Inert gas mixing in the upper and central airways of man.

The dispersion of an inert bolus of helium (He) or sulfur hexafluoride (SF6) was used as a direct, non-invasive measure of longitudinal gas mixing in the conducting airways of three human subjects. Penetration volume was determined as the milliliters of air inspired after bolus injection, and by computing the increase in mixing between bolus penetrations of 30 and 90 ml, mixing was characterized in the intervening upper airway compartment. Mixing in a central airway compartment was similarly evaluated as the increased mixing between bolus penetrations of 90 and 150 ml. To investigate mixing mechanisms during inspiration, inspiratory flow rate was varied between 0.2 and 1.8 L . sec-1 while the expiratory flow rate was held constant at 0.4 L . sec-1 and conversely, expiratory mixing mechanisms were studied at a fixed inspiratory flow of 0.4 L . sec-1 by varying expiratory flow from 0.2 to 2.4 L . sec-1. Under all eight experimental conditions (i.e. two inert gases X two airway compartments X two variable flow half-cycles), the extent of SF6 mixing was found to be three times that of He, and this is indicative of a dispersion process in which radial diffusion limits longitudinal mixing. This was further supported by the positive correlation between mixing and expiratory flow in the upper airway compartment. Mixing in this compartment did not vary with inspiratory flow, possibly because of the influence of turbulence generated by the glottis. In the central airways, the effect of changes in both inspiratory and expiratory flow was insignificant.