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: And now, your host, Dr. Sergio Zanotti.
Speaker: In a previous episode of Critical Matters, we discussed the initial management of patients with ARDS,
Speaker: During this discussion, we covered the concept of lung protective ventilation, which includes low tidal volumes, PEEP, and limiting airway plateau pressures.
Speaker: These are the interventions that every patient with ARDS should receive.
Speaker: However, there is a subset of patients in whom, despite the application of evidence-based lung protective ventilation, hypoxemia persists and can be life-threatening.
Speaker: Today, we discuss the management of such patients, what some refer to as salvage therapy, in patients with refractory hypoxemia and ARDS.
Speaker: Our guest is Dr. Robert Heise.
Speaker: Dr. Heise is the medical director of the critical care medicine unit and co-chair of the critical care committee at the University of Michigan Hospital.
Speaker: He's faculty, a professor of medicine, Division of Pulmonary and Critical Care at the University of Michigan in Ann Arbor.
Speaker: His research interests include ARDS, ventilator-associated pneumonia, and quality improvement.
Speaker: Dr. Heise is an accomplished investigator, clinician, and educator.
Speaker: He is a member of the American College of Chest Physicians Guidelines Oversight Committee and the American Thoracic Society Quality Improvement Committee.
Speaker: Dr. Heise has spoken nationally and internationally on several topics related to critical care medicine and has a special interest in ARDS.
Speaker: He has published multiple articles and chapters in medical journals and textbooks and has been a reviewer for the Annals of Internal Medicine, American Journal of Respiratory and Critical Care Medicine, Chest and Critical Care Medicine.
Speaker: Bob, welcome to Critical Matters.
Speaker: Thank you, Sergio.
Speaker: So today we want to expand a little bit on a previous conversation that we had in the podcast related to ARDS and really think about those patients in whom we apply low tidal ventilation, we apply appropriate PEEP, we protect the plateau pressures, but they're still in what we call refractory hypoxemia, are still having trouble.
Speaker: And I think that a good starting point might be, Bob, is
Speaker: How do you define refractory hypoxemia?
Speaker: I know in the literature there's a variation of definitions, but in your concept, what would you call refractory hypoxemia?
Speaker: Well, that's a great point because there is a lot of variability and no uniform definition.
Speaker: There was an article in the Annals of ATS I reviewed and wrote an editorial for where the Canadians resorted to their definition, which did not really include recruiting with PEEP.
Speaker: So for me, if you're talking about moving on to some of the therapies you mentioned,
Speaker: that would be patients who are not adequately saturating or oxygenating despite high levels of FiO2 and appropriate attempts to recruit with PEEP.
Speaker: So I can't give you some numbers, but ballpark numbers, let's say if your PO2 is still in the 50s when you're on 1820 of PEEP and you're not doing well, I mean, that's close enough.
Speaker: But I don't have a single working definition because there isn't a uniform definition agreed upon in critical care medicine.
Speaker: And from a practical and tactical standpoint, I think that you mentioned something very important.
Speaker: I mean, in terms of it's hard to say this is the exact number, but you did mention in somebody who's already on 20 or more of PEEP.
Speaker: So I guess that my question is, this is not something that you make a diagnosis as soon as the patient hits the ICU, but really something that you consider after doing some things first, right?
Speaker: I mean, exactly.
Speaker: You mentioned low tidal volume ventilation.
Speaker: Sure, exactly so, but it doesn't end there.
Speaker: when you're saying refractory hypoxemia you're generally talking about a what we call berlin severe patient who will have a pdf ratio under 100 on at least five a peep that's a that's a starting point for someone who uh might be uh a trouble to manage and then and then you recruit them and if you if they're eminently recruitable you might get by with say i don't know 15 a peep and next thing you know the fio2 is dialed down to 50 and i'm good to go i don't think i need to resort to these other things we're about to discuss
Speaker: Okay.
Speaker: And I think that this is important because I often encounter clinicians who very quickly, before a lot of these basic interventions are given any time to work or applied, are thinking of salvage therapies of all sorts.
Speaker: And I think that it's important to go in a stepwise approach and implement what's been proven to work first, see how patients respond.
Speaker: And then if we still have trouble, we now start thinking about this patient has refractory hypoxemia.
Speaker: Absolutely.
Speaker: We'll get to a sequence, ECMO, at the end, but there was one series that showed that only 31% of patients in this one series who received ECMO had an attempt to have prone ventilation before they pulled the trigger on ECMO.
Speaker: I think you're exactly right.
Speaker: There is an explicit evidence-based hierarchy here, but I do think there is some logic to this.
Speaker: Let's start with that logical approach.
Speaker: If you have implemented a lung protective ventilation, have optimized your PEEP,
Speaker: and after, let's say, a couple of hours, your patient's in the ICU, and you're still having hypoxemia, you still have a PAO2 that is not making you comfortable, like you said, what would be your next step, or what would be the first tier of interventions that you would try for that patient?
Speaker: Well, there's two things I think about then.
Speaker: One, at this juncture, with current evidence, would be neuromuscular RAC-K to avoid dyssynchrony, and that might improve the patient's stead.
Speaker: We're going to have more information about
Speaker: We're part of the Pedal Network, and Mark Moss and Derek Angus are the PIs on our publication.
Speaker: It will be out, I believe, at the SCCM meeting regarding whether or not the Acurisis article is really the way to go.
Speaker: We'll have more evidence on that.
Speaker: The other thing I often think about is body habitus.
Speaker: Danny Calmore has EpiVent 2 coming out.
Speaker: We participated here as a site for that.
Speaker: And patients whose abdomen is in play, post-op bellies, ascites,
Speaker: morbid obesity if I get to 20 a peep and think that I'm not helping the patient enough another thing to think about is we'll do transpulmonary pressure we'll put esophageal balloon in now admittedly that's not a technology that's uniformly available but we have that ability we have to switch out a ventilator for that and in that case you could argue that 20 people aren't really seeing that airway pressure because we think about a plateau pressure a plateau pressure
Speaker: incorporates the chest wall and the belly is part of the chest wall.
Speaker: So in circumstances where the abdomen is exerting significant pressure, if you will, on the lungs, that can be another thing to do.
Speaker: So those are the two sort of branch points I hit at that moment.
Speaker: So I think that those are excellent points.
Speaker: And like you said, I mean, first think about is there anything individual about this patient, such as morbid obesity, that might mean that what I'm doing right now is not optimized for their physiology.
Speaker: And I think that's a great example of where maybe you need more than just driving pressure or maybe where you need more than just measuring compliance.
Speaker: And esophageal balloons might become helpful for that subpopulation, like you said.
Speaker: And then the second thing that I would like to poke your brain a little bit more about is the use of neuromuscular blockers.
Speaker: And obviously, it's a great way to decrease patient ventricular asynchrony.
Speaker: There is one large randomized trial, the French trial, like you mentioned, that has shown positive outcomes.
Speaker: But could you give us a little bit more in terms of practical aspects?
Speaker: How long do you do it?
Speaker: Do you follow training for?
Speaker: I think some people have talked about that all you have to do is really target to your ventricular support.
Speaker: and how long do you do it?
Speaker: Well, you know, in a curasis it was done for 48 hours, but in patients in whom I do it, I do it as long as it's necessary.
Speaker: In other words, at 48 hours, if you remove neuromuscular K and you desaturate, well, I go longer, and if sooner than 48 hours you have gas exchange that has improved significantly, I'll back away.
Speaker: So, there's sort of a 48-hour notion, but I'd certainly go either way on that.
Speaker: In terms of training a four,
Speaker: You know, that was a criticism of the Curisys article.
Speaker: And frankly, I'm not sure it matters as much in the modern era with the Etercurium being the agent of choice for so many individuals.
Speaker: That's broken down in the blood.
Speaker: You don't get the kinds of prolonged paralysis issues we used to see with Vecuronium in the setting of organ dysfunction.
Speaker: So I'm not against Trinafore, but I think really it's not a critical piece here.
Speaker: If you're not triggering the vent, you're not triggering the vent.
Speaker: So I'm not against it.
Speaker: I think a more challenging and least evidence-based issue relates to monitoring of sedation during neuromuscular blackade.
Speaker: And it's a controversial issue.
Speaker: We still have a bis monitor, bispectral analysis, and the literature on that has been not particularly supportive or great.
Speaker: But what else do you have?
Speaker: In other words, the notion of a family member walked in a room and a patient's pulse went from 100 to 110, therefore they must be too light
Speaker: with their sedation when they're blockaded, that doesn't really help you that much.
Speaker: I admittedly, BIS is a mixed bag, and I'm not going to dispute that.
Speaker: But to me, training a four, I'm not against it, but I don't think it's completely essential for guiding you.
Speaker: I'm more concerned about ensuring your patients are sedated because, of course, neuromuscular blocking agents are not sedative.
Speaker: Yeah, absolutely.
Speaker: I think that you point out to great issues that not everything that we use has the highest level of evidence, but as long as you understand what are the limitations.
Speaker: And I guess the other aspect that I think is very important, and we'll probably touch on this when we get to pronin and ECMO.
Speaker: is also the routine or the discipline of doing things a certain way in your unit trains your team to be very efficient in handling that.
Speaker: And I think that people can learn from that as well, obviously at a lower level of evidence that may be large randomized trials.
Speaker: But I think that having expertise in your own unit with certain things and having a way of doing it is always, I think, a good thing while we wait for more evidence.
Speaker: I couldn't agree more.
Speaker: Process of care, we call that.
Speaker: And having a solid process of care...
Speaker: is very impactful and meaningful for good care delivery.
Speaker: And you mentioned one of the studies that looked at ECMO in the ATS annals.
Speaker: I've reviewed some articles that have looked at kind of the spectrum of what's going on with salvage therapy.
Speaker: And that was very interesting to me in two articles that looked at this at the ATS journals.
Speaker: They talked about the lack of an established process for salvage therapy in most ICUs, how very few of the ICUs that they examine have a very defined protocol and process of how they escalate this.
Speaker: Could you comment on that a little bit?
Speaker: Well, we have one here at University of Michigan, and we try to, within reason, adhere to it.
Speaker: It echoes at the bottom of the page, which is to say these other things are attempted prior to
Speaker: And as I mentioned to you, that article about some of these modalities not having been attempted prior to ECMO cannulation is a concern.
Speaker: So we have a protocol because we have a lot of these patients.
Speaker: We have patients referred in.
Speaker: And also because we're all here big believers in process of care and not just sort of approaching things willy-nilly.
Speaker: And I think that's an important point.
Speaker: And as people will find out that they don't know yet, ECMO has a long story in the University of Michigan, and there's tremendous experience there.
Speaker: So I think that we'll touch on that a little bit more.
Speaker: So after you've optimized, Bob, for the individual patient, an example you gave was a morbid obese patient, you've thought about neuromuscular blockers.
Speaker: If you're still having trouble, what's next?
Speaker: Well, prone positioning is on my list, right?
Speaker: And the Perseva article, which came out a few years ago now, I think had an incredible mortality benefit.
Speaker: And there's been a lot of reading of the tea leaves there with regard to what that meant.
Speaker: Either they got lucky or they finally got it right, if you will.
Speaker: I mean, earlier proning articles and studies have been negative, were criticized for lack of use of lung protective inhalation, not keeping them prone long enough.
Speaker: things of that sort.
Speaker: And so if you look at the historic timeline, it looks like things are getting better until finally Proceva hit and hit big.
Speaker: If you look at a force plot for treatment effect, it was pretty dramatic, but it doesn't matter to me.
Speaker: I mean, I'm sold.
Speaker: And it's become a go-to thing at that point, particularly in the context of other potential go-to things such as oscillation, which has had some negative studies.
Speaker: So I think proning should be done.
Speaker: Now, I've been
Speaker: Around the country, I've got a good friend, Ivor Douglas, at Colorado.
Speaker: They're big believers in proning.
Speaker: They'll do it pretty quickly.
Speaker: In Proceva, they waited 12 hours.
Speaker: And so, in other words, I'm going to optimize the patient.
Speaker: If I can optimize the patient, like I said, if I'm at 18 to 50% oxygen, I'm not going to prone.
Speaker: But if we're talking refractory, as I've kind of amorphously described it to you, that would be my next thing.
Speaker: And I think that in terms of talking about more of the practicalities and what we can learn from the proning trials that were positive, why don't we start with maybe reminding the audience, what are the theoretical benefits of prone position ventilation in a patient with ARDS?
Speaker: Well, so in ARDS, despite having lungs that leak, if you will, diffusely, there's still gas exchange heterogeneity due to compressive atelectasis.
Speaker: By that, I mean,
Speaker: the weight of the lung causes smush in the dependent lung zones.
Speaker: And so does the weight of the heart when you're lying on your back.
Speaker: So the distribution of ventilation is more equitable when you are lying prone and the weight of the heart is removed and the posterior inferior aspects of the lung where a lot of that smush occurred are now able to be aerated.
Speaker: So you're redistributing where the delivered mechanical ventilation breath will get to when you put a patient prone.
Speaker: And that goes back to what some people call ventral preservation, which are dorsal preservation, which is kind of a evolutionary, um,
Speaker: aspect of biology.
Speaker: And in terms of, Bob, in terms of, you talked about some of the things that people thought we were doing wrong.
Speaker: So one of the things that they did in the proning trials that were positive is they selected their population a little bit better, right?
Speaker: They only took patients who, A, were severe ARDS, so were very hypoxemic, but also they standardized what everybody got before they got proning.
Speaker: I think that those are two important aspects.
Speaker: I know clearly I should have mentioned that because I did mention other notions of earlier criticisms for not being lung protective and so forth.
Speaker: But that was another criticism was if you're not that severe, you're not going to be helped one way or another.
Speaker: And so if you enroll patients aren't that sick, then you're not going to see what you need to see in terms of a difference in outcomes.
Speaker: And one of the things that has always attracted me from prone positioning, especially with our group that practices in a diverse range of hospitals in the community, some very large and with a lot of technology, some, I mean, that are smaller, is that good proning is not about technology but about process.
Speaker: However,
Speaker: But before we go into some of the details of how you prone, there are a lot of commercial beds that have tried to capitalize on the proning.
Speaker: Do you use those?
Speaker: Do we need those?
Speaker: We do not have the luxury of having them.
Speaker: So my experience with them is limited in terms of visiting other institutions.
Speaker: So we do it the old fashioned way with hard work and lots of nurses.
Speaker: And it works fine, right?
Speaker: So as long as you train people and you have a team.
Speaker: Listen, these people, as you can imagine, one reason this is not embraced, if you will, is because these people are sick and they're plugged in with tubes and lines and such.
Speaker: And, you know, you want to get all tangled up.
Speaker: So it is not easy to...
Speaker: Other than the fact that it's the right thing to do, there's no reason to do it, right?
Speaker: In other words, it's not easy, but it is the right thing to do.
Speaker: And when you decide to prone somebody, so you said, I mean, you start with optimizing everything and they're still not responding.
Speaker: At that point, you might consider, okay, my next step is proning.
Speaker: Absolutely.
Speaker: How long do you usually let them prone?
Speaker: Minimum of 16 hours a day.
Speaker: I mean, this is the point.
Speaker: This is one of the issues.
Speaker: If you look at the – there's lots of ways to look at the –
Speaker: croning studies have done in the past, but I think there was a Cochrane analysis, for example.
Speaker: Clearly, you need to spend the vast majority of the 24-hour period face down.
Speaker: I think that has been shown to be one of the reasons this can be impactful.
Speaker: So, minimum 16 hours a day, and then flip them back for the remainder, and then back and forth, but mainly down until you really get a break in the action and see that they're oxygenating better.
Speaker: I had a colleague of mine
Speaker: He texted me about a case the other day, he said, when do you stop?
Speaker: I said, well, when they're better.
Speaker: But it is clear that you need to be face down for much of the day.
Speaker: And like you said, I mean, you see how the response is and when you stop getting that response is when probably when it, when a plateau is that maybe it's not working anymore.
Speaker: What about somebody that you prone and has no improvement 10 hours later?
Speaker: What do you do?
Speaker: Well, I mean, that's the point.
Speaker: We don't have – well, no, we're ECMO center, and oscillation used to be sort of in my algorithm, but the OSCR, but especially the Oscillate trial, which we did at this place, part of the Canadian critical care trials group, had a worse outcome.
Speaker: So, to me, that's kind of getting towards the end of the road towards ECMO.
Speaker: So, you say you're prone, and you're peeped, and you're paralyzed, and
Speaker: so forth, sedated, and you're still not getting anywhere, and it's early on, right?
Speaker: We're not talking about on day 10.
Speaker: We're talking more on day two or three.
Speaker: I will call my surgical colleagues.
Speaker: I don't do ECMO here.
Speaker: That's my surgical colleagues do that.
Speaker: You mentioned Bob Barlett, who is no question a visionary in this regard.
Speaker: And so we have a strong ECMO program and always have.
Speaker: I admit the fact that for me, it's easy, right?
Speaker: You don't have to make a transfer.
Speaker: They're already in the building.
Speaker: But if things aren't working, that's where I go with it.
Speaker: Now, I would mention that the oscillate trial, oscillators were not routinely distributed throughout the land either.
Speaker: Ours are collecting dust in the closet following the oscillate trial, but there was a post-hoc analysis of oscillate data last year showing that the PDEF under 50 may have had a mortality benefit, even accounted for confidence intervals.
Speaker: When you have a trial that shows no benefit, the true believers can say, like what they did with proning for many years, yeah, you didn't do it right until they finally got it to hit.
Speaker: Oscar in Britain had no mortality change, and Oscillate had a worsened mortality.
Speaker: So if things actually are made worse, then you've got to stop doing it.
Speaker: But like I say, the post hoc analysis suggested that maybe the very sickest of the sick
Speaker: might benefit.
Speaker: But that's not back in my algorithm, but I'm aware of that information and kind of a little bit tempted.
Speaker: And before we leave the proning procedure, any recommendations, any practical tips or what are the things that you worry about with proning that could be, I mean, useful for our audience?
Speaker: Well, I mean, there can be, in terms of sort of side effects or complications, you can get local necrosis of the forehead or bridge of the nose.
Speaker: There may be an increased
Speaker: ventilator-associated pneumonia risk.
Speaker: So there's no free lunch in critical care.
Speaker: I mean, everything we do has got limitations in terms of benefits and risks.
Speaker: But I think I am a believer at this point, and I think it, just like going for protective ventilation, unfortunately, is probably underutilized.
Speaker: In fact, not probably.
Speaker: I can show you the data.
Speaker: It is underutilized.
Speaker: So I think that before we jump into a little bit more of alternative modes of ventilation and ECMO, just to summarize, in terms of what the evidence would suggest right now is make sure that everybody's getting low tidal ventilation, that we're protecting their plateau pressures and utilizing PEEP.
Speaker: And those who we still have problems, evaluate for individual aspects of the patient, like morbid obesity that might require maybe a more invasive or more refined way of optimizing lung protective strategies.
Speaker: If that is done and we still have trouble, then think about neuromuscular blockers.
Speaker: And then the next step would be prone ventilation, which, like you said, is important to do for prolonged periods during a 24-hour period, so 16 hours or more.
Speaker: and to do it as long as we need to, and as long as the patient is showing improvement.
Speaker: If all that fails, we then have to move to other directions.
Speaker: So you did mention oscillations.
Speaker: I think that it would be worth talking a little bit about alternative modes of mechanical ventilation.
Speaker: And you talked about high-frequency oscillatory ventilation, and after the H1N1
Speaker: influenza epidemic and a lot of people thought that oscillation was the go-to therapy and I know a lot of hospitals that bought oscillators and even governments about oscillators and it made sense I mean because we were keeping people at a higher mean airway pressure we thought we were protecting the lung from from better induced lung injury yet like you mentioned the trials did not show that what are your thoughts on are we done with oscillation for the for now
Speaker: Well, we are for now because when you have a trial showing harm, you really can't justify routine clinical use in sort of your algorithm.
Speaker: And I worry that there might not be another trial.
Speaker: I mean, if you kind of think about a randomized trial, how about a randomized trial where at that point you go to ECMO versus oscillation?
Speaker: That's an interesting and very, very challenging trial to do.
Speaker: But I am intrigued by the notion of recruitment
Speaker: it would, like I say, PDEF under 50.
Speaker: I mean, that's bad, right?
Speaker: That's a very, very sick patient.
Speaker: And it would be unfortunate if there were no, there was not an opportunity in the future to maybe examine that more carefully.
Speaker: But an RCT with that population, it makes for a very, very difficult trial.
Speaker: So I have to say we're probably done with oscillation, but I'm intrigued by the post-hoc hypothesis generating notion
Speaker: that PDEF under 50 might benefit.
Speaker: Interesting.
Speaker: And the other, we were talking before we started recording the podcast that in terms of alternative modes of ventilation, Bob, I have observed, I mean, either in sign out or visiting some of our programs that there is a certain amount of clinicians who like to use APRV or airway pressure release ventilation in ARDS.
Speaker: Could you tell us a little bit about APRV and where it fits in your algorithm?
Speaker: Well, it's sexy and it's trendy and it lacks evidence.
Speaker: And I try to know as much as I can about it.
Speaker: Let me tell you.
Speaker: Well, what it is, of course, is you give a mean airway pressure, you're allowed to breathe spontaneously, and episodically you're released, if you will, to a lower airway pressure at some level of PEEP.
Speaker: And the idea here is that high mean airway pressure around which you're allowed to spontaneously breathe recruits the lung maximally.
Speaker: I'm not against it, I just need to, you know, sort of the old state of Missouri, you gotta show me.
Speaker: And I'm completely well versed in this.
Speaker: In fact, I had the opportunity to venture out earlier this year to Syracuse, where they do basic science research, and Neil Hibashi came in from Shock Trauma.
Speaker: I mean, I'm all about trying to do the right thing and figure it out.
Speaker: And there's no question that they have a heartfelt belief in this modality,
Speaker: And if it is superior, I want to know it.
Speaker: Now, the literature is wanting.
Speaker: About a year ago in intensive care medicine, there was a Chinese article which had a small single center.
Speaker: The randomization was they had sicker people in the control group, but it did seem to have more ventilator-free days.
Speaker: There was a pediatric trial in the Blue Journal earlier this year where APRV had a higher mortality.
Speaker: There was an EPUB done out of Utah.
Speaker: where they were thinking about doing an RCT, they stopped after 50 patients because they found the release tidal volumes were very high.
Speaker: There was a basic science article by Kavanaugh out of Toronto, just EPUB a week or so ago in the Blue Journal showing that, I think it was a mouse model or rat model, the release going from that high mean air pressure to that lower thing can precipitate lung injury.
Speaker: So now you know the world's literature.
Speaker: And I don't doubt that I have talked to, as you have, people
Speaker: who use it and swear by it.
Speaker: And I'm okay with that, that they believe what they believe, and I would like to see a larger trial.
Speaker: Now, I was recently awarded some money by the Chess Foundation, and we might attempt to try to look at APRV with biomarker endpoint as part of that small clinical sort of phase two.
Speaker: But I would like to see a big trial because
Speaker: There are people that really believe it.
Speaker: Now the problem with APRV is that there's a lot of ways, it's not like, it's just as refractory hypoxemia.
Speaker: There's not a single best definition.
Speaker: I think the folks out in Syracuse would argue that this must be tailor-made to the pressure waveforms of each individual patient rather than employing a single one-size-fits-all approach to this.
Speaker: And so they would argue any negative trials relate to not having done it right
Speaker: And so there's heterogeneity in the literature too with regard to what's been done.
Speaker: I looked at a recent submission for a meta-analysis that tried to look at the handful of very small RCTs, 300 patients.
Speaker: This is still under review elsewhere, but that's one of the challenges.
Speaker: Can you do a meta-analysis when you haven't done the same thing to every patient in every study, right?
Speaker: I mean, it's a gamish.
Speaker: So I am happy to...
Speaker: see whether APRV for the severe patient, the Berlin severe, the refractory hypoxemic patient, if you will, is superior.
Speaker: I would love to see a trial.
Speaker: But until such time as that trial occurs, everyone's got their anecdotes.
Speaker: I've got mine.
Speaker: And until we have evidence, I think it's very nice to look at lung recruitment, but that's not enough.
Speaker: That's not enough to say you've got a superior modality.
Speaker: And I think that those are great points because in some patients, if you've exhausted the evidence-based supported interventions, sometimes you might try things.
Speaker: I mean, truly as a salvage kind of left throw.
Speaker: But I do think that what is troublesome from an evidence-based perspective is that we still don't have 100% penetration in the things that are proven to work.
Speaker: And people are jumping to the new and sexy where there's really no good literature to tell us, are we harming or helping our patients?
Speaker: Yeah, I mean, just having a quote, get an oscillate.
Speaker: Let me take the opposite point of view.
Speaker: With an oscillator, that recruits the lung really well, too.
Speaker: And you think that ought to work.
Speaker: And it didn't.
Speaker: Or with the ARC trial, you know, the very large recruitment, you want to recruit the lung, these very large recruitment maneuvers that
Speaker: they use with a decremental PEEP approach.
Speaker: That's really recruiting and opening the lung.
Speaker: Everyone thought that would work.
Speaker: That improved harm.
Speaker: So you never know until you study it.
Speaker: And I'm fine with people having opinions provided they recognize the limitations of those opinions.
Speaker: Fair.
Speaker: So I think that at this point, really, we should maybe jump into ECMO.
Speaker: And ECMO, I think, is fascinating because a lot of people talk of it as new technology.
Speaker: It's been around for decades.
Speaker: It's evolved.
Speaker: But yet, I mean, despite some attempts to really study this, I think that we still have the people who believe it and people who don't believe it, and we don't really have the best answer.
Speaker: But clearly, it's utilized.
Speaker: And I think that
Speaker: how to utilize it with the available evidence in the best way, whether you do it or you don't, and you should be thinking of when to refer patients, I think is definitely a worthwhile discussion.
Speaker: Do you want to start maybe just telling us a little bit about the literature behind ECMO?
Speaker: And I guess CESAR and EOLIA are probably the most important for us.
Speaker: So the CESAR trial published in Lancet, and I believe 09, was done in the UK in
Speaker: And the randomization there was you get referred, you get randomized, and then you either were left wherever you were and told to be lung protective with a wild type, if you will, or brought to Leicester to get ECMO.
Speaker: And just as our experience here, you know, where only about a third of the ECMO transfers actually get ECMO, many patients referred to Leicester to randomized to ECMO, didn't get ECMO, and in fact,
Speaker: the mortality benefits seem to be accounted for by that fact.
Speaker: And that's fine, but the criticism of CESAR was, were you testing the technology, were you testing regionalization of care?
Speaker: And that people in Leicester know about RDS, they care about it deeply, they know what they're doing, and they didn't put everyone on ECMO because they got them better without it, which is the right thing to do, of course.
Speaker: So that's what led to Eolia, and I will tell you, it's one of the fun things is to
Speaker: At the ATS, they have a session where they announce trial results live in front of the studio audience, if you will.
Speaker: And EOLIA was announced live this past May.
Speaker: And folks, you can't make this up because it was powered for a 20% mortality difference.
Speaker: And then the trial was stopped early for futility with only 11% mortality difference.
Speaker: So the Kaplan-Meier survival curves look really widely separated.
Speaker: And yet it was a negative trial.
Speaker: stopped for futility.
Speaker: Now, there was a lot of problems, one of which was a 28% crossover between the control group and ECMO.
Speaker: And if you take those crossovers out and say they probably would have been dead without crossing over, you get a fairly impressive mortality difference.
Speaker: So the ECMO believers would argue this was a positive trial.
Speaker: The statisticians would say it was stopped early for futility.
Speaker: Now, there was a further analysis, a Bayesian analysis performed a
Speaker: that was published in JAMA within a month.
Speaker: And I think it probably has a mortality benefit.
Speaker: And one of the things, of course, was patient selection.
Speaker: And they had very rigorous, these were really sick people.
Speaker: We talked about the definition of refractory hypoxemia before.
Speaker: These were even more refractory than that.
Speaker: You had to be pretty hypoxemic over various intervals of time to qualify.
Speaker: So I'm willing to accept that this was a positive trial based on the Bayesian analysis, based on
Speaker: the crossover issue, but technically it was a negative trial.
Speaker: So it was so funny because everyone knew the study was coming out and wanted to, that's why the hall was packed, I believe, that day at the American Thoracic Society meeting.
Speaker: And then to have it come out as a technically negative trial with a big mortality difference.
Speaker: One of the things is, you know, you're going to power a trial for a 20% mortality difference.
Speaker: That's being pretty optimistic for critical care trials.
Speaker: You don't see that every day.
Speaker: So it was a fascinating moment when they came out.
Speaker: And then, of course, there's been a lot of back and forth since then.
Speaker: But the Bayesian paper in JAMA being perhaps the most recent word on that.
Speaker: But I mean, you can't argue that it's already hurt people, and it probably helped them.
Speaker: And you mentioned the H1N1 epidemic.
Speaker: That also was a big driver for ECMO,
Speaker: programs opening up because in that epidemic, these were young, previously healthy people who had extra-refractory hypoxemia.
Speaker: You just couldn't oxygenate them.
Speaker: I think that if you're a 24-year-old kid, you'd probably want them on ECMO.
Speaker: I think that in the severest cases, that's probably the case.
Speaker: Now, hey, I'd love to see, based on what I mentioned before, oscillator versus ECMO.
Speaker: That'd be an intriguing thing in that really refractory population.
Speaker: I'm willing to accept that something happened in Neolia, and it was probably a good thing.
Speaker: And I think that, obviously, on a larger or higher level, what it speaks also to is that the construct of the clinical trial, despite being
Speaker: what we consider the highest level of evidence based on scientific method, it has limitations in answering every question that we have.
Speaker: And it's not perfect.
Speaker: And it's very difficult sometimes to conduct these trials, to complete these trials.
Speaker: I know one of the commentary that has emerged out of Terriolia was that the enrollment rate was so low that to reproduce this trial again would take again, I mean, a very long time in a lot of centers and with believers and non-believers who
Speaker: right?
Speaker: Sometimes that might be very hard to pull off.
Speaker: Well, that's the dirty little underbelly of being a trialist.
Speaker: I use a quote by Bismarck that said, politics is like making sausage.
Speaker: I modify it and say clinical trials are like making sausage.
Speaker: It tastes great, but you don't want to know what's inside.
Speaker: When you see the PDF coming off the wire, it looks like revealed wisdom, but there's a lot going on there to get to that publication.
Speaker: And you recruit patients one by one,
Speaker: And I am a trialist hoping something good happens at the end, something makes sense.
Speaker: But on an individual patient by patient basis, it seems like chaos at times.
Speaker: And in terms of, I think just to refresh our audience,
Speaker: Some of the aspects that I think are important about this trial that we can learn is like you said, 28% of the patients who were randomized to conventional therapy actually did cross over to ECMO and the thought was that they were probably going to die.
Speaker: But none of those patients crossed over at day 10, 12, or 13, right?
Speaker: It was early.
Speaker: That's important.
Speaker: All of those patients… So you're saying that maybe they weren't going to die.
Speaker: I mean, that's absolutely true and there's no way to know that.
Speaker: Also, I think it's important to remember that the patients who were part of this trial were very sick just to enter.
Speaker: And I have the notes here, but basically, you had to have a PAO2 to FiO2 ratio of less than 50.
Speaker: for three hours, consecutive three hours, or a PAO2, FiO2 of less than 80 for more than six hours after optimization of the things that you talked about as well.
Speaker: So these were sick patients.
Speaker: It's not like, oh, they showed up hypoxemic and they just got randomized to ECMO or conventional therapy.
Speaker: which I think goes to your point that in the patients that you would consider ECMO, the first step is to make sure that you go through that algorithm of things that work.
Speaker: And if they're still having problems, you still think that they might die from hypoxemia.
Speaker: And then I think that that's when you would probably consider.
Speaker: Are there any patients, Bob, that why don't we talk about the people who would not be candidates for ECMO in your book in terms of you've done everything you can, but
Speaker: They're still having problems.
Speaker: Not every patient is going to get ECMO.
Speaker: Are there some contraindications that you would consider?
Speaker: Oh, sure.
Speaker: I mean, obviously, the most important one being comorbidities.
Speaker: If someone has widely metastatic cancer, for example, I don't think that it would be indicated.
Speaker: Duration of hypoxemic respiratory failure is also a consideration.
Speaker: Historically, my center here will not consider cannulating beyond about a week.
Speaker: So, I mean, I think those would be two precautions.
Speaker: pretty clear-cut ones.
Speaker: And, you know, extremes of age, you could argue, too.
Speaker: I mean, they're really going to put a 90-year-old on ECMO or something like that.
Speaker: So, but the comorbidities, at least in my world, again, it's not a question of waiting longer than a week.
Speaker: They're already in the building.
Speaker: But, you know, bone marrow transplant patients, we have a lot of people who have a lot of illnesses at the time they develop ARDS.
Speaker: And, you know,
Speaker: You can certainly manage things on a case-by-case basis, so you can get a consult and see what they say, and then you can tell the family you've done everything, I mean, with a clear conscience.
Speaker: But I think patient selection is an important and reasonable consideration.
Speaker: You talk about triage, you know, we use, and rationing, I mean, we use explicit rationing for organ transplantation.
Speaker: There's no question, who do you select?
Speaker: Well, we only have so many circuits, right, ECMO circuits.
Speaker: And there are some clear-cut pediatric evidence-based indications that we get referrals for as well for the circuit.
Speaker: So it is also a situation of some rationing and triage.
Speaker: And you're not morally obligated to give ECMO if you have comorbidities that are overwhelming.
Speaker: And in terms of timing, you mentioned earlier during the podcast that this should be an intervention that we should apply early, right?
Speaker: I mean, for people who've been in ARDS for a long time and not responding or getting worse, it might be too late.
Speaker: Can you talk about that a little bit, Bob?
Speaker: Well, that's what Bob Barlett always handed down.
Speaker: I mean, in general, I'll tell you what, when I come on service and get an ARDS patient after about a week, I'm not that happy, not because I didn't use ECMO, but
Speaker: But B, I happen to think that the situation with regard to the lung component anyway, putting aside any comorbidities, tends to be won and lost early, getting them recruited early.
Speaker: There was old days with literature about oscillate with the first three days.
Speaker: To me, the first couple of days, trying to really be aggressive and getting them recruited or treated, proned, or what have you, my bias is the dysent tends to get cast.
Speaker: And so the notion of...
Speaker: that the lung is going to heal particularly well when you're, whatever, say two weeks into an ARDS run and refractory at that point, is not that likely.
Speaker: In the bad old days of high tidal volumes, what we used to see were macroscopic airesis and recurrent pneumothoraces.
Speaker: So if the lung doesn't heal, it can be pretty nasty.
Speaker: Now that might have been more a function of a large tidal volumes, but if the lung doesn't repair it, we say that most patients with ARDS don't die hypoxemic death, but some do.
Speaker: And there comes a point at which that you can put them on ECHO if you want, but the lungs aren't going to heal.
Speaker: Now, we don't have a day-by-day rendering.
Speaker: There is clinical judgment involved.
Speaker: So it's not to say that day eight, you're completely unsalvable, and day six, you are completely salvageable.
Speaker: But it does get factored in.
Speaker: Terrific.
Speaker: And I think it's an important point.
Speaker: And I think that it's a balance, right?
Speaker: On one hand, implement therapies that have been proven to help first.
Speaker: Make sure you go through your arsenal or your toolkit of what's evidence-based.
Speaker: And if you're still having trouble early on, think about you're referring to an ECMO center,
Speaker: If you don't have one, I also think that ECMO is one of those things that probably should be done at places where they do it a lot.
Speaker: So doing one or two cases of ECMO a year probably does not make you an ECMO center.
Speaker: And that patient will be better served in a place that does one or two a month or a week.
Speaker: And I think that that's very important.
Speaker: Yeah, no argument.
Speaker: That's called the volume quality paradigm.
Speaker: Would you rather be two bypasses a year or 200, right?
Speaker: Exactly.
Speaker: And I think that that's something that everybody likes to do as an intensivist or most intensivists like to do new things.
Speaker: But I think it's also important to remember that and have clear criteria of what are the things that you do well at your unit and what are the things that maybe because of volume would be better served patients at another place, which I think is very important.
Speaker: So I think, Bob, that as a summary, I do think that some very tactical points for our clinicians at the bedside are that there are things that have been proven to improve mortality.
Speaker: That includes a lung protective strategy with low tidal ventilation, using PEEP and protecting patients from high plateau pressures.
Speaker: When that should, everybody should get that immediately.
Speaker: When that doesn't work, try to optimize based on individual patient aspects such as body habitus.
Speaker: And you talked about esophageal pressures being of use in that population specifically.
Speaker: Neuromuscular blockers early on for patient asynchrony, and like you said, probably in the next coming months, we'll hear from a larger trial that is being conducted or has been completed.
Speaker: So we'll have more about that to discuss.
Speaker: Prone positioning, proven to work in patients who have severe RDS.
Speaker: The key here is to do it for 16 hours or more per day and to do it, I mean, to find a process in your unit where everybody knows exactly how to contribute and do it in a good way.
Speaker: And alternative modes of ventilation either have been proven not to work or there's no data right now.
Speaker: So we can talk about where that fits.
Speaker: I mean, in individual cases, but it's hard to recommend at a widespread range.
Speaker: And then we talked about ECMO that's been around for a while and your takes on eolia and how you utilize it in your patients.
Speaker: And I think that all of those are very applicable to day-to-day practice.
Speaker: And I hope that our audience, I mean,
Speaker: If follows those, we'll link a lot of these studies that you've mentioned in the show notes so people can look at them.
Speaker: And what we like to do at Critical Matters, Bob, is kind of close with a couple of questions that tap on your wisdom but are not related directly to ARDS.
Speaker: Would that be okay?
Speaker: Sure, of course.
Speaker: So the first question relates to books.
Speaker: And is there a book that you have either gifted often or that has really impacted you in terms of how you think about life and critical care in general?
Speaker: Well, I'll be shameless and promote my own critical care textbook published by Springer.
Speaker: So I had the opportunity to create a new format case-based critical care textbook and
Speaker: What we tried to do, I also write for Up to Date.
Speaker: I'm very proud of what I write for Up to Date on mechanical ventilation.
Speaker: But we felt that that was not really contextualized sufficiently.
Speaker: So we made a case-based book.
Speaker: So I gift it.
Speaker: Well, our residents can go through our firewall and download it as a PDF.
Speaker: So I'm not quite giving it for free, but they are getting it for free.
Speaker: It's a PDF available to them.
Speaker: Excellent.
Speaker: So we'll make sure that we link that in the show notes for the book so people can find it.
Speaker: And you mentioned it's case-based, so it's really cases around different pearls in critical care.
Speaker: Is that correct?
Speaker: Absolutely.
Speaker: Every chapter starts with a case and sort of a question with the answer.
Speaker: And then we have the principles of management section.
Speaker: And then the last section is what I dubbed evidence contour, which is to say what are the
Speaker: things that are in evolution and controversial with regard to the management of that particular kind of patient.
Speaker: Excellent.
Speaker: And I think that there's obviously information available everywhere right now, but I think that putting things together around what's practical based on cases and also an appraisal, I mean, of what we know and what we don't know is always going to be very useful for our trainees, but also for our practitioners in real life.
Speaker: So the next question, Bob, is what did you believe to be true in medicine or life that most other people don't believe?
Speaker: Well, I don't know if they don't believe it, but you can tell from our earlier conversation, I'm kind of known a little bit at times to torture my house staff with quotes, and my favorite one is Pasteur, which is, chance favors only the prepared mind.
Speaker: And I think that's true in medicine, and I think that's true in life, particularly in medicine and critical care when you're confronted with a very complex milieu of variables to try to sort out in your brain, and you can't, I guess you can't get to where you want to get with the patient without having a sense of where you want to be, and
Speaker: and looking at things on the fly to manage.
Speaker: And so a chance favors only prepared mine.
Speaker: That's my all time favorite quote.
Speaker: Excellent.
Speaker: And I think that, like you said, I mean, quotes are very powerful in terms of they're easy to remember, but they capture really a lot of wisdom that applies to how we conduct ourselves on a daily basis.
Speaker: And the final question is, what would you want every intensivist who's listening to us to know?
Speaker: Well, I'll go back and quote again, because this is funny, because that Pasteur quote is so eloquent.
Speaker: The quote that I get teased about that is mine, if you want to give the evidence, the interest rather, one quote, it's dry them out and wean them.
Speaker: That's my aphorism.
Speaker: Certainly people accumulate fluid when they're sick and they need to pee it out when they're better.
Speaker: So I sometimes tell people, I only know five things and that's one of them and it's got me this far.
Speaker: So I'm sticking to it.
Speaker: Dry them out and wean them.
Speaker: And I think that that's what patients really want.
Speaker: Wean them and get them home, right?
Speaker: Exactly.
Speaker: And I know that.
Speaker: I think you have actually, I heard you talk about the ABCDF bundle as a bundle of a lifetime.
Speaker: And I think it applies exactly to that concept, right?
Speaker: I mean, in terms of drying up and wean them is part of that and getting people back to real life.
Speaker: So I think that's a great place to stop.
Speaker: Thank you very much for your time.
Speaker: We might talk a little bit more about after SCCM, see what new evidence comes in the direction of neuromuscular blockers and where trials might be coming up in mechanical ventilation.
Speaker: And we look forward to having you back as a guest on Critical Matters.
Speaker: Thank you very much.
Speaker: Well, thank you for having me.
Speaker: Thanks again for listening to Critical Matters.
Speaker: Make sure to subscribe to this podcast on iTunes or Google Play.
