Estimation of triggering work of breathing. The dependence on lung mechanics and bias flow during pressure support ventilation.

Abstract

Work of breathing necessary to trigger a ventilator (WOBtr) was calculated during pressure support ventilation (PSV), and the effect of bias flow on WOBtr was evaluated. A spring-loaded bellows type lung model with two bellows placed in series was used to simulate spontaneous breathing. A Venturi mechanism of jet flow generated subatmospheric pressure inside the diaphragm bellows simulated inspiratory effort. The lung compliance (CL) was set at 0.3 L/cm H2O or 0.05 L/cm H2O. The airway resistance (Raw) was set at 5, 20, or 50 cm H2O/L/s. Pressure support levels were increased from 0 to 45 cm H2O. Sensitivity was set at 2 cm H2O. No bias flow was used at first. The WOBtr was calculated using a pressure-volume (P-V) loop derived from the diaphragm bellows movement during the triggering period. We determined WOBtr and its dependence on the various pressure support (PS) levels, CL and Raw. To evaluate the effects of bias flow on WOBtr and triggering delay, a ventilator was put in the PSV mode, with various bias flow rates (from 0 to 20 L/min) at a sensitivity of 2 cm H2O. We found that when no bias flow was used, WOBtr increased with an increase in both Raw and end-expiratory lung bellows pressure which was considered as auto-PEEP. With bias flow, both triggering delay and WOBtr increased. An increase in bias flow at a given PS level resulted in both decreased pressure support time and tidal volume (VT). It is concluded that the bias flow system is not desirable for use during PSV.