Sequential statistical optimization of a positively-charged submicron emulsion of miconazole.

A positively charged oil/water (O/W) emulsion containing an antifungal agent was developed for ophthalmic use. An attempt was made using a sequential statistical methodology to optimize the O/W emulsion by varying both formulation and process parameters to obtain the smallest droplet size emulsion that can remain stable for a long period of time. During the first step of the study, not less than 7 parameters were found to be important--drug content, amount of lipophilic phase, poloxamer concentration, quantity of the phospholipids-stearylamine couple, pH adjustment, time of coarse emulsification and time of high pressure homogenization. A screening approach based on Hadamard's matrix was used to select the parameters displaying the most significant effects on response parameters. A first set of 8 experiments proved efficient enough to define the concentration of poloxamer and the quantity of the couple phospholipids-stearylamine, which confer the overall positive charge to the emulsified droplet, as the most significant parameters affecting the final droplet size of the emulsions formed. A 2k-type experimental design was then built with the two main factors in order to evaluate a first-order polynomial model with interaction. Poor analysis of variance results after an additional center experiment was performed revealed the lack of fit of the linear model as well as the importance of the response surface curvature due to a close optimum location. To find the optimal operating conditions the design was sequentially completed with 4 more experiments according to the Box and Wilson method. The response surfaces in 3-dimensional representation and their corresponding contour plots proved helpful in analyzing the validated models and in highlighting the precise optimum location. The optimized positive submicron emulsion is now under in vivo investigation.