Setting mean airway pressure during high-frequency oscillatory ventilation according to the static pressure--volume curve in surfactant-deficient lung injury: a computed tomography study.

BACKGROUND Numerous studies suggest setting positive end-expiratory pressure during conventional ventilation according to the static pressure-volume (P-V) curve, whereas data on how to adjust mean airway pressure (P(aw)) during high-frequency oscillatory ventilation (HFOV) are still scarce. The aims of the current study were to (1) examine the respiratory and hemodynamic effects of setting P(aw) during HFOV according to the static P-V curve, (2) assess the effect of increasing and decreasing P(aw) on slice volumes and aeration patterns at the lung apex and base using computed tomography, and (3) study the suitability of the P-V curve to set P(aw) by comparing computed tomography findings during HFOV with those obtained during recording of the static P-V curve at comparable pressures. METHODS Saline lung lavage was performed in seven adult pigs. P-V curves were obtained with computed tomography scanning at each volume step at the lung apex and base. The lower inflection point (Pflex) was determined, and HFOV was started with P(aw) set at Pflex. The pigs were provided five 1-h cycles of HFOV. P(aw), first set at Pflex, was increased to 1.5 times Pflex (termed 1.5 Pflex(inc)) and 2 Pflex and decreased thereafter to 1.5 times Pflex and Pflex (termed 1.5 Pflex(dec) and Pflex(dec)). Hourly measurements of respiratory and hemodynamic variables as well as computed tomography scans at the apex and base were made. RESULTS High-frequency oscillatory ventilation at a P(aw) of 1.5 Pflex(inc) reestablished preinjury arterial oxygen tension values. Further increase in P(aw) did not change oxygenation, but it decreased oxygen delivery as a result of decreased cardiac output. No differences in respiratory or hemodynamic variables were observed when comparing HFOV at corresponding P(aw) during increasing and decreasing P(aw). Variation in total slice lung volume (TLVs) was far less than expected from the static P-V curve. Overdistended lung volume was constant and less than 3% of TLVs. TLVs values during HFOV at Pflex, 1.5 Pflex(inc), and 2 Pflex were significantly greater than TLVs values at corresponding tracheal pressures on the inflation limb of the static P-V curve and located near the deflation limb. In contrast, TLVs values during HFOV at decreasing P(aw) (i.e., 1.5 Pflex(dec) and Pflex(dec)) were not significantly greater than corresponding TLV on the deflation limb of the static P-V curves. The marked hysteresis observed during static P-V curve recordings was absent during HFOV. CONCLUSIONS High-frequency oscillatory ventilation using P(aw) set according to a static P-V curve results in effective lung recruitment, and slice lung volumes during HFOV are equal to those from the deflation limb of the static P-V curve at equivalent pressures.