Coating characterization by hyperspectroscopy and predictive dissolution models of tablets coated with blends of cellulose acetate and cellulose acetate phthalate.

Abstract

The objective of current research was to develop the models of dissolution prediction of tablets coated with cellulose acetate (CA 320S or CA 398-10) and cellulose acetate phthalate (C-A-P) blends. Independent variables selected were coating percent (X1) and percent of CA 320S or CA 398-10 (X2) in the blend. Dependent variables selected were dissolution in 1 (Y1), 8 (Y2), and 24 h (Y3). Diclofenac sodium core tablets were coated with blend of either CA 320S and C-A-P or CA 398-10 and C-A-P at approximately 5, 7.5, and 10% weight gain. CA 320S and CA 398-10 content in the corresponding blends varied from 33.3-66.7% and 25.0-50.0% relative to C-A-P, respectively. Dissolution was performed in phosphate buffer 6.8 using USP apparatus 2. Coated tablets were also characterized for surface morphology and coating uniformity by near infrared hyperspectroscopy. Y1, Y2, and Y3 were statistically (p < 0.05) affected by X2 in CA 320S/C-A-P and CA 398-10/C-A-P blends coated tablets. On the other hand, X1 had statistically significant (p < 0.05) effect only on the Y3 in CA 320S/C-A-P while Y1 was statistically (p < 0.05) affected by X2 in CA 398-10/C-A-P. Analysis of variance also indicated statistically significant (p < 0.05) effect of the studied variables on the dependent variables for both the blends. The models were verified by independent experiment. Model predicted and empirical values of Y1, Y2, and Y3 were close with maximum residual of 7.0%. In conclusion, dissolution can be modulated by varying composition of blend, polymer type, and coating weight.