A study of microemulsions as prolonged-release injectables through in-situ phase transition.

Microemulsions (MEs) have been studied extensively as colloidal carriers for the delivery of both water-soluble and lipid-soluble drugs. Our previous study showed that addition of water to ME formulations resulted in phase transition to either liquid crystal (LC) or coarse emulsion (CE). The aim of this study was to investigate whether these MEs could be used as drug delivery vehicles for prolonged release through in-situ phase transition following extravascular injection. Three ME formulations from the same pseudo-ternary phase diagram were investigated with respect to their phase transition behavior, and in-vivo drug release; a coarse emulsion-forming ME (CE-ME), an oil rich LC-forming ME (LC-ME1), and an oil poor LC-forming ME (LC-ME2). CE-ME was a W/O ME and both LC-MEs were O/W type. The release profiles of (99m)Tc labeled MEs following subcutaneous (SC) injection in rabbits were investigated with gamma-scintigraphy. The CE-ME dispersed readily in water, forming a CE, whereas the LC-forming MEs formed 'depots' in water. Polarized microscopy revealed a LC boundary spontaneously formed at the water/ME interface for the LC-MEs with the LC-ME2 forming a substantially thicker LC layer. The CE resulting from the water-induced transition of the CE-forming ME had a higher viscosity than the MEs, but lower than the LCs resulted from LC-MEs. Compared to LC-ME1, LC-ME2 underwent more rapid phase transition and the resultant LC had significant higher viscosity. The LCs formed from both ME formulations exhibited pseudoplastic properties; increasing the shear rate decreased the apparent viscosity exponentially. Following SC injection into the animal thigh, the LC-MEs had more prolonged release of (99m)Tc in a first-order manner, than CE-ME. The oil poor LC-ME2 had the slowest release with a t1/2 of 77min, 2.3 times longer than the oil rich LC-ME1; consistent with the thickness of LC layer formation observed in-vitro and their relative viscosities. In conclusion, the present in-vivo study has demonstrated the application of MEs as extravascular injectable drug delivery systems for sustained release. The retention of the vehicles at the injection site and the release rate were determined predominantly by their phase transition rather than ME type or oil content.