Does COVID-19 cause ARDS?

Welcome back to Depth of Anesthesia. My name is David Hao and I'm an anesthesia resident at the Massachusetts General Hospital. This episode will be featured on Anesthesiology News. You can find them at anesthesiologynews.com My guest today is Dr. Corey Hardin. He is board certified in internal medicine, pulmonary disease, and critical care medicine.

And he's currently an assistant professor of medicine at the Massachusetts General Hospital and Harvard Medical School. Dr. Harden, welcome to the show. Thanks. Glad to be here. So one thing I just want to touch on before we get started with the episode is I want to talk to you a little bit about the work you've been doing with uh Flair.

Flare is a collaborative effort between the Division of Pulmonary and Critical Care Medicine and the Department of Medicine. And the mission is to appraise some of the rapidly evolving literature currently on COVID-19. Yeah, that's right. It it and and it certainly started out as uh just an attempt to give uh nice concise summaries of the evolving literature. One of the things that we've found as as we've moved along is that There's a lot of underappreciated older literature that's very relevant to events in this outbreak.

So we've spent some of our issues also just reviewing older literature on ARDS and some critical care topics as they become uh relevant sort of on daily IC rounds on uh in the COVID-19 outbreak and also covering new uh new papers with a uh a particular focus on new case series and uh data from uh from the outbreak itself. That's really awesome. And I just saw you guys were featured on Wired magazine. The article is the COVID-19 newsletter that's by doctors for doctors. And I'll be sure to link that in the episode description as well as um show the listeners a way to to go ahead and sign up to that listserv.

And one of the uh s lines from that article, which I really like, is the underlying message of most flare issues is to avoid the temptation of anecdote. the allure of the silver bullet that will make the pandemic go away. And I think that's very much representative of where we're at now in terms of literature and COVID-19. Absolutely. I mean I think um this is obviously a devastating event for the uh the world as a whole and the healthcare system in particular and I think uh it's just only human to to think that uh we can somehow make this something that's not happening. That if you just have the right idea that this will all go away and the truth of the matter is that um that you just you know there are uh there are definitely evidence based ways of of dealing with this both from a public health and a clinical medicine st standpoint and you just have to do the work of of engaging with this with this event and taking care of our patients and taking care of our uh of our neighbors and um You know, it's there's not gonna be uh one thing that makes it all go away and we wake up tomorrow morning and it never happened.

We just uh that's just unfortunately not the way the world works, but I think it is really tempting to wish it were. I think that's absolutely right. Um and it so hopefully through the case of this episode we'll be able to explore one of the more common claims that I've personally seen out there not only in social media but some of the um popular press and that's related to whether or not COVID-19 is actually ARDS or acute respiratory distress syndrome And I'll start by um we usually start with the case, but I think in this uh circumstance it'd be just nice to jump straight into some of the claims that or arguments that people have made uh in favor of labeling COVID-19 as not being ARDS or rather distinct from ARDS.

And so the three claims laid out uh in the flare issue is that for the first Imaging abnormalities in COVID-19 are mild. Number two, pulmonary mechanical abnormalities, so that includes things like respiratory system compliance are mild. The third is hypoxemia is out of proportion to either the mechanical abnormalities or to the look to the level of dyspnea. So to our listeners, what's the level of your agreement with the claims that we've just laid out? And what's the level of evidence for what you believe?

So, Dr. Hardin, I think it'd be great if we just start off with what exactly is ARDS? How do we define it? And where does COVID-19 fit into all that? Yeah, so there are uh the there are two ways to to define ARDS. The standard one is uh clinical, so ARDS or the uh acute respiratory distress syndrome is just that.

It's a syndrome It's uh a particular constellation of clinical findings that have disparate causes. In the pre-COVID era, you know, these can be things from aspiration to pneumonia, bacterial or viral pneumonia in particular. It can be blood transfusion, it can be uh another overwhelming inflammatory insult such as trauma or pancreatitis. And uh the syndrome is defined by acute onset of hypoxemia bilateral infiltrates on chest immerging that are not associated with uh elevated wedge, which is to say not associated with heart failure. And uh a particular level of hypoxemia is quantified by the ratio of uh partial pressure of arterial oxygen to fraction of inspired oxygen, the so-called P to F ratio, uh that being less than uh uh 300 on uh PEEP or positive index toy pressure of five or more.

Those are the so-called Berlin criteria. They're the standard inclusion criteria in all of the large ARDS trials, which is to say patients who meet the Berlin criteria would have been included in the ARDS trials, so should have should be assumed to respond to treatments similarly to the patients in those trials. A more specific way to define ARDS, which is not necessarily available clinically is pathologically. So the pathologic pattern that's typically associated on autopsy, for example, of the lungs of patient with ARDS is a pattern called diffuse alveolar damage. which is characterized by um widespread cellular apoptosis, hyaline membranes, uh fibrin, fibrin plugs in the vasculature and and uh uh fibrin proliferation. Um so those are sort of the two ways that you define it I've heard of some uh critics claiming that the Berlin criteria might be overly inclusive.

Does that matter in the context of how we interpret applying some of those studies related to ARDS using the Berlin criteria? No, it does it's not relevant. They are overly inclusive. They're deliberately over-inclusive because they were designed to help uh identify patients for enrollment in clinical trials. So yes, it's absolutely true that patients who meet the Berlin criteria have a variety of clinical illnesses.

But you in your question you alluded to the key fact. Those were the inclusion criteria in the trials So to the extent that a patient before you meets the Berlin criteria, they were similar to the patients in the in the trials. And you might argue, well, these are mild or They might meet the Berlin criteria, but they're not my ARDS or what I think of as ARDS. And I would say that largely misses the point. If they meet the Berlin criteria, then the trials ought to have external validity with respect to that patient. you can expect a patient who meets the Berlin criteria will respond to tested therapies that were tested in ARDS trials, similarly to the other patients, because the full spectrum of uh disease uh that what that we see in patients before us who meet the Berl-Berlin criteria was represented in the trials, which is to say ARDS is extremely heterogeneous.

The patients enrolled in prior trials were also extremely heterogeneous. So yes, they're overly uh inclusive, but that doesn't really matter in terms of applying the results of clinical trials. Gotcha. So you're essentially saying that because all of these trials that we have done to date we're using and already we're already using such a widely inclusive uh criteria to define ARDS, the outcomes would then be relevant again to our current application of the Berlin criteria in its already uh inclusive state. Correct.

And so I know that a lot of the claims related to not only the imaging, but the pulmonary mechanical abnormalities of COVID-19 are largely anecdotal. But I think we've gotten to a point where there's now some case series and some studies looking at what exactly COVID-19 associated respiratory failure looks like. What do we have in terms of uh the published experience of defining COVID-19 failure? Yeah, so I mean I think there's a lot of discussion um in the popular press and in commentary forums in the medical press, which is to say editorials and letters and things outside of uh of peer-reviewed studies Suggesting that there's a large number of patients with normal pulmonary compliance presenting with uh hypoxemia.

And I would say that's just not borne out by the published case series. So, you know, we've certainly seen lots of discussions of sort of example patients who might have a pulmonary compliance of 60 for instance and severe hypoxemia. That's not what's in the published case series. So we now have published case series from Seattle. We have a number of them from China in particular, a couple ICU-only cohorts from China.

We have a 1,400 patient or roughly 1,400 patient cohort from northern Italy. We have and just this week we got a new uh cohort of three hundred odd patients from Cornell in New York. And what we're seeing is that to the extent that those patients publish those series published pulmonary compliance, it's around 30. So it's somewhere between 30 and 35 in the various case series. That's consistent with our MGH experience.

So, you know, that's not a normal compliance. Normal compliance is 100. On a vent, you might expect about 50, because compliance does go down a little bit in on the on the ventilator. But 30 is is an abnormal compliance. It's squarely in the range of of mild to moderate ARDS.

And it's just not true that any of the published series are reporting large numbers of patients with stone cold normal compliance. Now I will say there are definitely patients, both uh in our own MGH experience and in some of the published case series. where there's pretty normal pulmonary compliance. I'm not saying there aren't any patients with compliances of of greater than 50. There certainly are, and we've seen those at MGH. They are largely patients with mild gas exchange abnormalities, which correspond to those mild mechanical abnormalities.

So there's this mythical patient out there with an O2 SAT of 70% and a compliance of 60 I have not seen that patient, and that patient has not appeared in the published case series. And to review some of the literature that you referenced in the Flair series, it looks like an initial report of 138 from China had a mean p median p to f of 136, but the range extended from 103 to 234, so that's severe to mild. And then same thing from Washington State, so median P to F of 142, but again ranging from 94 to 177, and a median system compliance of a range of 25 to 36. That's correct. So these are these are abnormal mechanics and abnormal gas exchange and they're largely uh moving in concert.

And how should we think about this in the context of the Perceiva study? Yeah, so Prosiva is interesting. So um one of the things ARDS if you'll allow me to just go a little bit into the into the weeds here. So Praceva is a large multi-center randomized controlled trial Which demonstrates a substantial mortality benefit to prone ventilation in ARDS and is the basis for sort of our enthusiasm for prone ventilation in the current outbreak. Praceba, like most ARDS trials, um is uh enriched or was attempted to be enriched in severe patients.

And the reason for that Is ARDS is very heterogeneous and it's been hard to demonstrate a um uh a strong signal in a lot of ARDS trials. And so um Most trials tend to enroll just patients that are more more on the severe side. So Prasiva, the entry criteria to Prasiva was P to S P to F less than 150. That is uh that is the more severe side. So you might ask yourself whether uh it's fair to apply Prasiva to the full cohort of COVID-19 patients who might have better gas exchange than the inclusion criteria in Prusiva?

That's absolutely fair. Um at MGH and our own guidelines we use that cutoff of PDF less than 150 for prone ventilation, we've seen a lot of patients who are meeting that cutoff. And as you say, the mean P2Fs um and the ranges that are reported in the published case series are consistent with um uh there being a lot of patients who would meet Proceva criteria. And how about the Lung Safe trial? So Lung Safe is interesting and and we we spend a lot of time comparing to Lung Safe because You know, one of the things about COVID-19 is we have PCR testing for COVID-19.

And so we can identify a lot of patients and and we're, you know, most hospitals for anybody who's presenting with the fever or any symptoms are testing for COVID-19. And what that means is we have for, you know, one of the first times identified a large number of patients who are at risk for ARDS prior to their presenting with such severe respiratory failure that they're in the ICU. So we have a lot of patients on the floor who in previous times might not have been thought of as at risk for ARDS, who are being thought of in that context. Lung Safe was a large observational study, basically a snapshot in time of ARDS uh around the world And basically what they did is they surveyed a number a large number of ICUs across the world and they looked at ARDS by the Berlin criteria, so gas exchange. radiographic abnormalities, regardless of whether those patients have been identified as having ARDS by their providers, and regardless of whether they were getting treatment that was consistent with uh ARDS treatment guidelines at the time.

Predictably what they uncovered was um there were a large number of primarily mild patients who weren't necessarily getting labeled with ARDS, but nevertheless had ARDS by the Berlin criteria. So Lung Safe is uh a cohort which has a fuller spectrum of disease than things like Prosiva. And I think it's a useful comparison for COVID nineteen where we're identifying all of these people who meet the Berlin criteria and as we're seeing with all the discussion of is it ARDS, despite meeting the Berlin criteria, there are many people who are tempted to say, well they don't have ARDS because they don't have my ARDS. They don't have what I think of as ARDS, even though they meet the Berlin criteria.

And if you look at the average P2Fs and the average compliances that are being reported in COVID-19, they're very similar to what was reported in LungSafe. So I think lung safe is a useful comparison because it emphasizes the full spectrum of disease in ARDS and with COVID-19 we're seeing the full spectrum of disease. So essentially, based on a lot of these prior studies, even though COVID-19 patients might have some mild issues with respiratory compliance or a gas exchange, that's still within the spectrum of disease that's been previously reported in all of these major trials. That's correct. And how should we think about this third claim, which is that hypoxemia is often out of proportion to either mechanical abnormalities or the level of dysphenia?

And I I I'll preface this by saying this is already a little bit difficult because I don't think there's any clear evidence to support or refute this. Well I think there's some evidence to to refute that that presentation is common. I mean certainly if you talk to people who've taken care of a lot of COVID-19 patients, you will hear that said. And I've heard it said, I've heard it said by our own providers. If you look at the actual published data, that does not seem to be the case Right?

I mean as we've just gone over, the compliance abnormalities are not unusually mild. And uh certainly um as we've compiled our own internal data at uh at MGH and um you know we're in the process of writing this up to publish it the gas exchange and the mechanical abnormalities do tend to move in concert. So uh it is a little bit difficult. This is one of these things in medicine that is talked about much more frequently than it's documented Maybe those patients are out there. They don't seem to be common in the published uh in the published series.

But why might there be uh mild abnormalities and severe gas exchange? The two most predominant methods of or mechanisms of hypoxemia in all circumstances, but certainly in COVID-19 and ARDS are shunt and VQ mismatch. In order to have shunt and vq mismatch you need to have some areas of impaired ventilation, or most commonly you need to have some areas of impaired ventilation. Undoubtedly patients with COVID-19 do. They almost always present with bilateral infiltrates on chest x-ray.

So in the the Wuhan ICU series, it's about 80% of patients had bilateral infiltrators on presentation. In the Seattle case series it's a hundred percent. In the New York case series, which includes um both ICU and some non uh non-innubated patients. Certainly in the New York case series, it's again about 80%. So the vast majority of these patients have impaired ventilation.

They're gonna have areas of low VQ and therefore they're going to have hypoxemia. Whether or not the areas of the uh infiltrates are in proportion to the hypoxemia is really difficult to say. We're not doing CT scans in these patients I don't know that it's even really been thought of. Like what does it mean? Has somebody the the kind of studies that we've previously done in ARDS where you go through and you count quadrants affected in a chest x-ray and that sort of thing.

That quantification hasn't been done. So once that's done, then maybe we can talk about whether imaging abnormalities are out of proportion to hypoxemia. My guess is they probably won't based on the large percentage of bilateral infiltrates, but we don't have that kind of quantitative analysis for COVID-19 yet. The uh imaging abnormalities out of proportion to dipsnia is another really interesting thing It's always been the case that dipsnia can be or sensation of shortness of breath can be out of proportion to hype to hypoxemia. Hypoxemia alone is not a very profound stimulus to dipsemia that makes some sort of TL logic sense because you know there's very there's not a whole lot you can do uh to improve your oxygen delivery by uh improve by increasing your minute ventilation because of the shape of the hemoglobin saturation curve.

You can modestly improve your oxygenation by breathing heavier, but it's not a lot. So the things that are a much more profound stimulus to dipsnia are these sort of mechanical sensations. So changes in respiratory compliance and uh that stimulate sort of mechanical efforts in the lung, changes in resistance as in asthma. These things are profound stimuli to dipsnia. So we would expect, in the setting of mild mechanical abnormalities, to have mild dipsnia.

Gotcha. And just to play devil's advocate, if you were to take the other position in arguing that this was not ARDS, is there any evidence you can think about that you'd be able to use to support that point of view? Well let me let me say a couple things. I mean I think that there there is there is some substance to that argument, but I don't think It's exactly the thing that's being articulated. I think what people really mean when they say that is a couple of things.

One, they mean that this isn't Severe, severe ARDS. Right? And and this isn't like uh, you know, that patients on an FIO two of one and they're Plateau pressure is 35 on a 4cc per kilogram tidal volume and you just can't ventilate them and maybe you're thinking about ECMO. This is largely not that. We have some of those.

We've done some ECMO at MGH, and they've done a lot more in other centers such as Paris, but that's not the majority of patients. And so if that's what you think of as A or D S, that sort of really low P to F, then that's correct. This isn't this isn't that In addition to the arguments based on the Berlin criteria that we've just gone through, let me point out that that's not a distinction with a lot of clinical relevance. Anytime you have any degree of infiltrate or heterogeneous inflation or stimulus to inflammation, as far as we know, you're better off using standard ARDS ventilatory approaches, for instance, with low total volume ventilations. As I'm sure you're aware, there's uh evidence of benefit in low tidal volume ventilation even in settings like abdominal surgery.

So whether or not this is severe ARDS probably doesn't impact on the recommended uh treatment course, but I do think what some some of what people are saying is that uh that this is not severe ARDS. The other thing that I think people are saying is that A lot of people have in mind a particular set of ventilatory protocols for ARDS that are both not the the consensus modern approach to ARDS in the ARDS literature and are also inappropriate for COVID-19 patients, which is to say formal recruitment maneuvers, really high peeps These things that we used to do in ARDS, which absolutely are likely to be harmful in these more mild abnormalities such as COVID-19, but also have never been demonstrated to have any benefit in in the pre-COVID-19 ARDS population.

And some of the this is not A or D S argument I think is really saying don't use high peeps in these patients. Don't use the high peep art net table. Don't do recruitment maneuvers. Don't do, you know, all of that I agree with. That's completely true You shouldn't do those things.

I would just add the caveat that you shouldn't have been doing them before COVID-19 either, and you should continue to not do them after COVID-19. Gotcha. And I saw that, you know, I'll also link link this to our show notes, but there's an entire flair on how you should manage PEEP and alveolar recruitment. And I thought I think a lot of what you just mentioned is related to that. And that's reviewing some of the studies in terms of whether or not those have made any differences in terms of mortality.

That's correct. That's correct. So, you know, uh just to to get into it a little bit, uh, in case it's of interest, right, the physiologic rationale for the use of PEEP in ARDS is really, really strong, right? The primary mechanisms of hypoxemia are VQ mismatch and shunt, and the shunt results from alveolar collapse and the VQ mismatch results from impaired alveolar ventilation to a large degree. So opening up those collapsed lung units with positive indexatory pressure, thereby decreasing shunt Seems to make a lot of sense.

It's a very compelling physiologic argument. And we have spent twenty-five years now trying to demonstrate a clinical benefit in clinical trials that uh is uh of equal uh persuasiveness to the physiologic argument we have not done so. Uh you know, there's a number of trials. There's alveoli, there's LOVES , and uh most recently ART. uh EP Vent uh EP Vent one and two I would throw in there, which have attempted to show a benefit to aggressive recruitment and high peep approaches. Most of those trials show no difference between high peep and low peep, and ART shows a harm associated with recruitment maneuvers and high peep.

So it just hasn't been true that those approaches benefit uh patients with ARDS, despite the fact that we associate those approaches. with patients with ARDS. I mean I think ARDS, in addition to being a heterogeneous syndrome, is heterogeneous with respect to the individual patient. Like some areas of the lung are affected and other areas of the lung are not affected. Therefore, if you apply a large uh pressure at the airway open opening to to recruit or open affected lung units, you're also over distending uh unaffected lung units, the so-called baby lung, and therefore potentially causing lung injury there. On balance, those high peep approaches haven't shown themselves to be beneficial.

So I think some of the argument is is coming from a correct place about is this A or D S. There are definitely ventilatory approach or ventilatory pro approaches that people associate with the treatment of ARDS, which should which are inappropriate. There's def that's definitely true. I think um you know Katnoni uh who had a letter in intensive care medicine recent or a letter to uh the Blue Journal, the American Journal of Respiratory and Critical Care Medicine, making this it's not a RDS argument, as well as a um editorial in intensive care medicine making the argument And I think what he's getting at is that these high peep and he said this explicitly, these high peep strategies are inappropriate.

They're definitely inappropriate. Um they just always have been. I'm glad that you mentioned that paper because I think that a lot of what the proponents of claiming that ARDS is a distinct pathophysiological phenomenon from COVID-19 will cite this article because in this article he subdivides it into type one, so that's near normal pulmonary compliance with an isolated viral pneumonia and type two with decreased pulmonary compliance. How should we think about his stratification of this and whether or not there's any evidence to support this subdivision Yeah, I mean, what what can we say about this? Is it true that there are people who have uh a pulmonary infection, a viral pneumonia for instance, that is not yet meeting cr meeting the Berlin criteria?

Yeah, that's true. That's always been true. Um because, you know, I mean you can have an ammonia that's not bilateral, and then you wouldn't meet the uh um the Berlin criteria based on having unilateral infiltrates. The low bar pneumonium would give you hypoxemic respiratory failure without meeting the Berlin criteria. And so you would have hypoxemic respiratory failure, but you would not yet have ARDS.

So yeah, those patients are out there. Gatnoni chooses to call them type L or type 1, depending on which of his articles You're referring to, I'm not sure why we need a new noun, we used to just call that pneumonia, and so I guess we could continue to just call that pneumonia And then his type 2 or type H, again, depending on which article, is is just severe ARDS, dense posterior uh collapse, low compliance. It's just severe ARDS, low P to F. So we could continue to call those uh hypoxemic respiratory failure and ARDS as we always have, or I suppose you could give them new nouns if you are inclined to. I'm not sure that uh that anything has contributed to the analysis by inventing a new term.

And let me let me just say to the extent that there are definitely patients who have hypoxemic prostroit failure who do not meet the Berlin criteria. That's true and also deeply unimportant because you should manage those patients the same. And Gatnoni himself suggests you should manage those patients the same. One of the reasons that you get lung injury is because you have heterogeneous inflation. So you have that that isolated viral pneumonia where that area of the lung has decreased compliance and the rest of the lung has normal compliance.

And there is therefore a junction, and this is uh nicely outlined in Gatnoni's letter, there is therefore a portion of the lung which exists at the the border of a region that has abnormal compliance and a region that has normal compliance. If you apply large transpulmonary pressures to a heterogeneously inflated lung, you get these so-called stress multipliers where you have a high risk of mechanical damage at that junction. And so you want to be really careful. in A, allowing large spontaneous uh spontaneous transpulmonary pressures, which is to say, patients with increased work of breathing and large negative pleural pressures are generating large transpulmonary pressures And as Gatnoni outlines in the intensive care medicine article, you can convert type one to type two by allowing a patient to work hard for too long.

If you wish to stop a patient from working hard and generating those large transpulmonary pressures, that's largely accomplished by intubation, again, as Gatinhooney outlined. So that sequence of events uh can happen. It's always been known to happen. There's an old uh article by OG Gajik. uh which demonstrates that the relative risk of turning non-ARDS hypoxemic respiratory failure into ARDS, which is to say the risk of taking a patient with hypoxemic respiratory failure that does not yet meet Berlin criteria and turning them into one who does meet Berlin criteria. In Gatnoni 's language this would be a type one to type two transition.

OG Gajik described this years ago and the relative risk is about 1.6 per milliliter of tidal volume above six milliliters per kilogram predicted body weight. So absolutely inappropriate ventilatory strategies or um uh increased work of breathing in the case of the spontaneous breathing patient can worsen lung injury. That's always been the case. And so to the extent that patients with isolated um non-bilateral abnormalities exist, they are at high risk for ARDS with inappropriate non-ARDS ventilatory. management. So that distinction is an old one.

It's one that does not have a significant therapeutic consequence. And so it's not one that's been terribly important in my mind and certainly not one that justifies the invention of new terms. So one of the things that's continued to confuse me is if we were to assume that COVID-19 was distinct from ARDS , What kind of treatment strategies are there that are alternatives that might be beneficial or evidence-based in any way? I've kind of scoured through Twitter, I'm probably seeing some of the same things you're seeing. Patients need oxygen, not pressure, COVID-19 's not ARDS.

What is the actual treatment strategy that's being advocated? For the serious players, because that's the more interesting argument to engage with, they're arguing for a relatively subtle difference, and it's not one that I think I disagree with that much. So they're saying start a patient on supplemental oxygen, and this is what Gatnoni outlines in his uh intensive care medicine piece. Start a patient on supplemental oxygen and watch them very carefully If they have increased work of breathing, then they're likely generating large transpulmonary pressures. And to the extent that they have any area of viral pneumonia or impaired ventilation, they're thereby likely to re to injure their lung with those large transpulmonary pressures, and you should innovate that patient and use what he termed gentle ventilation, by which he means low PEEP, no aggressive recruitment.

Low tidal volumes. That to me is a canonical ARDS ventilatory strategy. He's absolutely right about that. You know, one of the old the sort of Not old, but one of the more acute dilemmas in the ARDS field uh and certainly in day-to-day practice prior to COVID-19 was what do you do with the patient? Who may be meeting Berlin criteria or about to meet Berlin criteria with hypoxemic respiratory failure, who has not yet been innovated And you can decide that you're so concerned about large transpulmonary pressures that you want to innovate and put that patient on low total volume ventilation.

But to do so, you end up having to apply a lot of sedation, transfer that patient to the ICU, and and there's a mortality uh associated with those things and a morbidity associated with those things. We know that sedation is associated with poor outcomes in the ICU. And so at what point does innovating that patient, the benefit of innovating that patient and controlling the transpulmonary pressures outweigh the risk of sort of ICU fulmanands and sedating them and moving them to the ICU. That's always been a delicate question, and what most of us have done in those cases is You know, put the patient on supplemental oxygen in the pre-COVID era, sometimes uh we would be fairly likely to use high flow nasal cannula oxygen, and then watch that patient.

And when they start having increased work of breathing, decide that now they're at risk for spontaneous lung injury and innovating them and controlling the transpulmonary pressure. That's the treatment strategy that Katnoni is explicitly arguing for. I think it's absolutely the correct one. I would venture to say There are very few hospitals in this country where people's first therapy when a patient shows up in the emergency room complaining of shortness of breath is an endotracheal tube. Those patients always get put on oxygen first.

And all we're debating about is the fairly subtle distinction of at what point do you become so concerned about that patient's trajectory that you innovate them rather than leave them with an increased work of breathing on supplemental oxygen. So this is a small area of disagreement and I think It's always been the case, and this I'm sure that this has been your experience, if three different uh physicians roll up on a patient who's sitting in a room on, you know, some moderate level of supplemental oxygen and tried to decide where is this patient heading, is it time to pull the trigger on intubation? That's a difficult decision to make Predicting the future is hard. And most of us make it by sort of looking at that patient.

Are they struggling? Are they working hard? Making that decision based on the clinical impression of work of breathing And so I think that's a standard approach to ARDS. Uh and I think that the one that the serious people such as Gat Noni are advocating is very standard. And I take his point completely.

Once you've innovated that patient, you should not be doing these big aggressive recruitment maneuvers, you should not be using high peep tables or putting patients on really high peep, and you should be watching that patient's driving pressure and that patient's plateau pressure All of those things are true. And so I think that the actual when it comes right down to it, at least with respect to the people who are making serious arguments the actual difference in advocated treatment plans is extremely small. And I think that's reflected in the fact that uh most of the published case series basically are describing the same experience. They're describing a high percentage of ICU patients who are innovative. They're describing a relatively short period of time between presentation. and innovation and they're describing um a relatively um similar spectrum of disease because I think when you really boil down to it most reasonable physicians end up with about the same treatment plan.

Gotcha, yeah, I think that makes a lot of sense and again we we mentioned this early earlier, but it's important to contextualize a lot of these findings or these these discussions because irrespective of what you end up calling it, I'm not sure that Currently from what I'm hearing from you, there's a lot of evidence to advocate for an alternative strategy that may be more beneficial if this were called something else. I don't think there's a lot of evidence for anything other than standard lung protective ventilation, even if you were to call it something else, because there is a lot of evidence that standard lung protective ventilation benefits patients outside of the Berlin definition. And so even if you were to decide and most of these patients meet the Berlin definition, so there is probably no better proved fact in all of critical care medicine than that lung protective ventilation benefits patients who meet the Berlin criteria.

And there's pretty good evidence that lung protective ventilation benefits patients outside of the Berlin criteria. So I'm not aware of a serious argument that there's an alternative. uh to lung protective ventilation. Now there is some nonsense out there, right? Like there's things on Dr. Oz 's website or Twitter advocating for sort of extremely different um uh approaches, um even advocating for approaches that have been demonstrated in the past to increase mortality. And so that's a whole separate thing.

Um but those are by and large not being being made by people who I would arguments that are by and large not being made by people who I would characterize as serious. Gotcha. And to reference what you were saying, there there are studies by, I think, Davies and Fu Futier and That's those are the patients who uh benefit regardless of the reason for ventilation with low tidal volume ventilation strategies. That's correct. You know, I mean I think The the real the really only downside to low tidal volume ventilation in almost any patient, certainly any patient with hypoxemic respiratory failure um is is the sedation that comes along with it.

Patients with hypoxemic respiratory failure really want to have a large minute ventilation that's inconsistent with lung protective ventilation. So the downside to it is, you know, is really the propofol and the uh the versed and the things that you and the dilatid that you end up using to sedate them. And that's a real real dilemma and one of the reasons why you it might You might have pause just before innovating everybody and putting them on lung protective ventilation, but the alternatives to innovation and lung protective ventilation in a patient who has an increased work of breathing, I'm not aware of of good ones. There are, you know, people who advocate for vent modes like APRB.

There's even um some stuff going around Twitter advocating for high frequency oscillatory ventilation and things like that, but those are those are approaches for which we actually have trial data that suggests outcomes are poorer when you use those approaches. So I don't think it's possible to responsibly advocate. for those as treatments for COVID-19. I think the takeaway I'm getting is that Twitter's probably not the best place to be going if you're looking for evidence-based recommendations. So it's certainly fun to pick those apart because they're utter nonsense and they're easy to pick apart. All right.

Um so to get back to the uh couple of claims uh so for our listeners, the claims were imaging abnormalities in COVID-19 are mild. Pulmonary mechanical abnormalities are also mild and hypoxemia is out of proportion to either the mechanical abnormalities or to the level of Dyspnea. So to our listeners again, take a second, think about your level of agreement with those claims, and what's the level of evidence to support those claims. Dr. Harden, thank you so much for taking the time to talk to us. I know you're incredibly busy, but I think this will give everybody a uh strong evidence-based approach moving forward in terms of evaluating COVID-19 and its role in the ARDS argument.

Great. Thank you so much for having me. To all our listeners, thanks for joining us on the show today. I'll make sure to include some links to Dr. Hardin's references as well as to Flair. Stay hungry and keep asking.

Questions.