Abstract
Substituting nebulisers by another in non-invasive ventilation circuit (NIV) involves many process variables which must be adjusted to ensure patient optimum therapy. However, there is a doubt when nebulisers use the same technology.
Data mining technology based on artificial neural networks and genetic algorithms were used here to model in-vitro inhalation process and predict bioavailability from inhaled doses delivered by three different vibrating mesh nebulisers (VMNs) in NIV.
Modelling of data indicated that in-vitro performance of VMNs was dependent mainly on fine particle fraction, mass median aerodynamic diameter (MMAD), total emitted dose (TED) and to lesser extent on nebuliser type. Ex-vivomodel indicated that amount of salbutamol collected on facemask filter was directly affected by TED. In-vivo model showed that amount of salbutamol deposited into the lung (0.5 hQ) and amount absorbed systemically (24 hQ) were dependent directly on MMAD and TED. Female patients showed higher 24 hQ values than males. Nebuliser type affected TED, 0.5 hQ but not 24 hQ values.
Results indicate suitability of VMNs in achieving appropriate in-vitro inhalation performance model. The results also, indicate that the three VMNs are comparable and can be interchanged with no fear of any additional toxicity. (C) 2016 Elsevier B.V. All rights reserved.