So we wrote this up last week, but the (relatively) slow process of submission made us decide to just go ahead and release it as pure #FOAMed literature, because while highly relevant and somewhat controversial 10 days ago, it is now almost already accepted! What an incredible (in all senses of the word) time we are in…
So here are our thoughts:
I think it is important to clarify that none of us are saying that ventilators should NOT be used. HFNC or NIV approaches will not avoid intubation and mechanical ventilation for all patients, only for some, and for those it doesn’t, mechanical ventilation is an absolute life-saver. For many laypeople and perhaps non-critical care or acute care physicians, this may have been lost in much of the social media discussion…
For anyone interested in the development of the CAB-RSS score, do get in touch, we are in the process of an inter-observer validity study and eventually will integrate in a decision making algorithm.
cheers
Philippe
thinking critical care 
I wonder if anyone would like to comment on the levels of EPAP. Gattinoni has claimed that there is little recruitability in L-type COVID mechanics, however this has been disputed by some of you guys (Josh for example). Assuming there is actually some recruitability and that there is a higher viscus component of the elasto-viscus model of atelectotrauma, should not a higher PEEP or EPAP be used to move to the right of the lung compliance curve and away from where the alveoli are likely to collapse (as long as it doesn’t move too far to the point of limit of lung compliance)?
Your algorithm suggests EPAPs or PEEPs of 5. As long as hemodynamics tolerate slightly higher pressures and you haven’t reached the upper right lung compliance inflection point, I don’t see much downside, and only a possible upside in preventing further VILI with perhaps PEEP of 10 instead of 5 as a starting point.
Interesting point. On the emcrit podcast, i remember finding the discussion between Cameron, and Josh about the utility of APRV. It was suggested that some of these patients are not in need of increased mean airway pressures, and that simply increasing the FiO2 with minimal PEEPs are the best way to go about this. What are your thoughts? It makes sense that in the L phenotype that the patients would not be in need of recruitment, but I believe imaging of these patients and clinical experiences have suggested that both of these phenotypes are responsive to PEEP. Where is the disconnect?
Interesting work. I have been contemplating a lot on this topic myself. I put together my thoughts in this article (https://criticalcare.imedpub.com/dilemmas-in-covid19-respiratory-distress-early-vs-late-intubation-high-tidal-volume-and-low-peep-vs-traditional-approach.pdf). Happy to help in creation of algorithms if still needed!