Prospective, randomized, controlled experimental study.Setting:
Hospital research laboratory.Subjects:
Thirteen anesthetized, mechanically ventilated pigs.Interventions:
We induced three types of acute lung injury: oleic acid injury (n = 4); ventilator-induced lung injury (n = 4); and pneumonia (n = 5). All three models were designed to initiate a similar severity of oxygenation impairment. RM methods tested were sustained inflation, incremental positive end-expiratory pressure (PEEP) with a limited peak pressure, and pressure-controlled ventilation with increased PEEP and a fixed driving pressure. From a baseline PEEP of 8 cm H2O, all interventions were tested using post-RM PEEP levels of 8, 12, and 16 cm H2O. Cardiac output by thermodilution and systemic and pulmonary artery pressures were measured frequently during the RM and for 15 mins after its completion.Measurements and Main Results:
During the RM, cardiac output decreased to a greater extent in the pneumonia model (0.49 of baseline cardiac output) than in the oleic acid injury (0.67 of baseline) or ventilator-induced lung injury (0.79 of baseline) models. Cardiac output recovered to the baseline value by 5 mins post-RM in oleic acid injury and ventilator-induced lung injury models. However, cardiac output remained decreased 15 mins post-RM in the pneumonia model. There were no differences in hemodynamic parameters among RM methods in oleic acid injury and ventilator-induced lung injury models. In the pneumonia model, however, cardiac output decreased to a greater extent during the RM with sustained inflation (to 0.33 of baseline cardiac output) compared with pressure-controlled ventilation (to 0.68 of baseline).Conclusions:
We conclude that RM transiently but profoundly depressed cardiac output in three models of acute lung injury. The results imply that a lung recruiting maneuver should be used with caution, especially when using sustained inflation in the setting of pneumonia.