Introduction Mechanical ventilation promotes pulmonary inflammation and oedema formation, a process known as ventilator-induced lung injury (VILI). Various aspects of ventilation have been proposed as injurious, including over distension of alveoli and the repetitive collapse and reopening of lung units associated with atelectasis. Here we attempt to elucidate the impact of these distinct entities on the pathophysiology of VILI, using a mouse isolated perfused lung, which enables us to study the effects of stretch versus atelectasis in the absence of extra pulmonary factors.

Methods Lungs were obtained from male C57BL6 mice, and allocated to one of three groups, that is, control, atelectasis or high-stretch. All the lungs were ventilated with respiratory rate of 80/min, and perfused at 25 ml/kg/min in a recirculating manner with non-blood buffer for 3 h. In the control group, low tidal volume (7 ml/kg) with positive end-expiratory pressure (PEEP; 5 cmH₂O) and regular deep inflation (DI; 25 cmH₂O, every 15 min) was applied. The atelectasis group received the same low tidal volume, but neither PEEP nor DI. In the high-stretch group, lungs were ventilated with high tidal volume (30–32 ml/kg) and both PEEP (3 cmH₂O) and DI. Perfusate and lung lavage samples were taken at the end of experiments for analysis of total protein and chemokines.

Results The lungs in the atelectasis and high-stretch groups developed similar, severe pulmonary oedema as represented by increases in protein levels in lavage fluid. High-stretch induced substantial increases in both perfusate and lavage fluid chemokines, compared to controls. In stark contrast, the atelectasis group showed similar low levels of chemokines in perfusate, with only slight increases in lavage fluid chemokines, compared to controls (Abstract S107 Table 1).

Conclusion While both high-stretch and atelectasis can induce barrier dysfunction, only high-stretch induced substantial production of chemokines by the lung and their release into the circulation. These findings suggest that over distension of the lung, rather than collapse/reopening associated with atelectasis, primarily contributes to the exacerbated pulmonary and systemic inflammation during VILI. Our results may provide insights into why addition of PEEP to limit atelectasis has limited clinical outcome benefit in ventilated patients with acute lung injury.