Purpose: Intranasal deposition of aerosols is often studied using in vitro nasal cavity models. However, the relevance of these models to predict in vivo human deposition has not been validated. This study compared in vivo nasal aerosol deposition and in vitro deposition in a human plastinated head model (NC1) and its replica constructed from CT-scan (NC2).
Methods: Two nebulizers (Atomisor Sonique® and Easynose®) were used to administer a 5.6 μm aerosol of (99m)Tc-DTPA to seven healthy volunteers and to the nasal models. Aerosol deposition was quantified by γ-scintigraphy in the nasal, upper nasal cavity and maxillary sinus (MS) regions. The distribution of aerosol deposition was determined along three nasal cavity axes (x, y and z).
Results: There was no significant difference regarding aerosol deposition between the volunteers and NC1. Aerosol deposition was significantly lower in NC2 than in volunteers regarding nasal region (p < 0.05) but was similar for the upper nasal cavity and MS regions. Mean aerosol distribution for NC1 came within the standard deviation (SD) of in vivo distribution, whereas that of NC2 was outside the in vivo SD for x and y axes.
Conclusions: In conclusion, nasal models can be used to predict aerosol deposition produced by nebulizers, but their performance depends on their design.