So, hello and welcome to the acute respiratory failure case presentation. My name is Hillary Faust. I'm from the University of Wisconsin, and I am thrilled to be joined today by John Marini, Keith Lamb, and Tim Gerrard. I have no relevant disclosures, and with that, we'll get started. So, for case number one, we have a 55-year-old woman who presents with COVID-19. Her vital signs are remarkable for tachypnea with a respiratory rate of 26, a low-grade temperature, and hypoxia setting 85% on room air. Her chest X-ray is seen at right. And her hypoxia progresses. She's placed on six liters of nasal cannula, but setting only 89% and becomes increasingly tachypneic. So, for our first question, is the pathophysiology of COVID-associated lung injury different from that of ARDS? Well, I can start it off. It may be controversial to say this, but I think it is different. I think most of us have seen, over the course of the COVID experience, people who have relatively good compliance convert into patients with terrible compliance. Some patients present with terrible compliance, and their pathophysiology is driving that evolution and also, you know, accentuating the vascular components. Early on, there's a redirection of blood flow from, I don't know if you want to call it vasodysregulation, microthrombosis, et cetera, and eventually it'll be mainly microthrombosis that is the problem from the vascular side. The alveolar side, of course, is affected as well. The crude measures that we have of underlying pathology at the bedside are not very good, and we don't really use the right ones, to tell you the truth. But hypoxia from the start is disconnected from the mechanics. The hypoxemia may stay relatively stable, whereas the mechanics do evolve and change over time, or at least that's my impression of what's going on. Thank you. Anyone else have any thoughts on that? I guess the only thing I would add is that when you ask the question whether the pathophysiology of COVID-associated lung injury is different from that of traditional ARDS, that immediately leads me to ask the question, should treatment be different? Because regardless of what the pathophysiology is, what we're really trying to decide at the bedside is how do we treat the patient in a way to best improve their chances of recovery? And we know that treatment is different. There are clearly things that we do based on clinical trial evidence for a COVID patient that we don't do for a non-COVID traditional ARDS patient. Is the reason that they respond in the way they do based on clinical trial data because the pathophysiology is different? Probably so. But my point is just, I always ask, what should I be doing for this patient in this moment? And I think at this point, it's very clear that there are specific treatments that COVID patients would benefit from that we don't often do for non-COVID ARDS patients. So that led perfectly into my next question. I think one question I have based on that to follow up is, are we doing things differently in COVID patients that we should not be doing in ARDS patients? Or do you think some of the progress that we've made in COVID will be applied to management of ARDS patients? I agree completely with Dr. Girard that we have to consider the principles that are absolutely rock solid established and help us guide through. But the different pathology suggests that, you know, interventions such as high PEEP and recruitment maneuvers and so on may not be as successful or actually counterproductive. And we need to, I think if what we've learned is to down-regulate our reliance on PEEP and go earlier to prone positioning as a way to improve oxygenation and so on. Which operates in somewhat different ways in the prone position in COVID than it does with conventional ARDS. There's a vascular component there too. I agree with John. You know, no matter what the ideology of the ARDS or the severe respiratory failure is, I think that we have come a long way in terms of customizing or dosing the ventilator to the patient in a way that is most protected to that patient. So, you know, as an example, you know, you intubate somebody for hypoxemic respiratory failure, they either, if you believe in the two phenotype ideology of COVID ARDS, one has relatively normal compliance and one doesn't. But if you have one in one phenotype and one in the other, you may start both those patients on what you think is protective ventilation, i.e. 6 milliliters per kilogram of body weight. But you may then re-dose the ventilator to the patient based on the things that we have literature to support other than just the volume and that might be plateau pressure, driving pressures, you know, the potential for the patient to hurt themselves, self-induce lung injury, in addition to the primary insult. So I think that the most important thing in my mind, as I understand the literature today, and like Tim said, I'm always waiting for the next thing to come out because I'm not always convinced that we know exactly what to do, you know, that most of these questions haven't really been answered, I don't think. But you take that patient and you do what you think is right based on the literature that we have available now, and then you dose the ventilator to the patient. Those are all great responses, and I think, you know, being able to personalize management may be one of the best things that comes out of COVID. Just one note on some of the differences that are emerging between COVID versus non-COVID ARDS. I think a lot of us have appreciated that the duration of mechanical ventilation and the dynamics of the illness seem to be significantly prolonged in COVID versus ARDS from other etiologies. And this one cohort study showed a duration of mechanical ventilation of 13.5 days versus approximately 7.5. Interestingly, you know, I think that there's a big spectrum in COVID-19, and we've always appreciated there's heterogeneity within ARDS. But as you can see from some of these physiologic parameters, there are no overall differences in P to F ratio, but there's a big spectrum, and we do see some of these patients who have a very high P to F ratio. I was somewhat surprised to see this study in which minute ventilation was actually lower in COVID-19, since we do seem to see this high dead space phenotype. And compliance may trend slightly larger to be slightly higher, but was not overall different. There are differences in inflammatory markers and kinetics of inflammatory markers, with IL-6 being elevated more in the second week of ARDS compared with the first, overall different versus ARDS. So, let's go back to our patient and say that due to hypoxia and high work of breathing, she started on BiPAP, and her tidal volumes are ranging from 595 to 720 cc, and this is a 5-foot, 2-inch tall lady. So how important is targeting lung protective tidal volume during noninvasive ventilation, and how do you optimize noninvasive positive pressure to achieve this? You know, this is something that's discussed amongst ICU staff all the time, and I don't think that there's a clear answer other than, in my mind, yes, we should be concerned about tidal volumes and noninvasive ventilation because I don't have a good argument to not worry about them just because they're being noninvasively ventilated. So how you approach it, you know, the only way really, we've all taken care of patients on noninvasive ventilation, and you want the tidal volume to be smaller in terms of adjusting the machine, then you have to reduce the support, and is that the, you know, is that the right answer for somebody who's sick and you're trying to not have to escalate and intubate? So I think, yes, we should be worried about it. I think that we should look at these patients via risk and benefit, just like we would anybody else. Is there a risk in sedating somebody that's on noninvasive ventilation in order to slow their tachypnea or to reduce their tidal volumes? Yes, there is, and what is the benefit? So I don't think there's a clear answer. I'd be interested to see what others have to say. I'd like to point out that, you know, this idea of PCILI, patient self-inflicted lung injury, it certainly makes sense physiologically, and I think we've all seen patients rapidly deteriorate who are laboring to breathe. If the patient has a comfortably large tidal volume, in other words, they seem to be okay with that, then, of course, we don't want to give them more than that. But, you know, we may not need to aggressively attack what they're choosing and to bring it down, because some of these people will have, I hate the term, but happy hypoxemia and be doing quite well and actually following the trajectory is more important than anything else. I mean, if the patient is beginning to labor more and more, if they're deteriorating in whatever objective way you can delineate, then, you know, it may be time to scale back on all kinds of support that the patient is taking spontaneously or even intubate them at that point. I think that's been debated in the literature pretty aggressively on both sides. Yeah, those are great answers. Oh, go ahead. Well, I guess the one other thing, I don't want to get sort of too philosophical here, but one thing I'm wondering as we think about this question and we see the waveforms there and the numbers is would we be wondering the same thing if this were a patient that we didn't have these measurements for? You know, what if they were on high flow and that was enough support to give them adequate oxygenation and we didn't have a measure of their tidal volume? Would we, in that case, be wondering if we should proactively do something to reduce or limit their tidal volumes? I think we probably wouldn't. And so I worry a little bit. I do agree that in principle, we should be worried about PCLE and we should be worried about high tidal volumes in a patient who's non-invasively ventilated. But in practice, I don't know what to do about it and I don't have an evidence-based pathway to address it. And I'm very reluctant to think, well, I should just move directly to intubation and basic mechanical ventilation, neuromuscular blockade. I mean, I would never consider that in someone who was tolerating high flow nasal cannula oxygen delivery. So why would I consider that in someone who is tolerating non-invasive? It seems like the only difference between those two patients is that we actually have a measurement telling us their tidal volumes are higher than we would like them to be. I agree 100%. And of course, it's not tidal volume. We've learned this over the years. You know, it's not tidal volume, but the compartment into which the tidal volume is being cramped. If it's a fairly expansive tidal volume, I mean, container, then a given tidal volume is not going to generate the stresses that concern us about the possibility of PCOE. So, you know, looking at the patient tells you a lot. They're, you know, they're laboring to breathe, et cetera. There's a whole literature that we've basically forgotten about that, you know, those of us who are quite old do remember looking at the ratio of sustainable pressure versus maximal pressure that can be generated. And that's in the range of 0.15, something like that. So when you get, you're approaching 0.10 or 0.15, in other words, 10 or 15% of the maximum pressures that you can generate, then the patient begins to seriously labor. So we don't have esophageal balloons in place in most of these people or any other indicator like that. And it's really the stresses, the transpulmonary pressures that we're worried about, and the patient is possibly your best indicator of that, whether they're awake or not. If they're awake and comfortable, great. If they're deeply sedated or relatively sedated and can't answer you, then you have to look at the respiratory muscles and activities and so on and see if there's paradoxical breathing and that kind of thing. But we just don't have all the information yet. Those are great answers. One relevant study would be the CARTO study from Critical Care Medicine in 2016, in which patients with an over-median tidal volume had a higher rate of non-invasive ventilation failure compared with those with below-median. But this was really a different scene in patients with moderate to severe hypoxia, with P to F ratios less than 200. And so I think in addition to thinking about how the patient looks, it's important to consider the population and potential modification by other underlying factors, like potentially how severe the hypoxemia is. All right. So saturations dropped to 88% on 100% FiO2 on BiPAP. So would you recommend awake-prone positioning, and how does this affect hemodynamics and gas exchange? Yes. Yeah. I mean, I think that three years ago, I might have said probably not, but I think that clearly early-prone positioning has declared itself as a viable, awake-prone positioning has declared itself as a viable intervention, and I would absolutely recommend that. I think this may be a reflection of sort of those longer kinetics of illness that we see in COVID compared with non-COVID ARDS. I think in addition to having longer times on the ventilator, they may have a longer sort of pre-mechanical ventilation course, and so there may be more time to intervene with awake-prone positioning, whereas in a patient with ARDS, they don't have those days and days that they spend on high flow or non-invasive positive pressure. And I come back to a point that I made earlier, and that is that there is more of a disconnection between what's happening with blood flow and hypoxemia and mechanics. Now, PEEP is not your friend when the patient has a hypoxemic event or is trending the wrong way. Proning is for a couple of reasons. Of course, it redistributes the transpulmonary pressures more evenly, but studies that David Ciumello in Milan has shown very convincingly from a large population there is that the perfusion is different in the prone position, and that is helping you. What I'm trying to say is that the connection between prone positioning, benefit, and recruitment is not as strong as it is in early stage ARDS. So I think that prone positioning is a better option than raising mean airway pressure and risking the adverse physiology we've talked about already. Just to point to one recent study looking at awake prone positioning for COVID-19, the physiologic responses included an initial increase in S to F ratio, but then sort of a trend back down as equilibrium was achieved. But there was a decrease and a sustained decrease in respiratory rate, and patients with awake prone positioning had lower rates of treatment failure and intubation. Now, this was all strongly associated with duration of awake prone positioning like we know from the literature in invasive mechanical ventilation. All right. So our patient tolerates awake prone positioning well, but remains very hypoxic and significantly tachypneic. And so the question comes up of deciding timing of intubation. So how do you all decide timing of intubation, and do you use clinical scores or other sort of metrics to make your decisions? It's not that dissimilar from other patients, non-COVID patients that you sort of have in the back of your mind that delay of definitive treatment is bad, but you're never really sure that that's what you're doing, you know? And so, yeah, it's not straightforward to me. I'm not smart enough to figure, I just, somebody says, okay, we're going to intubate this person and I go get the stuff and we intubate it. Hopefully you guys are smarter than I am, but I think we would all agree that trending is, you know, of things we're confident about is the best thing to follow. Some patients can look pretty good and have oxygenation values that look very precarious and they're still comfortable and maybe they should not be intubated, but those who are vigorously breathing, and particularly if it's more vigorous than it was, maybe that's the time to intervene. Serious delay is not a good idea. Yeah. Yeah. And the literature in COVID is supporting that just like it does in non-COVID ARDS. Well, believe it or not, we're already at 20 minutes. And so that's the amount of time that we have. I just want to thank all three of you so much for joining us and lending your expertise. And hopefully everyone who's listening enjoyed and learned this as much as I did. Thanks. Thank you so much for leading.

acute respiratory failure

COVID-19

ARDS

treatment strategies

noninvasive ventilation

intubation