- Abstract The main purpose of this study was to investigate the nasal absorption of insulin from a new microemulsion spray preparation in rabbits. The bioavailability of insulin lispro via the nasal route using a W/O microemulsion was found to reach 21.5% relative to subcutaneous administration, whereas the use of an inverse microemulsion as well as a plain solution yielded less than 1% bioavailability. The profile of plasma glucose levels obtained after nasal spray application of the microemulsion (1 IU/kg lispro) was similar to the subcutaneous profile of 0.5 IU/kg at the first 90 min after application and resulted in a 30–40% drop in glucose levels. The microemulsion system was characterized by DLS, TEM, viscosity measurements, and by construction of pseudo-ternary phase diagram. The average droplet size of an insulin-unloaded and insulin-loaded microemulsions containing 20% aqueous phase (surfactants-to-oil ratio = 87:13) was 2 nm and 2.26 nm in diameter, respectively. In addition, the effect of the microemulsion on FITC-labeled insulin permeation was examined across the porcine nasal mucosa in vitro . The permeability coefficient of FITC-insulin via the microemulsion was 0.210 ± 0.048 cm/h with a lag time of 10.9 ± 6.5 min, whereas the permeability coefficient from a plain solution was 0.082 ± 0.043 cm/h with a lag time of 36.3 ± 10.1 min. In view of the absorption differences of insulin between 20%, 50% water-containing microemulsions and an aqueous solution obtained in vitro and in vivo , it has been concluded that the acceleration in the intramucosal transport process is the result of encapsulating insulin within the nano-droplet clusters of a W/O microemulsion, while the microemulsion ingredients seems to have no direct role.