Objective: To study the impact of low pulmonary vascular pressure on ventilator-induced lung injury.
Design: Randomized prospective animal study.
Subjects: Isolated perfused rabbit heart-lung preparation.
Settings: Animal research laboratory in a university hospital.
Interventions: Twenty isolated sets of normal lungs were perfused (constant flow, 0.3 L/min; left atrial pressure, 6 mm Hg), ventilated for 20 min (pressure control ventilation, 15 cm H2O; baseline period), and then randomized into three groups. Group A (control, n = 7) was perfused and ventilated as previously described during three consecutive 20-min periods. In group B (high airway pressure/normal left atrial pressure, n = 7), pressure control ventilation was 20, 25, and 30 cm H2O during each period. Group C (high airway pressure/low left atrial pressure, n = 6) was ventilated as group B but, in contrast to groups A and B, left atrial pressure was reduced to 1 mm Hg.
Measurements and main results: The rate of edema formation (WGR, weight gain per minute normalized for initial lung weight) and the ultrafiltration coefficient (Kf) were measured during and after each period and their changes from baseline [DeltaWGR (edema formation index) and DeltaKf (vascular permeability index)] calculated to compare groups. The incidence and timing of vascular failure were compared. Vascular failure was considered to be present if all the following conditions were met: pulmonary hypertension, accelerated weight gain, and occurrence of fluid leak from the lungs. At the end of the study, DeltaWGR (g.g.min(-1)) was higher in group C (0.54 +/- 0.17) than in groups B (0.08 +/- 0.04) and A (0.00 +/- 0.01; p<.05), as well as in group B compared with A (p <.05). Similar differences between groups (p <.05) were found for DeltaK (g x min(-1) x cm H2O(-1) x 100 g(-1)): C, 7.24 +/- 2.36; B, 1.40 +/- 0.49; A, 0.01 +/- 0.03. Vascular failure was not observed in groups A and B but occurred in all but one preparation in group C (p <.05; C vs. A and B).
Conclusion: Reducing left atrial pressure results in more severe ventilator-induced lung injury. These results suggest that lung blood volume modulates cyclic tidal lung stress.