Comparison of variable and conventional ventilation in a sheep saline lavage lung injury model

Objective: There has recently been considerable interest in alternative lung-protective ventilation strategies such as variable ventilation (VV). We aimed at testing VV in a large animal lung injury model and exploring the mechanism of improvement in gas exchange seen with VV. Design: Randomized, controlled comparative ventilation study. Setting. Research laboratory at a veterinary hospital. Subjects: Female sheep weighing 59.8 +/- 10.57 kg and excised calf lungs. Interventions. In a sheep saline lavage model of lung injury, we applied VV, whereby tidal volume (V-T) and frequency (f) varied on each breath. Sheep were randomized into one of two groups (VV, n = 7; or control, n = 6) and ventilated for 4 hrs with all mean ventilation settings matched. Measurements and Main Results: Gas exchange, lung mechanics, and hemodynamic measures were recorded over the 4 hrs. VV sheep showed improvement in gas exchange (i.e., oxygenation and carbon dioxide elimination) and ventilation pressures (i.e., reduced mean and peak airway pressures) but control sheep did not. VV sheep also displayed lower-lung elastance and mechanical heterogeneity in comparison with control sheep from 2 to 4 hrs of ventilation. To study the mechanism behind improvements seen with VV, we examined the time course associated with the enhanced recruitment occurring during VV in eight saline-lavaged excised calf lungs. We found that the recruitment associated with a larger V-T during VV lasted over 200 secs, nearly an order of magnitude greater than the average time interval between large V-T deliveries during VV. Conclusions: The application of VV in a large animal model of lung injury results in improved gas exchange and superior lung mechanics in comparison with CV that can be explained at least partially by the long-lasting effects of the recruitments occurring during VV.