A J Nam
,
R G Brower
,
H E Fessler
,
B A Simon
ABSTRACT
Mechanical ventilation in patients with acute respiratory distress syndrome and acute lung injury (ALI) remains a difficult challenge because of the conflict between maintaining adequate gas exchange and furthering lung injury via overdistention. In a recent study, Lefevre and colleagues (Am. J. Respir. Crit. Care Med. 1996;154: 1567-1572) suggested that mechanical ventilation with natural biologic variability (BV) in breath-to-breath respiratory frequency (f) and VT could reduce lung injury and improve gas exchange without increases in mean airway pressure (Paw) or peak inspiratory pressure (PIP). However, significant differences in cardiac output (CO), Pa(CO(2)), pH, and delivered VT between the treatment groups in their study could have influenced these results. Because of the potential implications of these findings for patient care, we attempted to confirm these findings by Lefevre and colleagues in a canine model of oleic acid-induced lung injury. Eighteen mongrel dogs were anesthetized in the supine position, paralyzed, and mechanically ventilated with 50% O(2) at f = 15 breaths/min, and VT was adjusted to achieve an end-tidal CO(2) of 30 to 35 mm Hg. Lung injury was produced by infusion of 0.06 ml/kg oleic acid solution into the right atrium over a 30-min period. Animals were then randomized to either conventional ventilation at the baseline settings (n = 9) or to BV at the same mean VT and f (n = 9). Both groups received comparable degrees of injury, and hemodynamic and ventilatory parameters were closely matched, with no differences in mean VT, PIP, mean Paw, Pa(CO(2)), pH, CO, pulmonary artery occlusion pressure, or arterial pressure (Pa). However, no differences between the two groups were found in Pa(O(2)), shunt, or static compliance over a 4-h period. When hemodynamic and ventilatory parameters were well matched in a canine model of ALI, BV showed no advantage over conventional ventilation at constant VT and f.