a) Characterize how ventilator and patient variables affect tidal volume during high-frequency oscillatory ventilation; and b) measure tidal volumes in adults with acute respiratory distress syndrome during high-frequency oscillatory ventilation. Observational study. Research laboratory and medical intensive care unit. Test lung and patients with acute respiratory distress syndrome. Using a previously validated hot wire anemometer placed in series with a Sensormedics 3100B high-frequency ventilator, an endotracheal tube, and a test lung, tidal volume was measured at different combinations of frequency (4, 6, 8, 10, and 12 Hz), pressure amplitude (50, 60, 70, 80, and 90 cm H2O), mean airway pressure (20, 30, and 40 cm H2O), test lung compliance (10, 30, and 50 mL/cm H2O), endotracheal tube internal diameter (6, 7, and 8 mm), bias flow (20, 30, and 40 L/min), and inspiratory/expiratory ratio (1:2 and 1:1). In patients, tidal volume was measured at baseline ventilator settings and at baseline frequency +/-2 Hz and baseline pressure amplitude +/-10 cm H2O. Measured tidal volumes were 23-225 mL during high-frequency oscillatory ventilation of the test lung. A 2-Hz increase in frequency and a 10-cm H2O increase in pressure amplitude caused a 21.3% +/- 4.1% decrease and 21.4% +/- 3.4% increase in tidal volume, respectively. Decreasing endotracheal tube internal diameter from 8 mm to 7 mm and from 7 mm to 6 mm caused a 15.3% +/- 1.7% and 18.9% +/- 2.1% reduction in tidal volume, respectively. Increasing bias flow from 20 L/min to 30 L/min increased tidal volume by 11.2% +/- 3.9%. Further increases in bias flow, changes in compliance, and changes in mean airway pressure had little effect. Tidal volumes measured in acute respiratory distress syndrome patients were 44-210 mL. A 2-Hz increase in frequency was associated with a 23.1% +/- 6.3% decrease in tidal volume. In contrast to the test lung data, a 10-cm H2O increase in pressure amplitude resulted in only a 5.6% +/- 4.5% increase in tidal volume. Tidal volumes are not uniformly small during high-frequency oscillatory ventilation. The primary determinant of tidal volume in adults with acute respiratory distress syndrome during high-frequency oscillatory ventilation with the Sensormedics 3100B is frequency. Test lung findings suggest that endotracheal tube internal diameter is also an important determinant of tidal volume.