To the Editor:
I have read with interest the study of lung inhomogeneities and time course of ventilator-induced mechanical injuries, recently published in Anesthesiology by Cressoni et al.1
This study aimed at investigating the genesis of ventilator-induced lung injury in healthy piglets, with particular attention to the interface between inhomogeneous lung structures, and at determining whether the induced lung injury was associated with collapse or consolidation.
In 6 of the 12 animals considered, approximately 7.5% of the total lung volume had abnormal density on computed tomography scan, indicating consolidated tissue, and affecting respiratory system elastance. It is interesting that, if computed tomography scanning had not been performed, this potentially important abnormality could have been neglected. This is an important factor in the study of animals that are presumed healthy, but where health status is often overlooked, possibly affecting the results and conclusions of the studies.
One of the most important experimental details in this study is that the piglets were ventilated with strain greater than 2.5 (tidal volume/end-expiratory lung volume), corresponding to a tidal volume of about 40 ml/kg. This value is more than three times greater than the 12 ml/kg found harmful for the ventilation of patients with the acute respiratory distress syndrome2 and was chosen to apply very high stress and strain to the lung parenchyma. As stated, it is known that “excessive stress and strain induce lung injury.” In this sense, the observation that abnormally large tidal volumes were associated with changes in lung density at the interface between structures with different extensibility becomes difficult to interpret in relation to a critical care scenario. Here, much smaller stress and strain would be applied to the lung, and the role of stress raisers may not be as important as other factors such as, for example, overdistension. As the authors state, the results from the experiments as they were performed are not transferrable to human patients. It is possible that similar experiments performed at stress and strain values resembling those used in clinical settings (and scaled to the pulmonary mechanics of the animal species used) may generate more clinically important evidence in support of the stress raisers hypothesis.
A minor but potentially confounding element of the experimental design is the large fio2 (50%) used throughout the study and the associated high Pao2/fio2. It is possible that absorption atelectasis may have occurred, especially during the long duration of the experiments, and contributed to the recruitable nature of alveolar collapse.
The authors declare no competing interests.