Corticosteroids in acute lung injury and acute respiratory distress syndrome

The authors reply:
We thank Drs. Bauer and colleagues for their comments on the Confalonieri study and whether this should have been included as part of our meta-analysis. The impact of this study on the overall findings of our meta-analysis is minimal, as the study carries very little weight in the random effects model (Fig. 2 in our paper). If this study is excluded, the finding is still very similar to our original result (relative risk = 0.67 vs. 0.62; the p value is unchanged at .01).