Assess the accuracy of four different methods of measuring tidal volume during simulated high-frequency oscillatory ventilation.
In vitro study.
Three differential pressure pneumotachometers, a modified Pitot tube, an ultrasound flowmeter, and an adult hot wire anemometer.
Each device was placed in series with a Sensormedics 3100B high-frequency ventilator and an 8.0-mm endotracheal tube attached to a 48.9-L plethysmograph. Inspiratory/expiratory ratio was fixed at 1:1 and mean airway pressure at 10 cm H2O. Tidal volumes were calculated at each combination of frequency (f: 3, 4, 6, 8, 10, 12 Hz) and pressure amplitude (ΔP: 30, 60, 90 cm H2O) by digital integration of the sampled flow signals from each device and compared with those calculated from pressure changes within the plethysmograph. The protocol was repeated after incorporation of frequency-specific calibrations to the flow-measuring algorithm of each device. The hot wire anemometer was further evaluated at Fio2 of 1.0, 37°C, 80% humidity, mean airway pressure of 20 cm H2O, and an inspiratory/expiratory ratio of 1:2.
Measurements and Main Results:
Tidal volumes were 36–305 mL. Bland-Altman analysis demonstrated that each device exhibited systematic bias before frequency-specific adjustment. After frequency-specific adjustment of the flow-measuring algorithm, the two most accurate and precise devices were the Hans Rudolph pneumotachometer, which exhibited a mean error of 0.2% (95% confidence interval, −3.0% to 3.4%), and the hot wire anemometer, which had a mean error of −1.1% (95% confidence interval, −5.5% to 3.3%). The hot wire anemometer remained accurate at Fio2 1.0, 37°C, 80% humidity, mean airway pressure of 20 cm H2O, and an inspiratory/expiratory ratio of 1:2.
Tidal volume can be measured during high-frequency oscillatory ventilation using a variety of techniques. Frequency-specific calibration improves the accuracy and precision of tidal volume measurements. Hot wire anemometry exhibits stable performance characteristics across the range of temperature, humidity, Fio2, and inspiratory/expiratory ratios encountered clinically, has a small deadspace, is unaffected by mean airway pressure, and is therefore suitable for clinical applications.