phenobarbital, FNa ñ phenobarbital sodium (a and b ñ two dif-
`ferent batches)
`
`A standard hardness tester used for com-
`pressed tablets was not appropriate to study
`strength of the lyophilized tablets since they did
`not break during the test. This is why the applica-
`bility of another instrument, a universal material
`testing machine Instron, was investigated. The
`results of the mechanical studies performed with
`this apparatus are shown in Figure 1. The curves
`indicate that the matrices are plastic: an increasing
`
`Mylan Ex 1038, Page 3
`
`
`
`28
`
`MA£GORZATA SZNITOWSKA et al.
`
`force caused deformation without breaking the
`tablets.
`At the initial step gradual deformation up to
`70% was observed with small forces applied, but
`30%-50% further increase in the force was required
`to get 80% deformation. The relationship between
`force and deformation is described by the curves
`which are linear in three segments: I ñ up to 20%, II
`ñ between 20% and 70%, III ñ above 70% deforma-
`tion. The initial 20% deformation occurs when the
`fragile lyophilizate structure is crushed, the step II
`corresponds to the deformation which is caused by
`compression of the porous matrix, while only at
`higher loads the structure of a tablet is completely
`destroyed (step III). The force required for 20%
`deformation can be a measure of resistance to crush-
`ing, and the one responsible for step II can be relat-
`ed to porosity and plasticity of the tablets. The
`slopes of the curve segments (aI and aII values)
`were calculated using linear regression analysis. On
`the basis of these parameters it can be assumed that
`the chemical and physical form of an active sub-
`stance influences the mechanical strength and the
`structure of the lyophilized tablets. The use of F in
`acidic form resulted in lower mechanical strength
`(aI=0.38 N/%) in comparison with tablets contain-
`ing FNa (aI=0.98 N/%), however the compression
`response at step II was similar: aII=0.31 and 0.34
`N/%, respectively. This means that faster precipita-
`tion and crystallization of a slightly soluble drug (F)
`resulted in the structure less resistant to crushing,
`however the porosity and plasticity of the matrix
`was similar. For FNa tablets the crushing resistance
`was similar, independent of concentration (aI=0.98
`N/%), but increased drug concentration resulted in a
`less porous and plastic structure, what can be
`demonstrated by higher aII values: 0.62 N/% for
`1.0% FNa formulation and 0.34 N/% for 0.1% FNa
`formulation. This indicates that FNa contributes to
`the structure-forming process. For FNa formulations
`good reproducibility of the curves was observed
`(RSD <15%).
`The results demonstrate that the Instron appa-
`ratus can be suitable for the mechanical analysis of
`lyophilized tablets. A comparison between the
`mechanical properties of tablets located at the
`peripheral and in the central part of the shelf in a
`freeze-dryer was done. The location of the blister
`during lyophilization may be important for the prop-
`erties of tablets because different time of freezing
`was observed ñ shorter for tablets in the peripheries.
`The deformation curves for one batch of tablets con-
`taining 0.1% FNa are shown in Figure 2. Small dif-
`ference in the mechanical behaviour was noted ñ
`
`tablets from the central part of the shelf, freezing
`slower, were more resistant, however the difference
`was not statistically significant. Corveleyn and
`Remon (4) reported that faster freezing results in
`higher degree of supercooling with the formation of
`small crystals. Small ice crystals result in a higher
`surface area and higher degree of porosity after
`lyophilisation, what can explain a decreased
`strength of the peripheral tablets.
`Lyophilized tablets show significant porosity,
`and one can expect that they can absorb moisture
`very easily. To determine hygroscopic properties,
`the tablets were stored in three levels of the relative
`humidity ñ 40%, 60% and 90%. The most intensive
`water uptake occurred at 90% RH and this resulted
`in tablet shrinkage. In Figure 3, the changes in tablet
`weights after 4 week storage at 40% and 60% RH
`are demonstrated. Only a small increase in moisture
`content was noted. This increase was higher at 60%
`RH (2,7%), compared to 40% RH (about 1%). A
`very small moisture enhancement does not allow for
`conclusion whether type of the drug and its concen-
`tration influences hygroscopic properties of tablets.
`Besides, the evaluation of two batches of tablets
`containing 0.1% FNa (a and b, Figure 3) shows that
`the difference within one formulation can be larger
`than between different formulations.
`All lyophilized tablets rapidly disintegrated
`in water regardless of the method of analysis. Due
`to a very short time of the process it could not be
`possible to define the disintegration time precise-
`ly. At 37oC all tablets disintegrated in few seconds
`and at room temperature disintegration time varied
`from 14 s to 2.5 min., but no influence of either
`drug form or concentration was observed. If the
`syringe apparatus was employed, the jelly struc-
`ture was formed due to the presence of gelatin, and
`this stopped the flow of water. This means that a
`new adequate method for analysis of disintegra-
`tion time of the freeze-dried tablets should be
`searched.
`
`CONCLUSIONS
`
`The results demonstrate that the ratio 5:1 of
`mannitol and gelatin as matrix formers is appropri-
`ate to obtain freeze-dried fast dissolving tablets.
`Chemical form, the concentration of the drug and
`physical properties of the freeze-dried liquid (solu-
`tion or suspension), have a significant influence on
`the mechanical strength of tablets. Even though the
`lyophilizates were porous, they did not absorb water
`during at least monthly storage in 60% RH. Further
`studies should be carried out in order to develop a
`
`Mylan Ex 1038, Page 4
`
`
`
`The physical characteristics of lyophilized tablets containing a model drug...
`
`29
`
`new method for determination of disintegration time
`of lyophilized tablets.
`
`REFERENCES
`
`Acknowledgements
`
`The Authors would like to thank Prof. Ass.
`Maria Sadowska and M.sc. Robert Tylingo from
`Department of Food Chemistry and Technology of
`Technical University of GdaÒsk for their contribu-
`tion in the analysis of tablets with the Instron appa-
`ratus.
`
`1. Seager H.: J. Pharm. Pharmacol. 50, 375 (1998).
`2. Dobetti L.: Pharm. Tech. Eur 9, 32 (2000).
`3. Rote Liste: Editio Cantor Verlag, Aulendorf
`2003.
`4. Corveleyn S., Remon J. P .: Int. J. Pharm. 152,
`215 (1997).
`
`Received: 3.08.2004
`
`Mylan Ex 1038, Page 5
`
`