Ewan C. Goligher MD, PhD
Assistant Professor of Medicine
Interdepartmental Division of Critical Care Medicine
University of Toronto
Scientist, Toronto General Hospital Research Institute
Critical Care Medicine
University Health Network, Toronto General Hospital
What is Myotrauma?
Myotrauma refers to the deleterious structural changes occurring in the diaphragm as a consequence of how mechanical ventilation is applied. Myotrauma may result from several different mechanisms, including over assistance (resulting in disuse), under assistance (resulting in load-induced muscle injury), dyssynchrony (resulting in eccentric contractions), and excessive PEEP (resulting in acute sarcomere drop out). The evidence for over assistance myotrauma is very strong and evidence for under assistance myotrauma is growing. The other mechanisms require further research to confirm their existence but the available data suggests they may be important.
Why is it relevant?
Patients are ventilated in order to save their lives and restore their health. If ventilation is applied unsafely, then lung or diaphragm injury can result and patient outcomes are significantly impaired. Mechanisms of lung injury are well recognized and widely appreciated (barotrauma, volutrauma, atelectrauma, biotrauma). The term ‘myotrauma’ is intended to help clinicians communicate the importance of preventing diaphragm injury and atrophy during mechanical ventilation.
What is a mediation analysis?
Suppose you have three variables, A, B, and C. Suppose further that you know that A is associated with B, A is associated with C, and B is associated with C. Mediation analysis allows you to determine the extent to which A is associated with C because of their mutual association with B.
What was the reason for performing the analysis?
In clinical research, associations can be observed because there is a causal connection between two variables or because of confounding by association with other variables. A causal connection would entail that modifying one variable would affect the other. If you know that the association between A and B is probably causal, and you know that the association between B and C is probably causal, then if the association between A and C is explained by B, then A probably has a causal association with C (at least in part).
In this scenario, A = diaphragm thickening (a marker of breathing effort); B = changes in diaphragm thickness (a marker of injury); and C = clinical outcome. In a previous study, we observed that diaphragm thickening predicted clinical outcomes. The mediation analysis is designed to test whether this association is the result of the association between diaphragm thickening and changes in diaphragm thickness over time.
What do the results of the mediation analysis suggest?
We found that the association between diaphragm thickening fraction and three different clinical outcomes was mediated in part by changes in diaphragm thickness. Therefore if it’s plausible that the diaphragm thickening causally affects diaphragm thickness (i.e. disuse causes atrophy), and diaphragm injury causes prolonged ventilation, then its also plausible that diaphragm thickening causally influences clinical outcomes. This is important because we can modify diaphragm thickening (i.e. patient breathing effort) by modifying ventilator support and sedation. The mediation analysis provides a very strong rationale for future trials of diaphragm-protective ventilation strategies and suggests that targeting an intermediate level of breathing effort (similar to that of healthy subjects breathing at rest) could improve clinical outcomes.
What are the next steps in clinical practice and research?
The next step in research is to design and test diaphragm-protective ventilation strategies that prevent Myotrauma. We need to find out when and in whom diaphragm protection can be achieved and we need trials to confirm that these approaches really improve outcomes. For now, clinicians should more regularly monitor patient breathing effort and try to maintain patient breathing effort at low normal levels, both for patient comfort (to avoid dyspnea) and (probably) to avoid diaphragm injury and weakness.
Goligher EC, Brochard LJ, Reid WD, et al. Diaphragmatic myotrauma: a mediator of prolonged ventilation and poor patient outcomes in acute respiratory failure. Lancet Respir Med 2018; published online Nov 16. DOI:10.1016/S2213-2600(18)30366-7.