Semi-empirical stochastic model of aerosol bolus dispersion in the human lung

Aerosol bolus dispersion, that is, the broadening of an inhaled narrow aerosol bolus upon exhalation, was simulated by Monte Carlo methods using a stochastic, asymmetric morphometric model of the human lung. Physical mechanisms considered to contribute to bolus dispersion were (1) axial diffusion in conductive airways, approximated by effective diffusivities, (2) convective mixing at airway bifurcation sites, (3) differences in inspiratory and expiratory velocity profiles, (4) mixing with residual air in alveoli, and (5) inhomogeneous ventilation of the lung lobes due to asymmetric flow spitting at bifurcations and asymmetric and asynchronous filling of the five lung lobes. Theoretical predictions of the bolus dispersion model were compared to experimental data for 79 healthy volunteers, which provide detailed information on statistical bolus parameters (half-width, standard deviation, skewness, and mode shift) and total bolus deposition as a function of the depth of bolus penetration into the airway system. Predicted bolus dispersion and deposition data show excellent agreement with the published experimental data, suggesting that axial diffusion in conductive airways and convective mixing in alveoli, resulting in irreversible particle transport, are the major determinants of bolus dispersion. The variability and asymmetry of the branching airway network, leading to asymmetric flow splitting at airway bifurcations, greatly enhances the effect of irreversibility and the resulting dispersion of the inhaled bolus.