Lung volume is a determinant of aerosol bolus dispersion

The technique of inhaling a small volume element labeled with particles ("aerosol bolus") can be used to assess convective gas mixing in the lung. While a bolus undergoes mixing in the lung, particles are dispersed in an increasing volume of the respired air. However, determining factors of bolus dispersion are not yet completely understood. The present study tested the hypothesis that bolus dispersion is related, among others, to the total volume in which the bolus is allowed to mix--i.e., to the individual lung size. Bolus dispersion was measured in 32 anesthetized, mechanically ventilated dogs with total lung capacities (TLCs) of 1.1-2.5 L. Six-milliliter aerosol boluses were introduced at various preselected time-points during inspiration to probe different volumetric lung depths. Dispersion (SD) was determined by moment analysis of particle concentrations in the expired air. We found linear correlations between SD at a given lung depth and the individual end-inspiratory lung volume (V(L)). The relationship was tightest for boluses inhaled deepest into the lungs: SD(40) = 0.068 V(L) - 1.77, r(2) = 0.59. Normalizing SD to V(L) abolished this dependency and resulted in a considerable reduction of inter-individual variability as compared to the uncorrected measurements. These data indicate that lung size influences measurements of bolus dispersion. It therefore appears reasonable to apply a normalization procedure before interpreting the data. Apart from a reduction in measurement variability, this should help to separate the effects on bolus dispersion of altered lung volumes and altered mixing processes in diseased lungs.