First here’s an update from TPM Reader WC (not their actual initials), a critical care doctor on the West Coast who our team has been in touch with since early in the crisis …
Saw your post on the “Developments on the treatment of Covid-19.”
As someone who has managed my fair share of ventilated Covid patients, I can say that the opinions offered by Dr Kyle-Sidell seem to have face validity and represent something of a developing consensus in the EM/Critical Care community.
The Covid-19 epidemic has been interesting from a medical perspective in that so much about it has been new and the management strategies have been in flux since the beginning. I don’t want to get to into the weeds technically, but the concepts have ranged widely as we have gotten more experience with this fairly new disease. We went from early intubation strategies with a paradigm of using ventilator protocols similar to ARDS (Adult Respiratory Distress Syndrome) to strategies trying to delay intubation as long as possible using modalities such as noninvasive positive pressure ventilation (aka CPAP or BiPAP) or High Flow Nasal Cannula oxygen — both controversial as they may raise the amount of virus aerosolized. And now there is this very wild notion of “permissive hypoxia” which is the concept that if the oxygen level is low but the patient looks OK you just let them run low (previously a low oxygen level – one can quibble about the threshold – was an absolute indication for intubation). Anyway, while Dr Kyle-Sidell’s comments were provocative, it’s just one more day, one more change in the management. Whether it’s a game-changer, I doubt, but rather one more step in the road of figuring out how best to serve the sickest of the sick patients with this disease.Here’s a note we received from longtime TPM Reader AM …
I am a physician and actively treating a number of COVID patients. Something you may not be fully aware of is that MDs nationwide have formed large (tens of thousands strong), private communities on social media to facilitate exchange of information on PPE best practices and PPE shortages, telehealth best practices, and most importantly information on what works and what doesn’t in COVID patients. We have been fortunate to have input from physicians in Italy in particular who have shared their experiences just as our first wave of patients was hitting the ICUs. I would go so far as to say that these groups have supplanted many traditional sources of information for front-line physicians. There is a broad lack of trust in our national leadership so we are relying on the only ones we can believe in ~ one another.
In this context, Dr. Kyle-Sidell’s video was widely shared and has garnered intense interest. It is buttressed by some very interesting laboratory data that suggests that COVID interacts with hemoglobin in a way that could explain both the hypoxia and other diffuse clinical manifestations of the disease ~ i.e., there is some plausible biochemistry supporting the clinical observation. We will see whether this line of thought goes anywhere but my impression is that many physicians are starting to think along these lines.WC sent this note in our on-going exchange of emails last night …
In most cases of respiratory failure (pneumonia, heart failure, ARDS) the small air cells, the alveoli, fill up with fluid or collapse. So we maintain a higher constant pressure (PEEP: Positive End Expiratory Pressure) in the airways to hold them open, and it works well. Um, usually. But COVID seems to be different. More oxygen (normal room air = 21% O2) by which I mean a higher fraction of oxygen delivery up to 100% definitely helps. But the positive pressure ladder which worked well in ARDS seems maybe not to work in covid, and maybe even harms the lungs. We’ve always known that a high level of positive pressure is a double edged sword, but in covid maybe the harms clearly exceed the benefits, which would be truly new.
This is not to say we should not ever intubate. But it does mean that a) we should avoid intubation as much as possible by increasing the inhaled oxygen and b) if we do have to intubate, then the high positive pressures we have traditionally used maybe are to be avoided, even if it means allowing the blood to have less oxygen in it that we would traditionally accept.
So why would we intubate if it’s harmful? So traditionally we would intubate for “low oxygen level” – and this (maybe) is out the window. But if a patient is too exhausted to continue to draw air in, or too out of it to protect their airway, these are still obligatory reasons to intubate. We can’t get around those. But it’s seeming clear that anything we can do to increase the number of ventilator-free days is a good thing, and this is a new understanding.
And having seen the videos, I also have to say the “crazy” hypothesis remains valid. But Dr Kyle-Sidell may be crazy but I don’t think he’s wrong. To be clear, he is also not a “lone voice” in the wilderness. He’s reflecting an evolving consensus among docs who have managed a lot of covid.Let me emphasize a basic editorial point. TPM is not the place to litigate emerging clinical protocols for novel diseases. And we’re not trying to. See this more as a window into emerging discussions among clinicians around the country wrestling with a novel disease. This isn’t unexpected. It is a truism of modern medical history that wars often see major advances in medicine. Doctors are overwhelmed with large numbers of novel or seldom seen injuries and through a grim process of trial and error they develop new insights into treatment. The origins of modern epidemiology are often traced to the work of John Snow who traced the origins of an 1854 cholera outbreak in London to hand pumped wells that were communicating the contagion. It would be surprising if new methods or certainly more targeted treatments for COVID-19 didn’t emerge from this. Whether this points to one of them is a different matter entirely.
I would recommend people who are interested in this read this article that was published this morning in the must-read StatNews. It looks at these general treatment questions, the different doctors I’ve referenced above in a more global (in both senses of the word) context. Anyone who is reading this for more than news curiosity and general knowledge I highly recommend you focus on the articles I’ve referenced in Medscape and StatNews in this and the earlier post. (Here is the preprint of the Gattinoni article I referenced yesterday.)
Meanwhile TPM Reader NC, an intensivist and pulmonologist at a major academic medical center, sounds a more cautionary note …
Please be cautious with pieces like this. To be horribly blunt, this came off not dissimilar to Trumps uninformed grasping at hydroxychloroquine. While there is a well know (and respected) expert in the distant background (Gattinoni), a grainy YouTube video of a random critical care doc without training in pulmonary disease is … well, the optics are bad.
For the details. What is being proposed isn’t something terribly new. Our field is full of horrible disappointments, treatments that were based on large, very well run randomized controlled trials and then turned out to be useless, or even harmful. The list is long, activated protein C for sepsis, tight glycemic control, early goal directed therapy for sepsis, pulmonary artery catheters, steroids for x, y and z. However the treatment of ARDS using low tidal volume ventilation is the exception. It not only works, its made a visible change in patient outcomes in ICUs over the past decade.These are cautions I hope it is clear I take very much to heart.
In a follow up email NC provided this helpful context.
Yes, the observation that this disease is very much hypoxia driven and less about stiff lungs is true, and being discussed a lot, but doesn’t really get into the media as its kind of esoteric. Its the driver behind the deaths, rather than how our usual ARDS patients die. ARDS isn’t a disease, its a fixed set of clinical findings that are used to define a syndrome that we treat with a single common model. COVID patients fit the definition, but behave outside the norm. So they absolutely have ARDS (since its a syndrome, not a disease), but maybe their form of ARDS doesn’t slide neatly into the standard treatment model.
Re Gattinoni. It’s hard to say. It’s very much inside baseball. Gattinoni is super smart and has made massive contributions to this very field, but we’re still talking about stuff thats pretty theoretical and his articles are not going through the normal peer review process to publication.
My take would be in a different direction. Gattinoni is noting that there is variability in the disease and that treatment can’t be entirely protocol driven. ARDS is the poster child for protocol driven care. So we badly need physicians on the front line that can make decisions about when to follow protocols and when not to. I think most of us who are Pulmonary Critical Care trained are doing that now. However when you burn through us due due to lack of PPE etc, then the remaining providers will have to use protocols.This is the most interesting part of the discussion to me as an outside, lay observer. The point of common agreement appears to be that the standards ARDS protocol is not appropriate for all patients; that you need experienced clinicians who can depart from standard protocol based on the nuances of specific patients’ course of disease. The complication is that clinicians with deep experience in respiratory critical care are in short supply at the moment. And an additional hurdle is that the less invasive interventions are actually more dangerous to the clinicians who are often using inadequate personal protective equipment. There is a highly complicated set of trade offs in a context where there is limited knowledge, limited resources (human and technical) and mass mortality.
Josh Marshall, TPM | Read more:
Image: YouTube
[ed. Do check out the original post. See also: This post about ventilators, and how they could be doing more harm than good (which seems to dovetail with what's being said here) (TPM). Finally there's this: From Fine to Flailing (Reuters).]
