EFFECT OF THE PHYSICAL-PROPERTIES OF RESPIRATORY GAS ON PNEUMOTACHOGRAPHIC MEASUREMENT OF VENTILATION IN NEONATES

The measurement of ventilation in neonates has a number of specific characteristics; in contrast to lung function testing in adults, the inspiratory gas for neonates is often conditioned. In pneumotachographs (PNT) based on Hagen-Poiseuille's law, changes in physical characteristics of respiratory gas (temperature, humidity, pressure and oxygen fraction [FiO2]) produce a volume change as calculated with the ideal gas equation p * V/T = const; in addition, the viscosity of the gas is also changed, thus leading to measuring errors. In clinical practice, the effect of viscosity on volume measurement is often ignored. The accuracy of these empirical laws was investigated in a size 0 Fleisch-PNT using a flow-through technique and variously processed respiratory gas. Spontaneous breathing was simulated with the aid of a calibration syringe (20 ml) and a rate of 30 min-1. Results: The largest change in viscosity (11.6% at 22-degrees-C and dry gas) is found with an increase in FiO2 (21...100%). A rise in temperature from 24 to 35-degrees-C (dry air) produced an increase in viscosity of 5.2%. An increase of humidity (0...90%, 35-degrees-C) decreased the viscosity by 3%. A partial compensation of these viscosity errors is thus possible. Pressure change (0...50 mbar, under ambient conditions) caused no measurable viscosity error. With the exception of temperature, the measurements have shown good agreement between the measured volume measuring errors and those calculated from viscosity changes. Conclusions: If the respiratory gas differs from ambient air (e.g. elevated FiO2) or if the PNT is calibrated under BTPS conditions, changes in viscosity must not be neglected when performing accurate ventilation measurements. On the basis of the well-known physical laws of Dalton, Thiesen and Sutherland, a numerical correction of adequate accuracy is possible.