Do drug release studies from SEDDS make any sense?

Self-emulsifying drug delivery systems (SEDDS) are considered as a potential platform for mucosal drug delivery. The in vitro-in vivo correlation, however, is in particular for this type of delivery systems considerably poor resulting quite often in a simple trial and error approach in order to optimize formulations. One reason for this situation is certainly the lack of appropriate methods to determine the drug release from SEDDS in vitro, as the process is particularly troublesome. For quantification of the drug in the release medium the oily droplets need to be separated. In most studies this is achieved by utilizing a separating membrane such as dialysis membranes or filters having a huge impact on the obtained release profile. Moreover, sink conditions are very often not provided. As drug release from SEDDS is based on a simple diffusion process from a lipophilic liquid phase into an aqueous liquid phase, a likely more meaningful way to characterize the release behaviour might be just the determination of the distribution coefficient (log DSEDDS/RM) of the drug between the SEDDS pre-concentrate and the release medium (RM). As log D is simply the measure of the difference in solubility of a compound in two phases, it can be determined by measuring solubility of drug or drug complex in the SEDDS pre-concentrate and in the release medium in a separate manner. The impact of log DSEDDS/RM on the in vivo drug release behaviour is discussed including various case studies.