Models and mechanisms of acute lung injury caused by direct insults

Abstract

Acute lung injury (ALI) and its more severe form acute respiratory distress syndrome (ARDS) are life-threatening diseases that are characterized by acute onset, pulmonary inflammation, oedema due to increased vascular permeability and severe hypoxemia. Clinically, ARDS can be divided into ARDS due to direct causes such as pneumonia, aspiration or injurious ventilation, and due to extrapulmonary indirect causes such as sepsis, severe burns or pancreatitis. In order to identify potential therapeutic targets, we asked here whether common molecular mechanisms can be identified that are relevant in different models of the direct form of ALI/ARDS. To this end, we reviewed three widely used models: (a) one based on a biological insult, i.e. instillation of bacterial endotoxins; (b) one based on a chemical insult, i.e. instillation of acid; and (c) one based on a mechanical insult, i.e. injurious ventilation. Studies were included only if the mediator or mechanism of interest was studied in at least two of the three animal models listed above. As endpoints, we selected neutrophil sequestration, permeability, hypoxemia (physiological dysfunction) and survival. Our analysis showed that most studies have focused on mechanisms of pulmonary neutrophil sequestration and models with moderate forms of oedema. The underlying mechanisms that involve canonical inflammatory pathways such as MAP kinases, CXCR2 chemokines, PAF, leukotrienes, adhesions molecules (CD18, ICAM-1) and elastase have been defined relatively well. Further mechanisms including TNF, DARC, HMGB1, PARP, GADD45 and collagenase are under investigation. Such mechanisms that are shared between the three ALI models may represent viable therapeutic targets. However, only few studies have linked these pathways to hypoxemia, the most important clinical aspect of ALI/ARDS. Since moderate oedema does not necessarily lead to hypoxemia, we suggest that the clinical relevance of experimental studies can be further improved by putting greater emphasis on gas exchange.

Introduction

Acute lung injury (ALI) is a life threatening condition with an incidence in the United States of 79 per 100,000 and an overall mortality that is comparable to the number of deaths attributed to breast cancer (Rubenfeld et al., 2005). The diagnosis for ARDS is based on those four well known criteria defined by the American-European Consensus Conference for Acute Respiratory Distress Syndrome (ARDS) in 1994 (see Table 1). These criteria reflect the underlying pathophysiology and the severity of the disease: (i) ALI is an acute, not a chronic disease. (ii) The bilateral infiltrates on chest radiographs indicate pulmonary inflammation caused by leukocyte infiltration and plasma exudation. (iii) The pulmonary wedge pressure cut-off emphasizes that the exudate is of inflammatory nature and not caused by increased hydrostatic pressure; in fact, increased vascular permeability is considered a hallmark of ARDS. (iv) To be called ARDS, the inflammation must be sufficiently severe and lead to hypoxemia: if PaO₂:FiO₂ < 300 mmHg, it is called Acute Lung Injury (ALI), if PaO₂:FiO₂ < 200 mmHg Acute Respiratory Distress Syndrome (ARDS). At present, therapy for ARDS is largely supportive and only one phase III trial has ever shown a beneficial effect on the mortality of this disease (ARDSNet, 2000). In addition to defining ARDS, the American-European Consensus Conference distinguished pulmonary ARDS, caused by insults directly affecting the lung parenchyma, e.g. pneumonia, gastric aspiration or inhalation, from extra-pulmonary ARDS, resulting from an acute systemic inflammatory response (Bernard et al., 1994). Clinically, direct ARDS is characterized by a stronger increase in lung elastance and appears more resistant to therapeutical interventions, e.g. PEEP, prone positioning or recruitment manoeuvres than indirect ARDS (Pelosi et al., 2001).

At present, therapy for ARDS is largely supportive and novel treatment options are urgently needed. Therefore, this is a highly active area of research and considerable progress has been made towards understanding the molecular mechanisms of pulmonary inflammation and ALI. Obviously, ALI involves a complex process and the pathomechanisms depend to some degree on the model studied. On the other hand, in the clinical situation ALI is embraced as one common entity independent of its aetiologies. Hence, from the clinical perspective it appears highly desirable that drugs for the treatment of ALI should be effective irrespective of the underlying causes, be they pneumonia, sepsis, acid aspiration or VILI. The identification of common pathways in different experimental models of ALI may help to identify pathways of clinical and therapeutic interest. Therefore, in the present review we have screened the literature for mechanisms that are shared in different models of ALI. Here we have focused on direct ALI, because it may represent a separate and relevant clinical entity and because commonly studied models of indirect lung injury such as endotoxemia or experimental peritonitis in most cases do only cause mild inflammation ((Matute-Bello et al., 2008); our own experience).

Thus, here we will discuss and compare three widely used models: (a) one based on a biological insult, i.e. instillation of bacterial endotoxins; (b) one based on a chemical insult, i.e. instillation of acid; and (c) one based on a mechanical insult, i.e. injurious ventilation. We only included studies if the mediator or mechanism of interest was studied in at least two of the three animal models listed above. As endpoints, we selected neutrophil sequestration, permeability, hypoxemia (physiological dysfunction) and survival (Table 2).

The focus of this review is not to compare models of acute lung injury (there are recent excellent reviews on that topic (Matute-Bello et al., 2008, Matute-Bello et al., 2011)), but to review the molecular mechanisms that are shared in the three models of direct ALI. It was our aim to collect evidence for the hypothesis that there are common mechanisms for direct ALI.