ALI, like sepsis, is a clinical description and common endpoint of many pathophysiologic processes and should be considered a syndrome and not a disease. In considering therapeutic strategies for ALI, clinicians attempt to treat these common processes, address underlying etiologic factors, and, when possible, tailor treatment to specific underlying pathology.
Classically, ALI has been described as progressing through three stages: exudative, proliferative, and fibrotic [
8,
9]. Although different mechanisms of lung injury and severity of illnesses significantly influence the severity and duration of these stages [
10], the three-stage model has remained largely intact for four decades and serves as a useful frame of reference for discussion.
Exudative
this initial stage of ALI encompasses the first seven days of illness and is marked by a net efflux of proteinaceous material from the intravascular to the alveolar spaces. By definition this efflux is related to increased capillary permeability (i.e., a reduced reflection coefficient) and not hydrostatic forces (i.e.,

an elevated left atrial pressure). The alveolar exudate reduces lung compliance and increases alveolar surface tension both by virtue of the increased viscosity of the exudate compared to air and by pulmonary surfactant neutralization [
11–
13]. As vascular leak occurs to varying degrees, lung compliance is heterogeneous leading to focal areas of atelectasis and the patchy bilateral infiltrate on chest X-ray classic of ALI. With positive pressure ventilation, this heterogeneous lung compliance leads to relative overdistention of more normal alveolar units and underinflation of lower compliance ones. Perfusion of inadequately ventilated lung units leads to pulmonary venous desaturation and the hypoxemia of ALI.
Proliferative
this second stage is a pathological fibroproliferative response to the initial injury and is classically defined as occurring during the second week. Until recently, endogenous fibroblasts were thought to mediate this response; however, emerging evidence suggests that transformation of injured epithelial cells to fibroblast-like cells (epithelial-mesenchymal transition) may play a prominent role [
14]. ALI resulting from different mechanisms of injury has also been associated with the presence or absence of myofibroblasts [
15,
16]. As myofibroblasts exhibit a substantially enhanced fibroproliferative response to cytokines such as transforming growth factor-
β [
17,
18], there may be a role for cytokine antagonism in these patients. Regardless of fibroblast origin or phenotype, the lung's ability to turn off the fibroproliferative response and begin tissue remodeling is a critical determinant of outcome.
Fibrotic
two to three weeks following the initial injury, the lung parenchyma either undergoes tissue remodeling leading resolution or the fibroproliferative response is not turned off and fibrosis results. Patients who initiate lung remodeling typically will have near-normalization of pulmonary function six months later [
19]. Patients who fail to initiate lung remodeling experience progressive fibrosis which leads to worsening respiratory insufficiency and death weeks to months later. In the adult population, some patients experience initial improvement in lung function only to develop idiopathic pulmonary fibrosis months to years later. Idiopathic pulmonary fibrosis also leads to progressive respiratory insufficiency and death over the course of several months to several years [
20].