Abstract��
OBJECTIVE To optimize the formulations of nimodipine matrix tablet using D-optimal mixture design. METHODS Independent variables were the amounts of HPMC, sodium alginate and lactose, and drug accumulated release at 3, 6, 9 and 12 h were dependent variables. Optimal mathematical models were chosen to estimate the relationship between the dependent and the independent variables,and to delineate triangular-dimensional contour plots. With appointed drug release at 3,6,9 and 12 h as target values, overlay plot was used to predict the optimal formulation. Validation of optimal formulation was verified at last. RESULTS In optimal regression models, there are reliable quantitative relationships between three factors and four evaluation indexes. In vitro release test of five optimal checkpoint formulations indicated that there existed high approximation between their predicted and experimental values.The values of similarity factor (f2) indicated the equivalence to the release profile of the optimum formulation and zero-order release profile. CONCLUSION D-optimal mixture experimental design facilitated the optimization of nimodipine matrix formulations. Optimal models are proved to have the ability of high prognostic and feasible.
2009, Vol. 44 