CO2 elimination as predicted by augmented dispersion and convection in an asymmetrical dog airway model

We have developed a computer model of the dog lung based upon an asymmetrically branching network of tubes to describe gas exchange during high-frequency oscillations. Impedances to oscillatory flows were calculated for each airway segment and used to determine flow distributions in all airway generations. Gas exchange was assumed to occur by convective and augmented dispersive mechanisms. Also included in the model were features commonly found in animal studies including the effects of bias flow of fresh gas at the airway opening and equipment dead-space volume. The magnitude of CO2 elimination predicted by the model closely resembled actual experimental data. Specifically, it predicted that CO2 elimination is proportional to frequency to the 0.82 power and tidal volume to the 1.25 power. It also indicated that the bias flow rate and equipment dead-space volume influence gas exchange characteristics and thus these variables should be considered when comparing data from different studies.