Read on Twitter
Interesting letter from UCLH regarding use of CPAP in Covid https://link.springer.com/article/10.1007/s00134-020-06304-y
Notable points include the high rate of CPAP use at presentation (86%), as I think the ICU community knows UCLH were enthusiastic about use of CPAP in COVID19- others less so. So this is a relatively non-selective cohort which makes juding outcomes easier
And those outcomes indicate a 66% failure rate (either death for those in whom CPAP was ceiling of care, or transition to mechanical ventilation). Thatâs quite a high failure rate by my book, and above the 47% reported for severe ARDS in Lung-Safe https://jamanetwork.com/journals/jama/fullarticle/2492877
The mortality amongst those who transitioned to IMV was 55%. Thatâs also pretty high, and other units with more liberal use of early mechanical ventilation report lower rates. HOWEVER failure on CPAP almost certainly selects out patients who are sicker (this letter shows that)
What this paper canât answer is whether these patients would have done better had they been ventilated earlier. There are good physiological reasons why this might be the case, self-ventilating patients can still injure their lungs-indeed this is more likely with more severe ALI
Animal models show this https://journals.lww.com/ccmjournal/Abstract/2013/02000/The_Comparison_of_Spontaneous_Breathing_and_Muscle.19.aspx
Indeed hyperventilation itself can induce injury https://pubmed.ncbi.nlm.nih.gov/3230208/
And from the Lung-Safe study mentioned earlier, early NIV was associated with increased mortality Vs mechanical ventilation - especially in severe disease. https://www.atsjournals.org/doi/10.1164/rccm.201606-1306OC
So the question is, in those who were failures (remember these were 66% of all CPAP patients) would they have done better if ventilated early- there is reasonable reasons to believe this would be the case. And for the 34% successes, would they have been harmed by IMV?
The risks of IMV are well known, sedation, VAP (very common in Covid) and lung injury. https://ccforum.biomedcentral.com/articles/10.1186/s13054-021-03460-5
But these occur with prolonged and injudicious use of the ventilator. Short-duration IMV, using lung protective strategies is unlikely to cause much harm.
So what should we take from this.
1) CPAP is no cure-all for COVID
2) the more severe ones presenting illness is and the less early response to CPAP the more likely it is to fail
1) CPAP is no cure-all for COVID
2) the more severe ones presenting illness is and the less early response to CPAP the more likely it is to fail
Beyond that we are into speculation and âexpert clinician judgmentâ. What do I do? I still have a low threshold for ventilation, especially in severe hypoxia and failure to respond rapidly to CPAP (with self-proning if required and tolerated).
Is this approach correct? Hard to know for certain, but I do know ventilated mortality in the unit I work in was considerably lower than 55%.
The inherent bias of clinicians is to think that what they are doing is the best approach, which is why we have to do trials to see what truly is the best management. In the absence of trials we have to ask if observations will change our practice- in this case, for me, no.
