Carbon dioxide diffusion coefficient in noninvasive high-frequency oscillatory ventilation.

Abstract

The carbon dioxide (CO2 ) diffusion coefficient (DCO 2 ) reflects CO 2 removal during high-frequency oscillatory ventilation (HFOV). We hypothesized that despite leak flow during noninvasive HFOV (nHFOV) DCO 2 continues to indicate ventilation efficacy. A neonatal airway model including CO2 production and an adjustable oropharyngeal leak was connected to a ventilator via bi-nasal prongs. Pressures and gas flows were measured at prongs, trachea, and leak. Oscillatory tidal volumes below (V T trachea ) and above the leak (V T prong ) were calculated from tracheal and leak flows. DCO 2 was calculated using V T trachea (DCO 2 trachea ) and V T prong (DCO 2 prong ) and compared with CO 2 partial pressure (pCO 2 ). Effects of leak flow (0, 5, or 10L/min) on DCO 2 were assessed at fixed pressure amplitudes or predefined oscillatory volumes under steady-state pCO 2 conditions in the modeled lung. DCO2 trachea correlated strongly with pCO 2 , independent of the leak flow level (P < 0.0001). DCO 2 prong correlated with pCO 2 without and with moderate leak (P < 0.0001) but not with maximum leak (P = 0.1432). V T trachea correlated with the quotient of tracheal pressure amplitude and frequency irrespective of the leak (P < 0.0001). Based on the pressure amplitude at prong level (A prong ) V T trachea continued to follow a linear model of which the slopes decreased with increasing leak flow. V T prong correlated with the quotient of A prong and frequency, irrespective of the leak (P < 0.0001). DCO2 obtained at the airway opening at prong level reflects ventilation efficacy during nHFOV even in the presence of moderate oropharyngeal leak.