We appreciate the comments from Drs Thickett and Perkins and welcome the opportunity to further discuss the potential roles of interleukin 1 (IL1) in the pathogenesis and repair of acute lung injury.

Regarding the differences in IL1β levels in bronchoalveolar lavage fluid (BALF) obtained from mice and humans, we do not believe that the differences are surprising. IL1β levels are influenced by the lavage volume and the specific assays used. The primary finding is that IL1β mRNA expression and protein levels are markedly increased in the lung early in the course of ventilator induced lung injury.

We agree that the potential broader role of IL1 in alveolar repair and lung fibrosis should be considered when designing future studies of IL1 blockade for acute lung injury. Because of space limitations, we could not elaborate on this important issue in our manuscript.1 Previous clinical studies have reported that the majority of the proinflammatory activity in BALF is attributable to IL1.2 Through both neutrophil recruitment and an effect of epithelial cells, IL1 induces an increase in permeability to protein.1 IL1β also downregulates epithelial sodium channel (ENaC) expression and impairs vectorial fluid transport.3 Together, these effects favour pulmonary oedema formation, the hallmark of acute lung injury and ARDS. Although we have found that IL1 impairs alveolar barrier permeability, previous work from our group has demonstrated that IL1 promotes alveolar epithelial cell migration.4 5 It is conceivable that blocking IL1 signalling could interfere with normal alveolar epithelial cell migration over the basement membrane during the repair phase of acute lung injury. However, one recent study found that mesenchymal stem cells prevented both acute lung injury and fibrosis following bleomycin administration in mice. The effect was attributable to IL1 receptor antagonist expression in the stem cells.6 Additionally, chronic overexpression of IL1β induces acute lung injury followed by pulmonary fibrosis,7 although the mechanisms for the acute inflammatory response and later fibrosis may be distinct.8 Together these data show that IL1 signalling may govern a broad spectrum of inflammatory and repair processes in the injured lung. Differences in the timing of IL1 blockade may have different effects on injury and repair. Our hypothesis is that early blockade of IL1 signalling may limit the quantity of pulmonary oedema by preserving barrier function and sodium transport, while later IL1 blockade may affect epithelial repair and fibrosis. Additional studies of transgenic mice and IL1 receptor antagonist in other models of acute lung injury and fibrosis may shed more light on how the timing of IL1 signalling during lung injury influences the diverse effects of this cytokine.

Previous clinical trials have not directly addressed the question of the efficacy of IL1 receptor antagonist in patients with acute lung injury. Given the lack of effective therapies for this syndrome of acute respiratory failure in critically ill patients, we believe that further investigation of IL1 receptor antagonist is warranted.