Transcript
Speaker: Welcome to Critical Matters, a sound critical care podcast covering a broad range of topics related to the practice of intensive care medicine.
Speaker: Sound Critical Care provides comprehensive critical care programs to hospitals across the country.
Speaker: To learn more about our programs and career opportunities, visit www.soundphysicians.com.
Speaker: And now your host, Dr. Sergio Zanotti.
Speaker: In today's episode of the podcast, we continue to discuss issues related to the management of COVID-19 respiratory distress.
Speaker: Our guest is Dr. John Marini.
Speaker: Dr. Marini is Director of Academic Programs in Research and Education for the Department of Medicine of the Regions Hospital in St.
Speaker: Paul, Minnesota.
Speaker: Dr. Marini is a pulmonary and critical care physician, a master educator, and a prolific researcher with hundreds of peer-reviewed publications.
Speaker: His research interests focus on cardiopulmonary physiology, management of acute respiratory failure,
Speaker: with an emphasis on patient ventilator interactions occurring in the clinical setting.
Speaker: In most of his research, he has positioned himself at the interface between basic physiology and clinical medicine to develop insights into advancing clinical practice.
Speaker: John, welcome to Critical Matters.
Speaker: Thank you, Sergio.
Speaker: So a real pleasure to have you today and talk about this fascinating disease that we're dealing with, obviously very interesting times with COVID-19.
Speaker: And I thought that may be a good
Speaker: place to start would be at a very high level with a Nietzsche quote that says, there are no facts, only interpretations.
Speaker: And talk a little bit about evidence-based medicine during COVID-19 and getting your thoughts in terms of what that really means and how we apply it at the bedside.
Speaker: What I have noticed, John, looking at what's going on throughout our country is that two very extreme positions, which seems to be the common fabric
Speaker: of our conversations in all arenas today emerge.
Speaker: On one hand, I see people who have no regard for evidence-based and are trying all sorts of experimental therapies outside of the context of experimental trials.
Speaker: And on the other hand, these EBM or evidence-based medicine cellists who basically say that this is what we should do for every single patient.
Speaker: And I think that we all wanna use the best of all evidence, but the truth kind of lies somewhere in between of how we interpret evidence-based medicine.
Speaker: So I'm curious to hear how you think about this.
Speaker: Sergio, as you might understand, I think that the idea of basing our practice on evidence is uncontestable.
Speaker: We should be basing it on evidence, but the evidence ranges from the laboratory through personal experience to population-based data.
Speaker: To have randomized clinical trials, you have to have a well-defined problem.
Speaker: You have to have alternatives that mean something and selected doses of therapies, if you're testing the therapy, that are perfectly applicable to the patient population you're studying.
Speaker: Unfortunately, and I have put this in paper, and I've said it many times in conferences,
Speaker: What we're doing is we're taking population-based information and trying to apply it to all of our patients.
Speaker: And that simply does not work, especially when we're dealing with something like COVID-19, which is a brand new confrontation to our practice.
Speaker: We don't know very much about it.
Speaker: And the evidence that we have is not to be denied.
Speaker: It's from the laboratory.
Speaker: The only thing that can be tested
Speaker: regarding ventilator-induced lung injury is in the laboratory, is in the animals that unfortunately we have to sacrifice, but who give us information that we can use that's ironclad.
Speaker: Even our clinical studies that have looked at ventilator strategies make the leap between ventilator strategy and mortality through ventilator-induced lung injury.
Speaker: We don't know that.
Speaker: In fact, some people have contested that that is the explanation.
Speaker: So I'm not saying it isn't, but I am saying that we don't know those things.
Speaker: And yet, people become very, very, shall I say, committed and passionate about applying only what's been published in papers, which itself
Speaker: is time stamped.
Speaker: We're doing things differently today than we did five years ago.
Speaker: And I think that's to be reckoned with.
Speaker: Absolutely.
Speaker: And I think that, like you said, I mean, at the end of the day, we're all trying to do what's best for the individual patient in front of us.
Speaker: And that means taking the best available evidence and integrating that into our clinical observations and our understanding of physiology, which is dynamic, right?
Speaker: I mean, what I did, especially in the ICU,
Speaker: what I did this morning might be very different this afternoon.
Speaker: But I did want to, before we dive into COVID-19 specifically, talk a little bit about what you, very broad strokes, John, what you would say is the best evidence we have right now regarding the management of ARDS, regarding avoiding further lung injury with a ventilator, and what most people call as lung protective ventilation.
Speaker: I know that you have
Speaker: written and thought and researched this for a whole career.
Speaker: And I will link a lot of your recent papers to the show notes so people can take a deeper dive.
Speaker: But just if you could give us like a kind of a ARDS, Current Evidence 101, in terms of how you look at this from a very broad perspective.
Speaker: Boy, that's a very- It's a tough one.
Speaker: That's a large assignment.
Speaker: Well, first of all, not
Speaker: all patients are equally vulnerable to what we do with the ventilator.
Speaker: For example, healthy individuals have strong lungs and are unlikely to injure them with any tidal volume that we apply or any PEEP that we use.
Speaker: For example, if you look at an exercising athlete, they are pouring tremendous amounts of power through their lungs with an intensity that
Speaker: is never seen in the intensive care unit.
Speaker: And yet these people do not injure their lungs significantly, if at all.
Speaker: So the predisposition is very important to begin with.
Speaker: Secondly, some patients will have vulnerabilities that relate not to what we do on the airspace side, but on the blood side.
Speaker: And COVID-19 is one of those conditions, I think, that
Speaker: exemplifies that.
Speaker: And the major concern here is that we don't injure the healthy lung units of the ARDS lung.
Speaker: Now, whether or not you believe the baby lung concept that says that the air spaces that are functional are normally compliant and we need to
Speaker: we need to understand that the lung is not stiff but small.
Speaker: I happen to believe that's mostly true.
Speaker: With this small baby lung, we are applying an energy to each cycle that has the potential of overstraining the tissue and inducing injury.
Speaker: If we continue to do that at a rapid clip, faster than the
Speaker: lung can adapt to it, then we will eventually start breaking lung units, reducing the size of the baby lung, and concentrating the power.
Speaker: Why do we concentrate the power?
Speaker: We concentrate it because we're asking it to do the ventilation work with a small lung that a healthy person does with two large lungs.
Speaker: So the power applied per unit at the microscopic level is very high.
Speaker: And we continue to break that lung unit, and the others are loaded with the energy and force and power to an increased degree.
Speaker: And so they become even more susceptible.
Speaker: And I've talked about this in terms of the shrinking baby lung.
Speaker: The baby lung is not a static situation.
Speaker: that's going to progressively get better, but it often gets worse with our treatments, even if we don't change the treatment.
Speaker: So what I'm trying to say is that ventilator-induced lung injury, in the expanded, newer view, pays attention not only to excessive strain per cycle, which in a rough sense,
Speaker: correlates with the pressures that we apply, but also to the minute ventilation that's required and the size and capacity of the baby lung to which it's applied.
Speaker: That's what we have to understand as clinicians.
Speaker: Not that we use X tidal volume or limit the pressure to this.
Speaker: Limiting the pressure is an important piece of it, but it's not the entire story.
Speaker: And when you think at the microscopic level and you think of what you're actually dealing with, then you have a lot of things to consider.
Speaker: And the main principles are to minimize the demand for ventilation and blood flow through that lung by reducing the metabolic requirements, the vigor of breathing, the transpulmonary forces, and distributing forces in a more...
Speaker: equal way by prone positioning the most severely ill patients and using everything that we have to reduce the power that's required.
Speaker: Obviously, we have to give power that's required to ventilate and oxygenate, yes.
Speaker: We have to use the PEEP that's necessary to open the baby lung to its full capacity, but not overstretch it.
Speaker: If we do those things, we have our best chance and we're not going to save everybody.
Speaker: Absolutely.
Speaker: And I think that another area that comes to mind as an analogy that I want you to comment a little bit more, John, on is hemodynamic support, something that we still haven't figured out.
Speaker: But we have learned that when we're trying to evaluate the hemodynamic status of a patient, what we used to rely on were very static measurements.
Speaker: and now we talk about dynamic measurements.
Speaker: And I think that my sense is that as we talk about a ventilator-induced lung injury and the shrinking baby lung, that by itself is a very dynamic process, but also that we have hanged our hat, or many have hanged their hat, on static measurements and that that might not tell the whole story.
Speaker: Could you comment further on that?
Speaker: Yes, I think I can.
Speaker: There are several aspects to the question, Sergio.
Speaker: Going back 20 years, John Hotchkiss, Ellen Brocard, Avinahoum, myself in our laboratories found that the vascular gradient, the precapillary to postcapillary gradient that we could control was extremely influential as to the manifestations of Vili
Speaker: in terms of rupture of the delicate membrane, et cetera, and the rate of weight gain in our experimental animals.
Speaker: And we became very convinced that it's the gradient, precapillary to postcapillary pressure, that is correlating with flow through the alveolus that
Speaker: that accentuates the expression of ventilator-induced lung injury.
Speaker: We published a number of papers back then.
Speaker: We showed splits in the alveolar capillary membrane, actual physical ruptures, not only in animals, of course, but in a patient that we published in Critical Care Medicine.
Speaker: This was put on the shelf, basically.
Speaker: Many people didn't pay much attention to it.
Speaker: But the implications are that anything that increases the gradient of vascular pressure across the alveolus, which includes the shrinking baby lung, so you have fewer channels for which the blood can flow, and the height of the cardiac output demand, which is linked to oxygen consumption and ventilatory demand.
Speaker: So it's all sort of tied together.
Speaker: You've got the ventilatory demand and power
Speaker: and duration, that's critical.
Speaker: And on the vascular side, you have the size of the baby lung, meaning that there are fewer capillary channels through which the blood must flow, and the gradient of pressure is increased by any increase in cardiac output.
Speaker: That gives us an implication for what we need to do at the bedside, reduce ventilator demand,
Speaker: ventilatory demand reduce the need for cardiac output because it's a two-hinged problem.
Speaker: I don't know if that's all clear or not, Sergio, but there are stresses on the vessels.
Speaker: There's also edema that forms through permeable vessels as we increase the mean vascular pressures in the lung.
Speaker: And again, we can do that by
Speaker: elevating the gradient of vascular pressure with the higher cardiac output.
Speaker: No, and I think that is very clear.
Speaker: And I think, like you said, it's an evolution of our understanding.
Speaker: And I think that a lot of clinicians at the bedside really look at, okay, my tidal volume is 6 mLs per kg.
Speaker: My plateau pressure is below 30.
Speaker: And I have X amount of PEEP, and now my oxygen is, my PO2 is above 60.
Speaker: I'm good.
Speaker: But like you mentioned, there's a lot of other considerations and those static measurements of that time stamp measurement that play into this as principles we should be very aware of and thinking of as our patients evolve.
Speaker: One of them is, like you said, the evolution of the patient itself, that baby lung might be shrinking, which means that the power we are producing on that lung with those same settings might be changing, but also
Speaker: I think that the other part is that there's other ways of thinking of measuring that's more dynamic that would also be valuable.
Speaker: Do you find that driving pressure is more of a static or a little bit more dynamic measurement, or you would put them in the same category?
Speaker: Well, again, I've said this many times, I suppose, that what is driving pressure, how is it calculated?
Speaker: It's calculated as
Speaker: Plateau pressure minus total PEEP.
Speaker: Those are both static values.
Speaker: It's how you get there and how frequently you get there and how much correlated that is with the static and dynamic strains that are occurring.
Speaker: The faster you reach a certain driving pressure, the more likely it is to put excessive strain on certain elements that are particularly vulnerable
Speaker: within the heterogeneous lung.
Speaker: So by the way, and this is quite an aside, I am very interested to know whether these people who are reporting low compliance in COVID lungs are actually measuring total PEEP.
Speaker: Because at high minute ventilations, you almost invariably have an auto PEEP component stacked onto your set PEEP.
Speaker: And some of the papers I've looked at are electronic medical record-based, meaning that the therapist is likely to have put in the set PEEP, not the measured total PEEP.
Speaker: And when you do that, you assume that the driving pressure is actually much higher than it really is.
Speaker: So that's just an aside, but I think it's an open question at the moment.
Speaker: And I think it's an important one, right?
Speaker: Because I think that if we measure things, we measure them in ways that are not fully accounted for, we are talking about different situations.
Speaker: And I think that are we measuring compliance or not is an important question because people are arguing about this a lot still, which I think that's a great point.
Speaker: But in terms of moving forward and diving a little bit deeper,
Speaker: And I think that before we go into COVID, actually, John, two things that you have mentioned that I think are worth reiterating and reemphasizing, because I think it's not something that most clinicians at the bedside incorporate into their thought process of avoiding better than the lung injury.
Speaker: And that is, it's not only the power of one breath cycle that you are delivering to the lung and how that's distributed along the baby lung, but also the impact
Speaker: that flow and frequency have on that situation, which I think is something that most people, when they talk about lung protective ventilation, don't really even mention.
Speaker: Could you tell us a little bit more about flow and the frequency and how that impacts the overall power and the potential damage?
Speaker: Sure.
Speaker: The lung is not a solid structure.
Speaker: It is a flexible structure that has what are called viscoelastic properties.
Speaker: And the faster you pull one area that can move away from another one that moves less well or is static, the greater is the stress focusing at their junction.
Speaker: And all of these junctions are very vulnerable.
Speaker: So the idea of slowing frequency
Speaker: Probably is more complicated than I'm making out here, but think of it as an accommodation, a stress distribution problem, where the less time you give for adaptation, the more likely it is that you're going to develop an excessive strain.
Speaker: Now, when you take a heterogeneous lung and start breaking
Speaker: the most vulnerable pieces, as I mentioned before, you load the ones that are still left.
Speaker: The ones that are still left are now at a higher risk for breaking.
Speaker: They're stronger than the one that just broke, but not perfectly rigid themselves.
Speaker: And so you get a second cycle, a third cycle in many cycles,
Speaker: that break off the vulnerable pieces and concentrate more power on the ones that are left.
Speaker: So that's the short answer here.
Speaker: The other piece that we haven't talked about is duration.
Speaker: I mean, you can talk about power, which is energy per minute, but how long do you persist with that?
Speaker: And in fact, with COVID-19,
Speaker: people are talking about very slow resolution of the viral pneumonia and what we call ARDS.
Speaker: And therefore, there's even more likelihood that such principles apply.
Speaker: Now, I'll stop after this.
Speaker: But remember that driving pressure in almost all ICUs is measured from an airway pressure, which
Speaker: distends the lung and the chest wall under passive conditions.
Speaker: So that a driving, when people start to talk crazily about, I say it's crazy, maybe they don't, you know, 21 is a safe driving pressure.
Speaker: Well, it might be.
Speaker: But if you're a very obese individual, for example, you have collapsed areas in juxtaposition to open areas and have a lot of
Speaker: of those marginal units to be overstrained by virtually any driving pressure that's applied.
Speaker: And a driving pressure of 21 in an obese individual may correspond to a driving pressure of 10 in one with a flexible chest wall.
Speaker: And we've studied this.
Speaker: We have papers and press that show that massively obese patients
Speaker: have very high driving pressures based on airway pressures alone.
Speaker: And in terms of another question with driving pressure before we go into COVID, if you are looking at a driving pressure and you're trying to get it to what's considered maybe a safe range, and a lot of people, like you said, I mean, talk about 21, other people talk about 15 or below, but again, it depends on the individual patient.
Speaker: but it makes a lot more sense from my perspective.
Speaker: And I wanted to hear what your thoughts are is to use your title volume to titrate that as opposed to people playing with the PEEP and other settings.
Speaker: Is that the way you think about it, John?
Speaker: Well, the missing piece here, Sergio, is the compliance.
Speaker: So if,
Speaker: If the compliance stays static, yes, the tidal volume does correlate with driving pressure.
Speaker: But if the compliance is variable because, for example, you're collapsing lung tissue or because of the PEEP that you apply is overstraining the lung, then your assumptions may be invalidated.
Speaker: I think it's fair to say that if you can use a lower driving pressure and achieve your clinical objectives, that's a good thing in any individual.
Speaker: Now, I said 21 before is safe.
Speaker: No, I didn't mean 21 is safe.
Speaker: I was actually thinking of plateau pressure, by the way.
Speaker: And most people think 15, as you've mentioned, is the upper limit.
Speaker: And they've even encoded this into protocols.
Speaker: That's fine as a guideline.
Speaker: But for some patients, that's not going to be correct.
Speaker: And certainly not if they're simultaneously trying to trigger the ventilator.
Speaker: Absolutely.
Speaker: So let's move into a little bit more about COVID-19 specifically.
Speaker: And then what I really want to focus on is how you would take these principles, what we know about COVID, and apply it at the bedside.
Speaker: And I know that there's a term that you coined that I really
Speaker: find very useful as a framework, which is avoiding the Vili vortex.
Speaker: And I think it's very applicable to how we think about these COVID patients, especially since over a short period of time, people have kind of moved from one extreme to the other in how they're managing these patients.
Speaker: But in terms of COVID-19 associated respiratory illness, John,
Speaker: when this first came out, it was all about ventilators and all about being ARDS.
Speaker: Then people say, well, maybe it's not ARDS.
Speaker: And all sorts of theories or propositions came out.
Speaker: Some of them, I think, based on solid observations and physiology.
Speaker: Others just based, I think, on crazy talk, like this is high altitude pulmonary edema and things along those lines.
Speaker: But what's similar about COVID-19
Speaker: with what we usually see in ARDS and what's different?
Speaker: How do you see where we are today?
Speaker: I think in many ways they are very similar, but it depends on when you are observing the patient.
Speaker: The big difference with COVID-19, pathologically, it's being attacked from the vascular side to a degree which we don't usually see.
Speaker: And the other thing is that an expression of that is severe hypoxemia that does not correlate with the amount of gas in the lung.
Speaker: The amount of gas in the lung initially is quite good.
Speaker: The patients do not have distress necessarily.
Speaker: It's not ARDS.
Speaker: They don't have distress.
Speaker: They're aware of distress.
Speaker: labored breathing, but not difficult breathing.
Speaker: And I make that distinction as I make the correlation with an exercising patient or exercising individual.
Speaker: At the end of a vigorous exercise, we breathe in a labored fashion, but each breath is reasonably easy to pull and we don't have a lot of dyspnea.
Speaker: Later on, we have in COVID-19, we have a much more typical ARDS picture.
Speaker: This is what Luciano has called H. And to be truthful, I think we both understand that the L phenotype and the H phenotype are in the ICU.
Speaker: You're unlikely to see the L phenotype because that's an early phase problem.
Speaker: When the patient does not look like they're in distress, does not look like they're in need of intubation and a mechanical ventilator, and can respond to nasal prongs, non-invasive ventilation, high-flow nasal cannula, et cetera, and a wake-prone positioning without intubation.
Speaker: Those things all seem to work reasonably well.
Speaker: And oxygenation is not often a major, major problem.
Speaker: It doesn't mean that the lungs can't be expanded and recruitable by PEEP.
Speaker: There always will be some recruitable lung units
Speaker: As Luciano and I and colleagues showed many years ago, if you look at the pressure volume curve, even of a normal individual, you have simultaneous overdistension and recruitment going all the way to total lung capacity.
Speaker: In a normal individual, there's not much potential for recruitment.
Speaker: In the acute phase of pulmonary edema, you have a lot of potential for recruitment, but in both cases,
Speaker: you're always, when you apply PEEP, over-distending and recruiting.
Speaker: So I think you must understand that by the time the patients are intubated and come to the intensive care unit, they are more resembling our typical ARDS patient, whatever that is.
Speaker: It's a huge umbrella term.
Speaker: in the sense that they do have infiltrates that look a lot like we usually see.
Speaker: They do have low compliance.
Speaker: Part of that is self-induced lung injury from laboring to breathe under the influence of hypoxemia and advanced viral illness.
Speaker: And some of it may be our own doing with non-invasive ventilation and CPAP levels that are higher than they really should be.
Speaker: So I guess what I'm saying, Sergio, is that the principles of treating ARDS in the later stages are not very different from the usual ARDS.
Speaker: But in the early phase, when you have a chance to interrupt that cycle and prevent the shrinking baby lung, you must take measures to do that, which may include early intubation in a patient who's showing a downward trend.
Speaker: of labored breathing and even fatigue, then that's a signal to go ahead, intervene, take control because of the self-induced lung injury.
Speaker: I don't like to call it pee silly.
Speaker: I think that's a silly term.
Speaker: Really, I think that could have been.
Speaker: It's memorable, but...
Speaker: P for patient, it's redundant to say patient self-induced.
Speaker: Yeah, it's self-induced, self-induced lung injury.
Speaker: And these patients are often pulling fairly high minute ventilations, as we've said before.
Speaker: And I think that there's a lot here to dissect, John, but I wanted to separate it first.
Speaker: And I think that the idea of this L phenotype and H phenotype is a framework in its extremes.
Speaker: But the reality is whenever we have frameworks and extremes, it's the middle messy that's complicated, right?
Speaker: And I think that's sometimes what people believe that it's either one or the other, and it's more likely a progression.
Speaker: And like you said, the series that have been published so far of ICU patients, like the one out of Boston, really are probably patients that fit what we consider to be this H phenotype.
Speaker: But I do believe, and maybe you can correct me and reinforce this, is that many clinicians, and myself included,
Speaker: early on especially when people were intubated very quickly, would see behavior from a pulmonary dynamic that really fits that L phenotype and CT findings that really fit that L phenotype.
Speaker: And my understanding is that the conception of this L phenotype is based on reported cases with measurements, with CT scans that people around the world have seen.
Speaker: Is that the evidence that we have?
Speaker: I mean, obviously it's not a randomized trial, but it's evidence based on what people have documented, correct?
Speaker: Absolutely.
Speaker: I think it was suggested by a recent letter that Gatnone's description of L and H was based on a small experience.
Speaker: Hardly.
Speaker: Luciano, who is a close friend and colleague, said,
Speaker: is a lightning rod for all the information coming in from Lombardy and Northern Italy because he spent almost of his life in Milan and also has connections throughout the scientific and clinical communities in Italy.
Speaker: They have thousands of cases, thousands of cases, and they take CTs routinely on everyone, not something that every institution does.
Speaker: They saw
Speaker: early on, the ground glass infiltrates scattered, not very much, actually, in the periphery, with patients who had profound hypoxemia.
Speaker: And that led us to the idea that vasoplegia was a big part of the initial presentation.
Speaker: Vasoplegia meaning that the
Speaker: pulmonary blood vessels could not regulate their blood flow to match the ventilation.
Speaker: And so we had a lot of functional shunt as well as some true shunt going through injured areas.
Speaker: And furthermore, I think that we keep learning more and more about the endovascular or the endothelial dysfunction in this disease in COVID-19.
Speaker: which is not, again, unique to COVID-19.
Speaker: Other infections have also shown this, but more and more publications in terms of macrothrombosis, microthrombosis on biopsies and autopsies is emerging that really, I mean, supports more evidence that there's something going on at a vascular level, correct?
Speaker: Absolutely.
Speaker: I would emphasize that from measurements of
Speaker: D-dimer, for example, they tend to escalate as the disease advances.
Speaker: The virus is attacking all the blood vessels.
Speaker: It's causing endothelitis, if you want to call it that, endothelium inflammation.
Speaker: That probably is the trigger or the main trigger for in-sitra thrombosis.
Speaker: That happens more late, we think, than early.
Speaker: And of course, most of the
Speaker: anatomic information we have is from autopsy patients who succumb late in their disease.
Speaker: Early on, it's a vascular problem, but it's more vasoplegia, as we said in our JAMA article.
Speaker: And I think that as we dive now into the management portion, John, I think it's important also to recognize that early on in the pandemic, especially here in the United States, I think there was a lot of emphasis on early intubation, which I still think is the right recommendation.
Speaker: But I think what people misinterpreted at the bedside was early intubation means you don't delay intubation for those who need it.
Speaker: It doesn't mean anybody who's hypoxic on a little bit of oxygen needs to be intubated.
Speaker: And I think that that has been a change where everybody was being intubated, they were above six liters per nasal cannula, to a swing in the pendulum, they're saying we should avoid intubation.
Speaker: And I think at the end of the day, what we really want to do is support the patients
Speaker: avoid as much as we can damage, and intubate them early when they need intubation, which I think is part of that, trying to break the cycle of the Vili vortex.
Speaker: So maybe we can start talking about management specifically for COVID-19 by maybe sharing with us what you mean by the Vili vortex.
Speaker: Okay.
Speaker: We've been talking about it indirectly anyway.
Speaker: There are, in routine ARDS,
Speaker: there is the propensity for the most severe patients to enter a progressive cycle of injury to the baby lung units that are functional.
Speaker: A vortex, of course, is a whirlpool.
Speaker: It's a progressive contraction of the baby lung in this context.
Speaker: And that results from what we've talked about
Speaker: before.
Speaker: The structural elements of the lung are in the matrix that is between alveoli.
Speaker: And that matrix has fibrils that will break and load the rest of them with increased levels of power and more damage is likely to occur, which will shrink the size of the lung, which will intensify the power from the airspace side and
Speaker: you have a vortex.
Speaker: That's in the severe cases.
Speaker: In most cases, you don't go that far.
Speaker: In COVID-19, you have another vortex, if you want to call it that, or contributed to the vortex, which is the progressive influence of the viral infection, the viral pneumonia, for which we have no treatment.
Speaker: More importantly, in this context, in the mechanical context, from the vascular side, there is a vortex.
Speaker: We've mentioned that as the lung shrinks, the velocity of blood flowing through that lung increases.
Speaker: The local velocity through the remaining baby lung increases.
Speaker: And the power, the energy,
Speaker: Now thinking about the vascular side and the transpulmonary vascular pressure elevates.
Speaker: We don't have much data, but there is some that indicates that pulmonary hypertension occurs late or later.
Speaker: So you have something working from the vascular side as well.
Speaker: As the baby lung shrinks, the power that's applied from the vascular side increases.
Speaker: The velocities increase.
Speaker: the potential damage from the vascular side increases.
Speaker: That's the concept of the Vili vortex.
Speaker: And we need to keep that in mind to interrupt it so that we don't wind up with a patient who is flattened by these two severe processes.
Speaker: And we need to be very cautious about making sudden changes just because we think the patient is getting better
Speaker: you make sudden changes of large magnitude and you're likely to precipitate another vortex cycle and take you right back to where you were.
Speaker: And I think that especially in a disease where we don't have a specific therapy, a proven therapy yet, I think that doing the best supportive care that we can obviously is the best chance that these patients have.
Speaker: And I kind of see two extremes, and I've seen these patients, unfortunately,
Speaker: But I think that the right approach is obviously as many times a thoughtful walk through the middle.
Speaker: But on one extreme, I saw a lot of people or heard of a lot of people who come in with COVID-19 are very hypoxic and very quickly get intubated and despite of their physiology, get put on a very low tidal volume, a very high PEEP, right?
Speaker: And a high frequency maybe.
Speaker: And those patients potentially could be harmed.
Speaker: On the other hand, I see the pendulum swing, John,
Speaker: to a point where people say, let's avoid intubation, let's put them on high flow.
Speaker: They're laboring, they're laboring, I put them on BiPAP, they're laboring, they keep laboring, probably producing a tremendous amount of self-induced lung injury.
Speaker: And then three, four days later, they're almost an extremist and they get intubated and those patients also probably don't do very well.
Speaker: And I think that a lot of what people are trying to say is we should take what we understand so far of the disease, what we understand of lung injury,
Speaker: and kind of along the middle, walk a thoughtful approach to supporting these patients.
Speaker: And I think that you can break that up into different periods.
Speaker: So maybe we could start by, before these patients get intubated, they show up to the ED and they're hypoxic.
Speaker: How do you think about a COVID-19 patient along those lines, John?
Speaker: Well, we outlined it.
Speaker: I think, fairly well in the JAMA paper, which was restricted to just a thousand words and one table.
Speaker: And so we put these thoughts together.
Speaker: Early on, just make the patient comfortable by restoring their oxygenation and keeping them calm.
Speaker: It's a very important thing to avoid strenuous efforts if you can.
Speaker: Many times you cannot.
Speaker: And if they are progressively pulling hard on their lung, they will have SILI, silly, and need intubation.
Speaker: To delay as long as we usually do until we're forced into intubation, that is not rescue.
Speaker: That's letting the process go too far too long.
Speaker: And in the first stage, I would...
Speaker: worry about giving them enough ventilatory support with non-advasive ventilation or high-flow nasal cannula.
Speaker: And by the way, these things are not contraindicated, as many people have written about, at least not in my mind.
Speaker: They haven't shown that the aerosolization is threatening the universe, as has been suggested by high-flow nasal cannula.
Speaker: Yes, you can get little bits of information here and there that suggest it's not the best approach, but it may be the ideal approach for the patient, keeping them comfortable and relieving their hypoxemia and relieving anxiety, which is a big driver of oxygen consumption and the necessity for power applied to the lung, which is quite vulnerable.
Speaker: from two sides, and pretty soon you will need to intubate those patients if you're not successful.
Speaker: If you are successful, great.
Speaker: Run with it.
Speaker: Let it play itself down to a point where the trend is positive.
Speaker: If the trend is negative, then you need to intubate soon.
Speaker: And I think that it's very interesting how early on, I think, in reports and treatments and
Speaker: I think we kind of fell into the circular thinking, right?
Speaker: I decided to intubate somebody, so now they require mechanical ventilation, as opposed to treating them, like you said, with high flow oxygen, supporting them, and watching them very closely.
Speaker: And then I find when I start doing that, there's more and more people who obviously did not require intubation and who actually were able to be treated like that.
Speaker: Now, many, like you mentioned, might require intubation because
Speaker: they're still struggling.
Speaker: And I think that an important distinction for the clinician here is that we should probably base our decisions on intubation less on fear of contagion, less on infection control issues, and just to the degree of hypoxia, which can be supported with oxygen, and more on whatever objective measure we can find of increased work of breathing, of labor breathing, like you said, and I think like physical exam retractions of the supra,
Speaker: the intercostals, feeling the sternomastoid, placing your finger on the trachea and seeing it, there's tracheal tugging, all things that can help you identify somebody who's really taking these deep, deep breaths, large tidal volumes, which I think would correlate with swings and pleural pressure and high transpulmonary pressures, which is what we're worried about, correct?
Speaker: I couldn't have said it better.
Speaker: Yeah, we need to concentrate on the vigor of the patient's breathing.
Speaker: All of those physical signs are very important.
Speaker: I personally have swallowed more esophageal balloons than any patient in the course of my research career in the laboratory.
Speaker: And I know it's not that bad.
Speaker: But people don't seem to want to put esophageal balloons in patients.
Speaker: OK.
Speaker: And maybe that's not the wisest thing to do in this setting.
Speaker: but certainly pay attention to the vigor of the breathing by the signs that you just pointed to.
Speaker: And before we move on to the intubation and mechanical ventilation portion, I just think that asking a patient who can cooperate in which position they feel better and checking the pulse ox is easy to do sometimes.
Speaker: I think people are adding all sorts of fancy names to this, but
Speaker: basically this awake prone positioning or non-intubated prone positioning or conscious prone positioning.
Speaker: Could you comment a little bit on how you would approach this in these patients who you are supporting with oxygen and what would be the rationale behind it working?
Speaker: Well, Sergio, if the comparison is the semi-supine position in bed, then proning makes the most sense.
Speaker: Upright positioning, if the patient's fully awake and can be placed in a chair-like bed, that may work just as well as prone positioning for their oxygenation.
Speaker: And it may even be a little more comfortable and functional than the prone position.
Speaker: The prone position in most patients is comfortable enough for, you know, a few hours maybe.
Speaker: but then you start to get tired and want to do something else, sit up again.
Speaker: And I certainly think that, you know, awake-prone positioning has a role, but functionally, if they're going to be put back into a face-forward position,
Speaker: I would try the chair bed as a technique to get maximum value.
Speaker: The idea of prone positioning is several.
Speaker: In previous work,
Speaker: It's been looked at and FRC and the number of recruited units really doesn't change all that much.
Speaker: It does change the distribution of what's recruited and the recruited areas tend to match up well with the perfused units in the normal setting.
Speaker: In COVID, there's no guarantee that that's the mechanism.
Speaker: In fact, I'm quite sure that Luciano does not think it's the mechanism.
Speaker: from studies that have been done with the synchrotron and other fancy techniques.
Speaker: It looks as if the prone positioning changes the perfusion distribution, which makes some sense when you think of the vasoplegia part of it.
Speaker: But I'm digressing here.
Speaker: The prone positioning distributes stresses in a little more even fashion than the supine position does, even the semi-supine position.
Speaker: that we use clinically.
Speaker: It's more of a balance.
Speaker: If the strains are at the margin, then proning makes some sense to distribute the strains better.
Speaker: And the shifted perfusion actually helps the oxygenation as well.
Speaker: So I think that a key learning point that I think has changed over the last several weeks is that from intubating everybody, now we recognize that there's a group of patients who probably can be supported without intubation.
Speaker: But the key here is that we should really target adequate gas exchange.
Speaker: We should target non-vigorous breathing and be very conscious that those patients with vigorous breathing are probably the candidates who would benefit the most by from early intubation and not delay intubation in those patients.
Speaker: So once they get intubated and placed on mechanical ventilation, I have also found, John, that some patients without a lot of infiltrates might have very low plateau pressures and may not need a 4 mL per kg at the get-go.
Speaker: Could you talk about how you would approach, once they're intubated, how you think about providing lung protective ventilation and continue to try to interrupt that Vili vortex?
Speaker: Well, as we said before, the lungs vary in terms of their vulnerability.
Speaker: And I think if you have a choice between using six or eight and the patient is awake and tolerates both, use six.
Speaker: If they're awake and they prefer eight, then use eight.
Speaker: By the way, minute ventilation, as it increases in exercise, tidal volume goes up with minute ventilation.
Speaker: To talk about constraining a tidal volume, a tidal volume, I'm not talking about the pressures associated, but to talk about constraining a tidal volume to a value without considering what the minute ventilation is, is asking for trouble.
Speaker: You need to assure at all times that the patient is comfortable and that your gas exchange goals are being met.
Speaker: And that may be your best guide as to what tidal volume to use in the early phase.
Speaker: Now, it's not that you don't monitor plateau pressure.
Speaker: You want to keep that as low as possible in the passive individual.
Speaker: In the patient who's vigorously breathing, the plateau pressure only tells you the lower end of what's being applied.
Speaker: So obviously, you want to...
Speaker: You want not to overstrain the lung.
Speaker: And that means keeping the patient comfortable and applying a plateau pressure that's within what you think are acceptable limits.
Speaker: But nobody knows in COVID-19 whether a driving pressure of 15 and a plateau pressure of 21 is safe.
Speaker: even if under passive conditions, we simply don't know that.
Speaker: It may be more safe than 25, but it's not necessarily perfectly safe.
Speaker: Fair enough.
Speaker: And two more questions in terms of this early phase.
Speaker: I think that one of the comments I would hear very frequently from friends in different states as they were having more cases, and obviously this was very heterogeneous in terms of the spread around the country, but
Speaker: people would say, oh, I use a lot, they respond great to PEEP, I'm using a lot of PEEP, and then I would hear some people say that, well, I mean, I just got into trouble hemodynamically early on, but how would you apply, I think that people confuse also with PEEP, and I know a theme in your thought process has always been that more is not necessarily better, that sometimes less is more, and I think that if 10 of PEEP is good, 20 is not twice as good sometimes, right?
Speaker: So early on, how would you think about PEEP
Speaker: the least amount of PEEP that gives you what you need?
Speaker: Is that the way you think about it?
Speaker: That's the way I think about it.
Speaker: And because the lungs are often, to some degree, recruitable, I am a fan of a decremental approach to setting PEEP.
Speaker: As is shown with prone positioning and COVID-19, when you return the patient to the supine position,
Speaker: usually their compliance is somewhat better and their oxygenation is somewhat better.
Speaker: There has been some recruitment that has occurred.
Speaker: Lung units that were collapsed are now open.
Speaker: That's a little unusual for ARDS, but it can happen.
Speaker: So what that tells me is a sign that the lung may need to be opened.
Speaker: in collapsed areas, but you want to use the least PEEP that keeps the lung open and not use more than usually about eight centimeters, even 10 centimeters is okay, I suppose, of positive end-expiratory pressure.
Speaker: Even advocates of the open lung approach, which include folks who push very high PEEP levels, might agree that
Speaker: these patients have a lot of functional dead space.
Speaker: The more PEEP you apply, the more blood flow you redirect, and the less efficient ventilation becomes.
Speaker: Ventilation is not really a major trouble in these patients, but they do have a large dead space, and expanding the PEEP increases the power, increases the stretch, and redirects the blood flow.
Speaker: And since everybody's targeted on oxygenation and oxygen delivery,
Speaker: You may oxygenate better with higher PEEP, but also limit the cardiac output, which is in many cases not measured.
Speaker: And when patients have flexible lungs, the hemodynamic component is very important.
Speaker: And I think that a lot of the things that we talked about in this early phase obviously minimize the transpulmonary stresses, but you were getting to that in terms of the other part of this vortex that we worry about is the vascular stresses that might be very
Speaker: particular in COVID, what are simple clinical interventions that you think we could do at the bedside to try to minimize or try to optimize preventing vascular stresses?
Speaker: Well, the more evenly you can ventilate, the better.
Speaker: So prone positioning is part of the answer there.
Speaker: The major answer is reduce oxygen demand.
Speaker: meaning keep the patient calm.
Speaker: Don't be afraid to use sedation.
Speaker: And if the patient's fever is in evidence, bring it down.
Speaker: That'll help to some degree, but nothing helps quite as much as relieving the anxiety and any pain that they may be feeling.
Speaker: You want to reduce the blood flowing through the lungs
Speaker: and still satisfy what the patient requires.
Speaker: And that means reducing oxygen demand.
Speaker: Excellent.
Speaker: And as the patients progress and become, that baby lung continues to decrease in size and has a picture that's much more typical, what we call typical ARDS, some of these strategies might change a little bit, right?
Speaker: You might be considering
Speaker: a little bit different in terms of how you approach these patients.
Speaker: Could you talk about how you would imagine the ventilator in somebody who you think is a typical type H kind of phenotype at this point?
Speaker: I don't think I need to vary very much from what we wrote in the JAMA paper and from what folks are traditionally doing with ARDS.
Speaker: If they're an H-type, until you see
Speaker: major improvement in oxygenation and ventilation efficiency.
Speaker: Put them in a prone position.
Speaker: Use all the benefits of prone positioning to your advantage.
Speaker: You may want to test the recruitability and decrementally set PEEP following driving pressure if you're using volume controlled ventilation.
Speaker: to set the positive end expiratory pressure in the later stages.
Speaker: You always want to keep in mind not applying plateau pressures that are inappropriate for that patient.
Speaker: And of course, we want to minimize the ventilation frequency and allow the risk of hypercapnia if the patient will tolerate it.
Speaker: If the patient has metabolic acidosis for another reason,
Speaker: for example, they have kidney disease, you know, they're diabetic with kidney disease or whatever, then renal replacement therapy to reduce the acidosis is certainly indicated.
Speaker: Is there, before we go into just a couple of comments, John, on weaning, and I think that a lot of what we discussed so far is really based on
Speaker: attaining principles and objectives to protect the lung and to provide adequate gas exchange, minimizing the amount of vigorous breathing or the amount of patient-induced lung injury as well.
Speaker: A lot of people become very biased regarding the mode of ventilation.
Speaker: And I hear a lot of, and I've seen a lot of protocols for COVID specifically talking about
Speaker: you should do APRV in these cases, or you try APRV here, try APRV there.
Speaker: Does the mode really make a difference?
Speaker: What are your thoughts specifically on APRV, or is it what we achieve with the mode that really matters?
Speaker: Sergio, some institutions are very adept at using APRV and using the minimal plateau pressure with APRV, letting the patient set the breathing rhythm, et cetera.
Speaker: It depends on how you use any mode, whether or not it's going to be successful.
Speaker: Some people have reported SIMV being the ideal here.
Speaker: It could be.
Speaker: Yeah, I mean, it's almost laughable that somebody's reinventing the wheel, but it's true that in some cases, you escape some of the gas trapping that you get with assist control ventilation by using SIMV.
Speaker: Now, you know, it's, it always is annoying to me when I get a paper to review that talks about this mode versus that mode.
Speaker: It all depends on how they're used.
Speaker: I could give some of my friends who I respect very much any mode and they would be able to make it work.
Speaker: I'm not saying APRV is a bad idea, but don't use it with the objective of fully opening the lung and protecting the lung that way.
Speaker: Because if you do that and you have a mix of L with your H, you may make things a lot worse and the hemodynamics may actually come back to bite you.
Speaker: And plus, it's a mode that
Speaker: often is dependent on some spontaneous breathing activity.
Speaker: And unless the patient is kept comfortable, that can get out of control.
Speaker: So my advice is to use what you're familiar, comfortable with, and is logical to use given the principles we've talked about.
Speaker: And I think that that's an important message because one of the things that I always tell people, and it also relates to
Speaker: to other modalities like ECMO, for example, we're not going to really go down that rabbit hole.
Speaker: But what I tell people, if you normally do ECMO in your institution, you should do ECMO to the right to the patients you feel would benefit.
Speaker: If you don't do ECMO in your institution, a COVID pandemic is not the time to start doing it, right?
Speaker: So you stick with what you know, especially when the numbers go up.
Speaker: But I do think it's important because I do think a lot of people read about APRV and they think that there's something magical about the mode.
Speaker: But really the point is we're trying to, with all the things that we don't know still, trying to apply these principles and that's really the key here.
Speaker: Any words on weaning?
Speaker: And I think that this is something that you did mention in the JAMA paper.
Speaker: and that will be a link to the show notes.
Speaker: But also, I think it's a lot of people have talked about, and we've all observed that these patients don't have a short course.
Speaker: I mean, it takes usually days for them to get sick to the point where they get intubated, right?
Speaker: I mean, 10 days or more sometimes.
Speaker: It probably won't revert in two days, which I think makes sense.
Speaker: I mean, you have a mirror kind of recovery at the best.
Speaker: But I also know from my experience and people in our groups and people I've talked with that early on,
Speaker: being very aggressive with the weaning backfired and these patients got re-intubated.
Speaker: So what are your thoughts in terms of the weaning phase for COVID-19 specifically?
Speaker: It can be summarized easily.
Speaker: Go slow.
Speaker: Don't try to abruptly do spontaneous breathing trials every day and then say the patient is ready for extubation.
Speaker: Because just as soon as you start to lighten up
Speaker: you're loading that patient with more work to do.
Speaker: And if it's the baby lung that you're worried about, it could shrink down again.
Speaker: So go slow is what I would say.
Speaker: Use the techniques that are familiar to you.
Speaker: But it's not the usual problem we face in mechanical ventilation of too much sedation and lighten up and all that.
Speaker: No, you've got to keep them quiet and you've got to keep them
Speaker: not necessarily paralyzed, but quiet, comfortable with as much sedation as you need to use to achieve that objective.
Speaker: I don't know how many studies have been out there with dexmedetomidine in this setting, but that tends to be our go-to drug when we try to keep patients alert and comfortable as we try to wean them.
Speaker: And I think that also, since you mentioned dexmedetomidine,
Speaker: John, if you have a patient in the ICU on high flow or BiPAP, a low-dose dexmetomidine might be all they need to avoid that vigorous breathing.
Speaker: It might be a short course of that might be enough to avoid intubation.
Speaker: I think it's a great drug in that respect as well that you can use it in non-intubated patients in the ICU, obviously.
Speaker: Well taken point.
Speaker: That's exactly the way I would have stated it, Sergio.
Speaker: DEX can really help you calm the patient down and avoid intubation in the first place.
Speaker: Well, I think that definitely a lot to talk, John.
Speaker: I really appreciate your time.
Speaker: Definitely, I think we're learning a lot about COVID-19 and how it impacts the lungs and the amount of information being produced is amazing.
Speaker: I want to thank you for the things that you have put out in the last couple of weeks that I think are very thoughtful and
Speaker: and they'll all be linked on the show notes for our audience.
Speaker: One of the things we like to do in the podcast, John, is at the end kind of tap into the wisdom of our guest and talk about a couple of things that are unrelated to the clinical topic.
Speaker: Would that be okay?
Speaker: Yes.
Speaker: Okay.
Speaker: So the first question relates to books, and I wanted to know if there are books that have influenced you the most or that you have gifted most often to others.
Speaker: Well, in medicine, clearly what set me on my current course was a book by J.F.
Speaker: Nunn that most of the readers have at least seen, Applied Respiratory Physiology.
Speaker: He pulled together an incredible amount of detailed information that made it logical what we do.
Speaker: So Applied Respiratory Physiology, I think it's probably in its sixth or seventh edition now.
Speaker: It's really a terrific book.
Speaker: A second one that helped me a lot in the academic setting has been Why Not Say It Clearly.
Speaker: I'm blanking on the author right now.
Speaker: He was a JAMA editor in the 70s, I believe.
Speaker: And it's a simple, short book.
Speaker: I often give it to my fellows to help them with writing.
Speaker: Why Not Say It Clearly?
Speaker: You know, using the active voice, simple language, taking out all the debris and making it dense enough to transmit all the information, but not so dense that people can't understand it.
Speaker: And I think that not only in medicine, but in general, the ability to write clearly reflects an ability to think clearly.
Speaker: Right.
Speaker: And the beauty of writing, I think, is that it benefits the person who writes it.
Speaker: but also benefits the person who reads it, right?
Speaker: So it really, I mean, is a big plus.
Speaker: So definitely that will definitely link those.
Speaker: Any books outside of medicine that have been of interest for you or that you've gifted often?
Speaker: Well, yeah.
Speaker: I don't gift them very often.
Speaker: And I don't know that I've ever gifted this one.
Speaker: But
Speaker: The book's by Burstyn, B-O-O-R-S-T-I-N, from Yale University.
Speaker: As a scholar, I don't think he's still writing anymore, but the first book I read by him is a very long one, but it's called The Discoverers.
Speaker: And The Discoverers include people who found out fundamental facts about the way the world works.
Speaker: And of course, he covers people like Galileo and the discoverers who learned how to actually navigate longitude.
Speaker: Latitude is easy.
Speaker: Longitude is difficult.
Speaker: And sailors needed that in order to hit the open ocean and cross the seas.
Speaker: So just the process of
Speaker: of discovery is really critical to my development as a scientist.
Speaker: And I really think if I can make a very controversial statement, I think in medicine, we're given an impossible task.
Speaker: And the only way that I feel comfortable in treating patients is to stay nervous about your decisions, check on what you think was right,
Speaker: because you don't know how that patient is going to respond.
Speaker: You can guess, but you don't really know.
Speaker: And one of the things that we do in medicine that may be ethically debatable is to put a white coat on a young person and say, now you're a doctor, go out and make decisions for people.
Speaker: And then put them on a time schedule and make it critical that they make decisions very carefully.
Speaker: very quickly and sometimes in an overloaded fashion.
Speaker: They don't know, but they do prescribe and they do send them for tests and so on.
Speaker: And I think almost every doctor who really is honest with themselves says, you know what, I don't feel comfortable unless I'm sure.
Speaker: And I can only be sure if I know mechanism.
Speaker: And when I don't know mechanism, I've got to explain it to the patient that I don't know what's going to happen.
Speaker: But this is the direction I think it's going.
Speaker: And be honest with yourself.
Speaker: I think that's a great point.
Speaker: And I think that it also speaks to the idea and the way I think about it is that you become a better doctor by having better questions, not by having answers.
Speaker: And I think it just speaks to the immensity of what we don't know, right?
Speaker: And being comfortable with that.
Speaker: And recognizing that I think that early on people are a little bit over competent on what they think they know and don't really appreciate how much they don't.
Speaker: Well, if I can close with a quote.
Speaker: Absolutely.
Speaker: That I use a lot.
Speaker: Bertrand Russell was a great philosopher, British philosopher.
Speaker: And not that I read a lot of his work, but I have read this quote, which I think is wonderful.
Speaker: Fools and fanatics are always so sure of themselves and wiser people so full of doubt.
Speaker: Fools and fanatics.
Speaker: And I can't think of a political example off the top of my head, but maybe you can.
Speaker: Wiser people are always so full of doubt.
Speaker: And so I don't feel embarrassed to be doubting myself.
Speaker: And I...
Speaker: I have principles in mind.
Speaker: The ones I'm fanatical about are very few.
Speaker: The one I'm absolutely sure about are the physiologically based, physics based, mathematics based principles.
Speaker: And that's what I've tried to put down in the energy papers, the power, the shrinking baby lung,
Speaker: and watch for it.
Speaker: There's an editorial coming out probably next month in Critical Care Medicine called, are you ready for this?
Speaker: Dealing with the cards of COVID-19.
Speaker: And it's a longer exposition of the principles that we try to put down in the JAMA paper.
Speaker: So we'll definitely look out for that one and I'll link the other ones that are out already in electronic format.
Speaker: John, I really enjoyed the conversation.
Speaker: I really appreciate you sharing your time with us and hope to have you back on the podcast and to be able to see you soon.
Speaker: Okay, Sergio.
Speaker: Really good.
Speaker: Thanks a lot.
Speaker: Thank you for listening to Critical Matters, a Sound Critical Care podcast.
Speaker: Make sure to subscribe to Critical Matters on Apple or Google Podcasts and share with your network.
Speaker: Sound Critical Care is transforming the way critical care is provided in hospitals across the country.
Speaker: To learn more, visit www.soundphysicians.com.
