In Reply

We thank Drs. Dreyfuss and Saumon for their comments on our recent work on one-hit models to study the biotrauma hypothesis.¹  I agree that our study supports some of his earlier concerns and that my view has been too simplistic. However, it is important to note that our recent work does not discredit the biotrauma hypothesis itself. What our work suggests is that the biotrauma hypothesis is difficult to study in one-hit models using ventilation as the only hit, because in such models, there is either mild inflammation without lung injury or severe mechanical injury followed by secondary inflammation. One-hit models, therefore, do not well recapitulate the clinical situation where injured and inflamed lungs are exposed to a second proinflammatory stimulus, namely ventilation.

To me, the biotrauma hypothesis still offers a relevant explanation for the findings of the low tidal volume Acute Respiratory Distress Syndrome Network (ARDSnet) trial.²  In that study, neither barotrauma, oxygenation, nor hypercapnia correlated with mortality—only inflammation did.^2,3  Similar correlations were found in a second, independent trial.^4,5  For obvious reasons, such studies cannot be repeated, and we will need complex and more realistic experimental animal models mimicking intensive care unit–like conditions to understand the complex interplay between ventilation and inflammation in patients with adult respiratory distress syndrome. In contrast to Dr. Dreyfuss, I believe that such studies are possible.