© 2003 American Thoracic Society
Taking a Peep at the Upper AirwaysResearch Centre Borstel Borstel, Germany Cells sense and respond to numerous physical stresses. One clinically relevant source of physical stress is mechanical ventilation, which may affect cell function in a variety of ways (1–3). Deeper insight into these responses is important to improve our understanding of lung physiology and pathophysiology with the ultimate goal of improving clinical practice. A major impetus for such studies has been the desire to understand the iatrogenic effects of mechanical ventilation in acute lung injury. With this in mind, most studies have focused on the pulmonary parenchyma. Such studies led to the hypothesis that ventilation itself may be a cause of acute lung injury, among other things, by recruiting active neutrophils to the lungs (4), a concept that has been termed biotrauma (5). Mechanical ventilation, however, exerts forces throughout the entire tracheobronchial tree, not only in the lung periphery. It is there that Lim and Wagner (6) have made a novel contribution in this issue of the Journal (pp. 1068–1074). By using intravital microscopy of the rat trachea, they discovered that positive end-expiratory pressure (PEEP) of 8 cm H2O, but not of 4 cm H2O, causes neutrophil adhesion and recruitment in the tracheal vasculature. These findings show for the first time that mechanical ventilation may alter leukocyte transit not only in the pulmonary circulation, but also in the bronchial circulation that supplies the trachea. How does PEEP cause leukocyte activation in these vessels? A number of possibilities come to mind: The leukocytes could be activated by mediators released from overdistended alveoli, as a result of the hemodynamic alterations caused by increased PEEP, or by local mechanical forces. In elegant control experiments, Lim and Wagner were able to exclude blood-borne mediators and hemodynamic alterations. Thus the weight of evidence points to local mechanical forces. These local forces could be pressure (compressive stress) or stretch, both of which have been shown to activate airway epithelial cells in culture (7, 8). Indeed, airway epithelial cells can be activated by ventilation (9, 10). From these studies, one can envisage a scenario where activated epithelial cells send signals to nearby endothelial cells and to circulating leukocytes to mount an inflammatory response. And in fact, the present study does not rule out this possibility. Lim and Wagner (6) also suggest an alternative mechanism. They propose that the activation of endothelial cells is a direct consequence of vascular distension and—although they do not explicitly state this—may occur independently from the activation of epithelial cells. Lim and Wagner's proposition is based on the quantification of vascular distension by video microscopy as well as abatement of neutrophil activation by the selectin inhibitor fucoidin or by endothelin receptor antagonists. Endothelial cells are a major source of endothelins. It is well known that mechanical stress induces endothelin release (11), but this is the first study to demonstrate that this response can arise secondary to ventilation-induced cell activation. This study also reminds us that there is much more to endothelins than their well known vasoconstrictor and bronchoconstrictor properties (12). It thus appears important to further characterize the role of endothelins in ventilator-associated lung injury. Another intriguing finding is the protective effect of fucoidin, a nonselective inhibitor of the adhesion molecules that mediate leukocyte rolling: P-selectin on endothelial cells and L-selectin on leukocytes. This is intriguing, because another study that used in situ video microscopy demonstrated that vascular distension (application of increased pressure in constant-flow perfused nonventilated rat lungs) caused expression of P-selectin on pulmonary endothelial cells by a calcium-gated mechanism (13). The latter observation lends further support to the hypothesis that vascular distension may lead to activation of endothelial cells without needing signals from other cells. These mechanisms may be quite similar in different vascular beds, such as the bronchial circulation (6) and pulmonary veins (13). Thus, stretching the vascular endothelium, be it from the vascular side or airway side, appears to have similar consequences, something that is also illustrated by the observation that increases in both ventilation and perfusion pressure stimulate the production of nitric oxide in the vascular endothelium (14). The central role of the pulmonary vascular endothelium for the initiation of inflammatory responses is well appreciated. Currently, this concept largely embraces ‘biological’ (meaning receptor-mediated) stimuli, such as sepsis, but we may have to extend this concept to include mechanical stretch. At present, the clinical implications of the reported findings are less clear. To date, widespread tracheal inflammation has not being recognized as a critical issue in ventilated patients. And furthermore, patients with acute lung injury (see the ARDSnet ALVEOLI trial) or severe thoracic trauma (15) appear to tolerate PEEP levels of 15 cm H2O and above. This may well be because deformed cells can adapt to continuous stretch (and that is what PEEP provides) by inserting new phospholipids into their membranes to effectively reduce that stress (16). Nonetheless, the present work suggests tracheal inflammation as a potential complication of mechanical ventilation that should be investigated in the clinical setting. And if it is true that the different vascular beds in the lungs respond to stretch in a similar fashion, a major contribution of the work by Lim and Wagner may be that it provides us with an elegant model for studying the initial mechanisms leading to biotrauma. FOOTNOTES Conflict of Interest Statement: S.U. has no declared conflict of interest. REFERENCES
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
