Preparation and characterization of cellulose acetate butyrate pseudolatex for multi-particulate controlled drug delivery



Journal Title

Journal ISSN

Volume Title


Texas Tech University


pH dependent release has been reported from the formulations coated with dispersions of cellulose polymers. Unlike ionizible methacrylate polymers, such as Eudragit® polymers (L-l00 and S-100), cellulose polymers are pH insensitive. However, incorporation of ionic stabilizers in the process of pseudolatex preparation could render the coating membrane pH sensitive. Therefore, the objectives of the present dissertation were: (1) to develop an aqueous pseudolatex dispersion of cellulose acetate butyrate (CAB) with a non-ionic stabilizer, (2) to evaluate and compare the films made from the pseudolatex and organic solution of CAB, and (3) to test the ability of aqueous pseudolatex films of CAB for a pH dependent release from controlled release multi-particulate beads.

Pseudolatex was prepared with cellulose acetate butyrate and polyvinyl alcohol (PVA) as stabilizer by polymer emulsification after screening different stabilizers. The mean particle diameter was found to be 300nm. The pseudolatex was stabilized by steric forces and the stabilization layer was found to be 23 nm around each particle. The pseudolatex displayed viscoelastic behavior with higher solids content. The sedimentation volume was found to be 0.03%) over a period of six months. An increase in plasticizer concentration increased the free volume fraction and degree of elongation of the casted film. Also, an increase in plasticizer concentration decreased the Tg and Young's modulus. Films made from CAB pseudolatex were found to be less permeable to tritiated water than the ones prepared from aqueous dispersion. All the essential properties of coating membrane, such as permeability, mechanical properties, and Tg, were fairly controllable as a function of plasticizer concentration.

Excipient compatibility was studied by thermal analysis (differential scanning calorimetry) and x-ray powder diffraction. Blends of equal proportions of verapamil HCl and excipient with 5% water were stored at 55 °C for three weeks and then subjected to analysis. More than 95% of the verapamil HCl was retained in drug excipient blends. X-ray powder diffraction pattern revealed absence of any crystallinity change in verapamil HCl. However, thermal analysis indicated an interaction between verapamil HCl and excipient.

Controlled release multi-particulate beads were prepared by loading the verapamil HCl on inert beads and subsequently coating it with CAB pseudolatex. The process and formulation factors were screened by Plackett-Burman screening design in order to identify the most important factors affecting the amount of verapamil HCl released in 12 hours. Factors and response were linearly modeled. The mathematical model fitted the data and explained 98.05% of variability in the response.

The difference between observed and predicated values of any given run did not exceed 6% of maximum cumulative release at 12 hours. Plackett-Burman screening design identified coating weight gain, duration of curing and amount of plasticizer as the most important factors determining cumulative percent released in 12 hours. Increase in coating weight gain and duration of curing decreased the rate and extent of verapamil HCl released. However, increase in plasticizer concentration increased the rate and extent of verapamil HCl released. Amount of Polydexfrose/HPMC (Opadry II®), spray rate, fluid bed coater outlet temperature, and atomizing pressure had no statistically significant (p < 0.05) influence on the response.

In order to optimize the dissolution profile for zero order release, coating weight gain, duration of curing, and plasticizer concentration were modeled with in-vitio release profile of verapamil HCl. Experiments were designed and data was collected according to three-factor, three-level face-centered central composite design. The factors and responses were modeled and optimized by Response Surface Methodology (RSM) and Artificial Neural Network (ANN). The model fitted the data and explained 90%) of variability in response in the case of RSM and at least 70% in the case of ANN. Release profile was optimized for a zero-order.

Optimized formulations were prepared according to the factor combinations dictated by RSM and ANN. In both cases the observed release pattern of the optimized formulations was close to the predicted release pattern. However, the modeling and optimization abilities of RSM, as evaluated by the R-squared values, were found to be higher than that of ANN. X-ray powder diffraction and content analysis of optimized formulation suggested the integrity of verapamil HCl and excipients incorporated.

Optimized formulations were subjected for dissolution at pH 1.2 and 7.4. The release profile of verapamil HCl was pH independent, which was attributed to the nonionic nature of the polyvinyl alcohol. The optimized formulations were stored at 25 C, 30 °C, 40 °C, 50 °C, 60 °C and at different conditions specified by the International Committee on Hormonization. The rate of release of verapamil HCl from the beads stored at room temperature and at 30 ^C was found to be stable. However, a decrease in rate and extent of release was observed from the beads stored at 50 °C and 60 ''C



Delayed-action preparations, Verapamil -- Administration & dosage, Cellulose -- Analogs & derivatives, Drug delivery systems