A calibrated volume ventilator (Puritan-Bennett 7200) was used to perform the experiment. Patients were observed for one hour on settings previously chosen by their attending physicians. The subjects were then allowed to breathe spontaneously (without assistance) through the ventilator for a target period of five minutes. Tidal volume, inspiratory flow rate, and diaphragmatic EMG were recorded simultaneously for five to seven sequential spontaneous breaths at the end of the five-minute period. Patients 1 and 7 were unable to breathe spontaneously for this length of time, so data were recorded after only two to three minutes of spontaneous breathing. Subjects then rested for ten minutes on their original ventilator settings. Five to seven assisted breaths at nine combinations of tidal volume and flow rate derived from the measured values during spontaneous breathing were delivered in random order. Tidal volumes of 1, 1.5, 2, and 2.5 times the patients spontaneous tidal volume and flow rates of 0.5, 1, 1.5, 2, and 2.5 times the spontaneous flow rate were used. Only nine tidal volume-flow rate combinations (of the 20 possible) were attempted with each patient, in order to reduce the possibility of fatigue. Combinations with high tidal volume and low flow rate were frequently omitted, as they interfered with adequate expiratory time. After another ten-minute rest, controlled breaths were delivered at the same tidal volume-flow rate combinations. other
For each tidal volume-flow rate combination recorded, the last three breaths showing a clear electromyographic peak, ie unobscured by ECG, were selected for analysis. For each breath the inspiratory volume at the peak of electromyographic activity was determined by direct inspection of the raw recordings of diaphragmatic EMC and tidal volume. The inspiratory volumes at the peak of electromyographic activity for the three breaths were then averaged to obtain the Voff at a particular tidal volume-flow rate combination. Voff was expressed as a fraction of the measured spontaneous tidal volume, as were the delivered tidal volumes and flow rates, in order to facilitate interpatient comparison. The Voff values for all of the patients were plotted against ventilator-delivered flow rate at isovolume; flow rates for an individual patient at the same tidal volume were related by a line through the points (Fig 1A). The Voff was plotted similarly against ventilator-delivered tidal volume at isoflow (Fig IB). Regression equations and the slope of the regression line were determined for each Voff vs flow rate (isovolume) and Voff vs tidal volume (isoflow) relation. The mean Voff for the entire group was determined using the Voff at all settings, for all patients.
Figure 1 A (left). Ventilator-delivered flow rate vs Voff. B (right). Ventilator-delivered tidal volume vs Voff. Tidal volume, flow rate, and Voff are expressed as fraction of spontaneous breath values. Each line in A (left) represents series of flow rates delivered at same tidal volume for one patient; in B (right), each line is series of tidal volumes delivered at same flow rate.