J. Pharm. Pharmacol. 1986, 38: 686688
`Communicated March 18. 1986
`
`0 1986 J. Pharm. Pharmacol.
`
`COMMUNICATIONS
`
`Polysorbate 20 as a drug release regulator in ethyl cellulose film
`coatings
`
`TUULA LINDHOLM:, BENGT-AKE LINDHOLM, MERVI NISKANEN, JAAKKO KOSKINIEMI, Department of Pharmaceutical
`Technology, University of Kuopio, P. 0. Box 6, SF-70211 Kuopio, Finland
`
`Tablet coatings of h drophobic ethyl cellulose have been
`made more hydro {ilic by the addition of a non-ionic
`surfactant, polysorgate (Tween) 20, to the film. As its
`content increased, so did the release of sodium salicylate
`from the coated tablets. With a certain content of surfactant
`and specific thickness of the tablet coat, zero order release
`kinetics were observed. Leaching of the polysorbate 20
`occurred from all formulations. Scanning and transmission
`electron micrographs showed that the structure of the coats
`consisted of several layers parallel to the tablet surface.
`Polysorbate 20 was seen as small drops in some coats.
`
`In the coating process, surfactants have been shown to
`facilitate spreading of the coating mixture on tablets
`(Banker 1966; Pickard & Rees 1974). Small amounts of
`non-ionic surfactant have been used to wet and homo-
`genize the coating mixtures (Lehman & Dreher 1972;
`Pickard & Rees 1972; Takamura et a1 1973; Parker et a1
`1974; Patt 1975; Tomassini et a1 1975; Lehman 1975,
`1982). Surfactants can also change the permeability of
`the coats (List & Kassis 1982). We have previously
`shown that the release rate of salicylic acid from coated
`tablets depended on the surfactant added (polysorbate
`(Tween) 20,81 or Span 20) to the ethyl cellulose coating
`(Lindholm & Juslin 1982).
`In the present study we have used tablets with ethyl
`cellulose film coats that contained 1040% w/w poly-
`sorbate 20. The aims of this study were firstly to
`determine the effect of surfactant content and coat
`thickness on the release of sodium salicylate, secondly
`to obtain information about the structure of the coats,
`and thirdly to investigate the relationship between coat
`composition and release kinetics.
`
`Materials and rnethoh
`The tablet composition was: sodium salicylate (Ph.
`Eur., particle size 300 pm) 33.1% w/w, microcystalline
`cellulose (Avicel PH 101, FMC Corp.) 66.4% w/w, and
`magnesium stearate (Ph. Eur.) 0.5% w/w as lubricant.
`Tableting, coating and measurement of the release of
`
`* Correspondence.
`
`sodium salicylate from tablets were as described previ-
`ously (Lindholm & Juslin 1982; Lindholm et a1 1985).
`The release data were fitted to the cube root, first
`order, zero order and Higuchi's diffusion equation as in
`Lindholm et a1 (1985).
`The amount of polysorbate 20 dissolved from the free
`coats (detached carefully from the cores) was deter-
`mined gravimetrically. The pieces of the coats were
`weighed before and after they had been soaked in water
`at 37 "C and dried. In addition, the undetached coats
`containing 50% polysorbate 20 were studied. The
`tablets were weighed, kept in the rotating basket
`apparatus for different time intervals, dried and then
`weighed again. The amount of dissolved sodium salicy-
`late (determined spectrophotometrically) was subtrac-
`ted from the difference between the two weights.
`The structure of the coats was studied using scanning
`and transmission electron microscopy. In the former the
`samples were either cut with a microtome or fractured
`under liquid nitrogen before being coated with gold. In
`the latter the samples were embedded in Ladd's epon
`(Griffin 1972) and then cut with an ultramicrotone.
`
`Results and discussion
`Much of polysorbate 20 leached away (Table 1).
`Leaching was fast in the beginning and increased slightly
`with time. In the micrographs that were taken from the
`coats containing 50% polysorbate 20 and 50% ethyl
`cellulose, the surfactant can be seen as small drops in
`ethyl cellulose (Fig. 1A). These micrographs also show
`the leaching of surfactant; no dark points are left in the
`5 h samples, where only small, empty regularly ordered
`holes are left (Fig. 1B). The coats containing 10 or 30%
`surfactant were homogenous (Fig. 1C).
`However, some of the polysorbate 20 could have
`dissolved during work, when the thin sections were
`collected on the surface of the water bath after having
`been cut with a diamond knife. In that instance the
`number of points or pores may not be considered an
`exact measure of the amount of the surfactant in any
`
`Page 1
`
`KASHIV EXHIBIT 1051
`IPR2019-00791
`
`

`

`687
`COMMUNICATIONS
`Table 1. Amount of polysorbate 20 dissolved in water at
`37 "C from coats containing ethyl cellulose and polysorbate
`20. Except for one case, the coats were detached from
`cores. Results were determined gravimetrically. E = ethyl
`cellulose, Tw = polysorbate 20.
`
`Composition
`of thecoat
`E90%,Tw 10%'
`E 70%, TW 30%'
`E SO%, TW 50%'
`E 50%. TW 50%b
`
`Thick-
`ness of
`the coat
`(run)
`74 f 1
`74 ? 2
`148 L 3
`148 ? 3
`
`Tw dissolved
`(Y~oftheamountofTw
`in the coat)
`3h
`55.6 5 3.0
`74.1 ? 0.5
`84.3 L 0.2
`42.8 k 2.1
`
`6h
`66.4 ? 5.7
`78.4 L 1.4
`87.1 f 1.8
`49.3 f 3.3
`
`l h
`47.4 ? 3.1
`65.3 2 1.0
`82.9 f 0.8
`27.5 f 2.2
`
`a Coats detached from the cores.
`Coats not detached from the cores.
`
`FIG. 1. Transmission electron micrographs from cross
`section of the ethyl cellulose coats containing polysorbate
`20. The coat with 50% polysorbate 20 from the tablet (A)
`before and (B) after bein 5 h in the dissolution test, (C) the
`coat containing 30% pofysorbate 20 after the tablet had
`been 3 h in the dissolution test (4500~).
`sample. Minor dissolution of it into Ladd's epon can not
`be excluded, but no evidence of this could be detected
`from the micrographs.
`In this work, the coating formulation was sprayed
`onto the tablet mass, and the tablets were allowed to dry
`several times during the coating process to prevent them
`from sticking to each other (Lindholm & Juslin 1982).
`The effect of such a coating process can be seen from the
`scanning electron micrographs, where there are many
`layers parallel to the tablet surface (Fig. 2).
`Sodium salicylate can move in the water-filled holes
`
`FIG. 2. A scanning electron micrograph from a cross section
`of an ethyl cellulose coat containing 50% polysorbate 20.
`The section was broken after a piece of the coat had been
`immersed in liquid nitrogen. The section was from a tablet
`that had been in the dissolution test for 1 h (500~).
`
`and cavities between the layers of the coat (Fig. 2), and
`because of the amount of sodium salicylate released
`from the tablets (Fig. 3), it can also move between the
`layers. On the scanning micrographs, irregularly for-
`med, connecting pores or cavities can be seen (Fig. 2).
`Release curves for coated tablets with cores of
`different breaking strengths (9,11,13 kg) did not differ
`significantly from each other. When the amount of
`polysorbate 20 in the coat increased, release of sodium
`salicylate from the coated tablets was accelerated and
`the lag times were shortened (Fig. 3). Similar results
`have been obtained with another hydrophilic additive,
`hydroxypropyl methylcellulose, in ethyl cellulose (Kan-
`nikoski 1984; Kannikoski et a1 1984). In most cases the
`release kinetics of sodium salicylate was of the first
`order or according to the cube root equation. With quite
`thick coats, when the amount of polysorbate 20 was
`50% w/w of the dry coat, linearity of the release profile
`
`Page 2
`
`

`

`688
`
`1007
`
`COMMUNICATIONS
`the coat during the dissolution test, which may be a
`reason for zero order kinetics. In addition, it was
`necessary to use the appropriate content of surfactant
`and suitable coat thickness.
`
`The authors wish to thank Alpo Pelttari, M. S., for his
`help in electron microscopy.
`
`REFERENCES
`Banker, G. S. (1966) J. Pharm. Sci. 55: 81-89
`Donbrow, M., Benita, S. (1982) Ibid. 34: 547-551
`Friedman, M., Donbrow, M., Samuelov, Y. (1979) Drug
`Devel. Ind. Pharm. 5: 407-424
`Goldman, R. (1970) Drug Cosmet. Ind. 107: 52-64,
`151-156
`Griffin, R. L. (1972) Ultramicrotomy, 9-14, The William &
`Wilkins Company, Baltimore, London
`Kannikoski, A. (1984) Acta Pharm. Fenn. 93: 135-145
`Kannikoski, A,, Fock, H., Marttila, E., Uotila, J. (1984)
`Ibid. 93: 147-157
`Lee, V. H.-L., Robinson, J, R. (1978) in: Robinson, J. R.
`(ed.) Drugs and the Pharmaceutical Sciences: Sustained
`and Controlled Release Drug Delivery Systems, vol. 6,
`Mercel Dekker Inc., New York, pp 123-209
`Lehman, K. (1975) Acta Pharm. Technol. 21: 255-260
`Lehman, K. (1982) Acta Pharm. Fenn. 91: 225-238
`Lehman, K., Dreher, D. (1972) Pharm. Ind. 34: 894-899
`Lindholm, T., Juslin, M. (1982) Ibid. 44: 937-941
`Lindholm, T., Juslin, M., Kekalainen, S. (1985) Ibid. 47:
`109?-1098
`Lippold, B. H., Forster, H. (1981) Ibid. 27: 169-179
`List, P. H., Kassis, G. (1982) Ibid. 28: 21-33
`Parker, J. W., Peck, G. E., Banker, G. S. (1974) J. Pharm.
`Sci. 63: 119-125
`Patt, L. (1975) Acta Pharm. Technol. 21: 277-285
`Pickard, J. F., Rees, J. E. (1972) Pharm. Ind. 34: 833-839
`Pickard, J. F., Rees, J. E. (1974) Manufact. Chem. Aerosol
`News 45: 19-22
`Shah, N. B., Sheth, B. B. (1972) J. Pharm. Sci. 61: 412-416
`Takamura, K., Takamura, N., Kondo, T. (1973) Ibid. 62:
`610-612
`Tomassini, L., Goschkarjan, M., Radeva, K., Andonova,
`V. (1975) Pharmazie 30: 779-782
`
`Time (hl
`FIG. 3. Release profiles of sodium salicylate from uncoated
`cellulose and 30'0 I polysorbate 20 (thickness 74 k 2 pm), 0
`(0) and coated tablets as a mean of six samples & s.e.m.
`Composition of the coats: A 90% ethyl cellulose and 10%
`thickness 101 f 1 pm), V 70% ethyl
`polysorbate 20
`(thickness 148 f 3 pm and 80 rt 1 pm), * 40% eth 1
`and 0 '50% ethyl cellulose and 50% polysorbate 20
`cellulose and 60% polysorbate 20 (thickness 155 f 2 pmy.
`200 mg, 8 kg, 0 , O and * $00 m , 8 kg, V 400 m , 9 kg, 12
`Masses and breaking stren ths of the cores were: 0 and A
`kg and 13 kg. Statistically signikcant (P < 0.05y linearity
`was obtained for the release
`rofile when the content of
`polysorbate 20 was 50% and tgickness was 148 f 3 pm.
`
`was observed. With some thick coats, however, the
`release became
`too slow, and linearity was not
`achieved. When the surfactant content was 6O%, the
`coat was so hydrophilic that release was rapid (Fig. 3).
`First order kinetics, the theoretical equation dealing
`with the release mechanism of film-coated tablets and
`free films, do not apply when the film changes with time
`(Goldman 1970; Shah & Sheth 1972; Lee & Robinson
`1978; Friedman et al 1979; Lippold & Forster 1981;
`Donbrow & Benita 1982) and, in some cases, zero order
`release kinetics can be obtained with appropriate
`changes in the coat. Leaching of polysorbate 20 changed
`
`Page 3
`
`