The regional lung response to mechanical ventilation

Mechanical ventilation is a lifesaving therapy for patients with acute respiratory failure. Unfortunately, mechanical ventilation can contribute to mortality in these critically ill patients through a process known as ventilator-induced lung injury (VILI). VILI is thought to occur through overdistension and/or cyclic stretch of the lung which leads to tissue damage and inflammation (biotrauma). However, due to the heterogenous response of the lung to mechanical ventilation, both overdistension due to high lung volume ventilation and cyclic stretch due to low lung volume ventilation can occur simultaneously within the same lung. In addition, the magnitude of VILI is influenced by the pre-existing injury that caused respiratory failure in the first instance. To date, the link between variation in the regional lung volume response and regional lung inflammation, and how different lung injuries alters these responses, is poorly understood. The aim of this Thesis was to assess the impact of mechanical ventilation on the regional lung volume response and biotrauma in the healthy, indirectly injured (endotoxemia) and directly injured (acid aspiration) lung. To assess the impact of mechanical ventilation on the regional response in the healthy lung (Chapter 2), adult female BALB/c mice were ventilated for 2 hours using either a protective (moderate PIP with PEEP) or injurious (high PIP with no PEEP) ventilation strategy. The regional FRC (the regional volume of air retained in the lung at end expiration) and tidal volume responses were obtained from analysis of dynamic high-resolution 4D CT lung images at baseline and after 2 hours of ventilation. The expression of 21 genes were measured by qPCR array and correlated with regional lung volume responses. Protein expression and neutrophilia were quantified by immunohistochemistry. To assess the effect of indirect lung injury on the response (Chapter 3), mice were exposed to lipopolysaccharide (LPS) by IP injection in saline or saline alone. Four hours later, lungs were imaged by 4D CT at baseline and after 2 hours of ventilation. The inflammatory response was evaluated by the expression of seven inflammatory genes by qPCR and neutrophilia was quantified by immunohistochemistry. To assess the effect of direct lung injury (Chapter 4), mice were exposed intra-tracheally (IT) to hydrochloric acid or saline. Lungs were imaged by 4D CT at baseline and after 2 hours of ventilation and the regional expression of four inflammatory cytokines was assessed by ELISA. In the healthy lung, there was regional variation in the lung volume response and gene expression, whereby regional tidal volume, and the expression of two genes (IL-6, P = 0.02 and Ccl2, P < 0.01), varied regionally depending on the ventilation strategy. The expression of IL-6 and Ccl2 was positively associated with regional tidal volume, but not with FRC, suggesting that overstretch is detrimental in the healthy lung. In endotoxemia (LPS exposure), there was also regional variation in FRC and the expression of the two inflammatory genes (IL-6 and Ccl2). The expression of IL-6 was negatively associated with FRC in the LPS treated mice, whereas there was no significant association between tidal volume suggesting that the endotoxemic lung is susceptible to low lung volume ventilation. In contrast, acid aspiration had no effect on regional FRC or regional tidal volume. There was also no effect on regional levels of the inflammatory cytokines (TNF-α, MCP-1, IL-1β and IL-6). Further analysis to assess the effect of IT fluid administration on the response was undertaken by comparing lung volume data from the IP saline exposed group (Chapter 3) with the saline IT exposed mice (Chapter 4). Saline exposure via the IT route led to significant increases in FRC in some regions (P = 0.01). This Thesis showed that healthy lung is susceptible to regional overdistension, while the endotoxemic lung is susceptible to low lung volume ventilation. In both cases, these were associated with increased expression of IL-6. Unfortunately, pre-injury in the acid aspiration model was not established, which may be due to the high pH used, short exposure duration and the impact of liquid aspiration. Therefore, it is unclear whether this injury alters the regional stretch and inflammatory response compared to endotoxemia and further experiments are required. Overall, the data from these experiments provide critical insight into the regional lung response to mechanical ventilation and the influence of existing injury on the response. In particular, these results highlight the importance of the balance between cyclic stretch and over-stretch (tidal volume) and how each of these can contribute to lung inflammation and, potentially, patient outcomes