Physiological and pathological considerations for aerosol deposition: Expiration and models of deposition

Theoretical models are often used to predict fractional and regional deposi tion of inhaled particles in the respiratory tract. The distribution of par ticle diameters in the aerosol, airway geometry, breathing pattern, and loc al flow profiles are major determinants of deposition in the lung. However, most models predicting deposition consider airway geometry to be fixed and concentrate on inspiratory events in their calculations. When particle los ses during expiration are estimated, inspiratory and expiratory flow patter ns and airspace geometry are usually considered to be similar with similar effects on deposition. The theme of this presentation will be the analysis of events during expiration that influence particle deposition. In the norm al lung, during quiet breathing, experiments performed on excised lungs hav e suggested that convective forces may be different between inspiration and expiration that significantly affect deposition. Bennett and Smaldone, in excised dog lungs, by regulating the duty cycle of tidal breathing found th at more particles deposited during inspiration than expiration and that the effects were density dependent. In human subjects with obstructive lung di sease, the situation is reversed. Major differences in large airway geometr y between inspiration and expiration can occur with each tidal breath. Once the FEV1 decreases to about 60% of the FVC, flow-limiting segments (FLS) a re known to form in central airways. Large pressure drops can occur over sh ort lengths of airway indicating disturbed regions of convective streamline s that are not present during inspiration. Using radiolabeled monodisperse particles, Smaldone and Messina have determined that FLS can be a major det erminant of deposition in central airways. Theoretical predictive models of particle deposition and clearance should consider inspiratory and expirato ry differences in airway physiology in health and disease.