Abstract Background:

In vitro—in vivo correlations (IVIVC) for lung deposition may be established by testing inhalers in vitro with realistic mouth–throat (MT) models and inhalation profiles (IP). This study was designed to compare the currently available MT models and their ability to predict in vivo lung deposition.

Methods:

Budelin^® Novolizer^®, Ventolin^® Evohaler^®, and Respimat^® fenoterol were chosen to represent a dry powder inhaler (DPI), metered dose inhaler (MDI), and soft mist inhaler (SMI) in tests using eight MT models: small, medium, and large Virginia Commonwealth University (VCU) models; small, medium, and large oropharyngeal consortium (OPC) models, the medium adult Alberta Idealized Throat (AIT), and the United States Pharmacopeia (USP) Induction Port, with IPs that simulated those used by volunteers in lung scintigraphy studies. Drug deposition in MT was compared across the models, and IVIVCs evaluated by comparing values for total lung dose in vitro (TLD_{in vitro}) to those reported in the clinic.

Results:

MT deposition was dependent on both the flow condition and MT geometry for all the inhalers, while the deposition rank order was independent of both factors. The overall ranking was USP <OPC_L <AIT <VCU_L <VCU_M <OPC_M <VCU_S <OPC_S. All model groups (VCU, OPC, AIT, and USP) produced TLD_{in vitro} comparable with TLD_{in vivo} for the DPI, where flow conditions dominated aerosol deposition. Only the VCU and OPC models produced good IVIVCs for the MDI, where MT geometry dominated deposition. In vitro tests with the SMI at 15–45 L/min underestimated MT deposition and overestimated lung deposition with all MT models except OPCs, although testing at higher flow rates showed good agreement with in vivo results.

Conclusions:

While realistic in vitro tests may produce results that correspond to drug deposition in vivo, MT model selection was most important for the MDI and SMI, but much less important than inhalation strength for the DPI.