Deposition of Inhalation Aerosols in the Respiratory Tract

Abstract

Chapter one reviews the interconnected relationships between dry powder formulation, dry powder inhaler device, the inhalation flow rate and the lung deposition. However, the relationship between the resistance of a dry powder inhaler and the in vivo airway deposition is not known. To investigate this, chapter 2 looked at the airway deposition of orally inhaled mannitol powder (99mTc-DTPA radiolabeled) in healthy humans by using the low resistance Osmohaler™ at 110 – 130 L/min and 50 – 70 L/min peak inhalation flow rate (PIFR) on separate study days; and high resistance at 50 – 70 L/min on the third study day. The results showed that the resistance of Osmohaler™ was not a significant factor in the lung deposition at 50 -70 L/min PIFR. Instead, the lung deposition depended on the fraction of aerosol loss in the extrathoracic region, which was affected by the inhalation flow rate and the aerodynamic particle size of the aerosol. To describe the relationship between the extrathoracic deposition, the inhalation flow rate and the aerodynamic particle size of the inhaled aerosol, empirical modelling can be used. However, no human studies have been conducted to look at the accuracy of existing empirical models for predicting the in vivo extrathoracic deposition. To evaluate existing empirical models using human data, the inhalation flow rate needs to be known, but currently there is no suitable method for measuring the inhalation flow rate of polydisperse aerosols while preserving the particle size distribution. To solve this problem, the utility of a Thor D-30 ultrasonic flow meter for measuring the flow rate of nebulized aerosol was assessed in chapter 3. Laser diffraction measurement of the nebulized aerosols showed that the Thor-D30 ultrasonic flow meter did not affect the particle size distribution and when a duckbill valve was added after the D-30 flow meter, the exhaled fraction of the aerosol can be isolated and collected onto an exhalation filter. By incorporating the Thor D-30 flow meter to measure the inhalation flow rate of aerosols, chapter 4 investigated the deposition of the orally inhaled aerosols (0.9 % saline) in the extrathoracic region of healthy human subjects during tidal breathing. The resulting extrathoracic deposition fraction for each subject was then compared to the predicted values by four existing empirical models (Golshahi et al, ICRP , Cheng et al. and Stahlofen et al). The results showed that the Golshahi et al. correlation, which accounted for the average upper airway diameter, was the most accurate in predicting the mean in vivo extrathoracic deposition fraction. However, the Golshahi et al model was not satisfactory for providing subject specific predictions in extrathoracic deposition. This was likely attributed to geometric dissimilarity between the rigid physical models used to develop the Golshahi et al. correlation and the dynamic nature of a real human upper airway during inhalation. The existing empirical models and in vivo lung deposition studies also have not accounted for deposition caused by electrostatic charge of the aerosols. To date, it is still unclear on the extent that electrostatic charge contributes to the in vivo aerosol deposition in the airways because of the difficulties in generating and measuring the charged aerosols during an in vivo lung deposition study. Thus, a suitable setup was developed for in vivo lung deposition study by using a vibrating orifice induction aerosol generator to produce monodisperse particles at selected elementary charges per particle. Sep-Pak QMA (anionic exchange) and Sep-Pak amino propyl cartridge were used to purify the [18F]Fluorodeoxyglucose from its preparation in phosphate buffer for radiolabelling the charged aerosol particles. Both the purification of [18F]Fluorodeoxyglucose and generation of the aerosols could be done remotely by Elveflow OB1 controller to minimize radiation exposure. Remote sampling of the aerosol charge was done by a custom made faraday cup attached to a linear actuator that extends directly beneath the atomizer. The resulting vibrating orifice induction aerosol demonstrated repeatable aerosol generation, showing consistency in particle size distribution and aerosol number concentration.