FEMALE SPEAKER: Welcome to Mayo Clinic, COVID-19, Expert Insights and Strategies. The following activity is supported in part by an independent medical education grant from Pfizer Inc., and is in accordance with ACCME guidelines.

OGNJEN GAJIC: Hi, welcome, everyone to METRIC's 2020 virtual critical care conference. It's about the only thing that we'll be doing for the foreseeable future in critical care conferences. And I'm really pleased to have a fantastic panel of speakers on a topic important in these times, so how do you deal with severe hypoxemia? And I'll just try to have you briefly introduce yourself. And I know that one of our panelists from as far from Adelaide, Australia, Dr. Andrew Bersten, has to leave us in a few minutes for clinical duties. Let's start with you, Andrew.

ANDREW BERSTEN: Well, good morning. Good afternoon over there. Yeah, Andrew Bersten from Adelaide, Australia, Flinders Medical Centre. Looking forward to participating in it, thanks, Oggie.

OGNJEN GAJIC: Matt?

MATT SIUBA: Hi, I'm Matt Siuba. I'm a medical intensivist/critical care physician at the Cleveland Clinic in Ohio.

OGNJEN GAJIC: Philippe?

PHILIPPE BAUER: Well, Philippe Bauer, intensivist, Mayo Clinic.

OGNJEN GAJIC: Sarina?

SARINA SAHETYA: Hi, I'm Sarina Sahetya. I am a pulmonary critical care physician at Johns Hopkins in Baltimore, Maryland.

OGNJEN GAJIC: Alice?

ALICE GALLO DE MORAES: Hi, I'm Alice Gallo. I'm an intensivist here at Mayo Clinic in Rochester.

OGNJEN GAJIC: Steve?

STEVEN HOLETS: Hi, Steve Holets. I'm a respiratory therapist here at Mayo Clinic in Rochester.

OGNJEN GAJIC: So the first is starting with [INAUDIBLE] So we're not going to be talking about intubation for shock, for mental state, for un-hypocapnia reasons. And nowadays, obviously, we'd call it a pneumonia epidemic. We're talking about hypoxemia, severe hypoxemic respiratory failure. So what are your thoughts of any specific definition? There is P/F ratio less than 100 if you do the arterial blood gas analysis. But is there anything specific to the definition that we would need to consider at this time? Andrew?

ANDREW BERSTEN: So I think it's got to be more than just blood gas analysis. It's got to be the whole clinical picture, respiratory rate, work of breathing, distress, and the anticipated cause of what you're seeing in front of you. So sometimes it's going to be better to anticipate early. I don't think we want to wait for the patient to be severely hypoxemic before we decide to intubate them, because at that stage, it's a much more dangerous procedure. So I think it's a whole clinical picture, it's clinical assessment. And it's a repeated thing that we always do at the bedside in a continuous process.

OGNJEN GAJIC: Any other thoughts of defining severe hypoxemic respiratory failure?

MATT SIUBA: Yeah, I think one issue that comes to mind is a lot of the definitions that are used-- and there's a number of them-- don't really have a uniform definition of refractory hypoxemia. That'd be really helpful to help us sort of refine the way that we study this. The important factor that often is not taken into consideration is what are we doing to the patient. So somebody might have a P/F ratio less than 100 when they're on 100% FiO2 and 5 of PEEP, and then there might be another patient on 20 of PEEP who is in a very different circumstance. And to compare those two as if they're the same is probably problematic.

OGNJEN GAJIC: Sarina or Alice? Go ahead.

SARINA SAHETYA: Also, just to get into what Matt's saying, also, the time for how long they have been hypoxemic matters to me. So just to take the same example as Matt was saying, someone who has been on a PEEP of 5, and we are going up on the FiO2, for example, for an hour, I would worry more than someone who was on a PEEP of 5, same 80%, and still holding their oxygenation. So to me, the timing of the lens of hypoxemia matters also.

OGNJEN GAJIC: Any other thoughts?

PHILIPPE BAUER: Yeah, I would just question about the use of P/F ratio, especially in spontaneous breathing. P/F ratio of 120, you intubate the patient, and now it's 300. So I think for me--

[INTERPOSING VOICES]

PHILIPPE BAUER: --refractory hypoxemia would be something very simple, like when you get a [INAUDIBLE] oxygen level, let's say, above [INAUDIBLE] with the mean that you have. [INAUDIBLE] the nasal cannula high-flow, closed face mask, [INAUDIBLE] or even mechanical ventilation. P/F ratio is probably the worst in this situation. I mean, it's useful in some circumstances, but we should not focus on the P/F ratio.

OGNJEN GAJIC: OK, let's move forward. So let's start severe hypoxemia even before intubation. Because it's hypoxemic respiratory failure. It's not mental state. You don't have a reason to intubate because of shock or mental state. So it's only severe hypoxemia.

We know that not everyone who is not intubated in hypoxemic respiratory failure actually dies. In [INAUDIBLE] study, in patients we do not intubate, there are quite a bit of them who actually survived, even with a, let's say, a P/F ratio less than 100. So Andrew hinted a little bit to that. But how do you decide that the risk benefit of intubation is on the side of benefit in hypoxemic respiratory failure, compared to maybe waiting, giving some time, giving some, let's say, opioids for work of breathing, doing tender loving care, and trying to see while your intervention-- whatever the intervention-- is tincture of time is working?

So Andrew, while I have you there, maybe you could tell us. Because especially its critical now. We want to avoid these crash intubations, because we don't have PPE equipment. But on the other hand, over the years, my thinking of who actually has to be intubated in these situations has dramatically evolved with more experience with high-flow nasal cannula. Is that something that you can use to kind of have the nurses be happy, et cetera? So go ahead.

ANDREW BERSTEN: So look, I agree. One thing I've learned over the last 20-30 years intensive case is to do less, not to do more. So this the circumstance where-- can you avoid intubation? And even though we're not doing it for mental state, if the patient is calm and co-operative, that's completely different to somebody who is anxious and moving around the bed, maybe using up oxygen with agitation, those sorts of things. So I think that that's very important.

For me, it's the repeated looking at the patient, seeing what I've done, seeing what the response to that is, and coming back to the bedside, trying to be independent of that, and then understanding what the clinical course is. Things are changing rapidly over a short period of time-- maybe not one hour, but over three or four hours-- I think you can anticipate where you might be at the end of that.

Clearly, we all would all like to keep the patient fairly dry during that time. That's not the answer to everything at all, but if there's any hint of fluid overload, then we would be addressing that very early on. And I think using high-flow nasal oxygen-- and in this circumstance, because of the PPE issue, maybe intubating a little earlier than we would have before, because we know that it's going to take a little bit more time to prep up.

OGNJEN GAJIC: Alice?

ALICE GALLO DE MORAES: And Dr. Gajic, I know we're talking pure hypoxemic respiratory failure, we're not going into causes. But I think timing of intubation should take into consideration why the patient is hypoxemic also. And I know we're not supposed to get into causes, but for me, for example, if they're hypoxemic from a massive PEN, their RV is not doing so well, those are the ones that I'm probably going to delay a little bit, try [INAUDIBLE] flow for a little longer to not-- to protect the RV. So I know Matt was like, yeah. So timing to intubation, for me, we have to take into consideration why they are hypoxemic, in my opinion, at least.

OGNJEN GAJIC: Sarina?

SARINA SAHETYA: Yeah, I agree with everything that's been said. The two most important things for me are timing and trajectory. And all of that plays into what's causing it, what's the patient look like? Are they tolerating their hypoxemia, or are they working so hard to breathe that I'm [INAUDIBLE] that they're causing-- they have very high transpulmonary pressures, are causing themselves lung injury. So I think looking at the entire clinical picture is important.

One thing that we tend to see is that the patients who are tolerating high-flow or their non-invasive ventilation tend to declare them themselves within hours. And the people who do well on non-invasive ventilation with their severe hypoxemia-- you will know that relatively quickly with repeated assessments, just like Andrew said. One thing we've started doing in the current era we're practicing in is something I think that's a little bit more popular in Europe.

But we're actually proning some of our spontaneously breathing non-intubated patients on high-flow nasal cannula. Other people have done it. We hadn't done it in our unit until the last one or two days. But I'm shocked at how well people are tolerating that, and we're staving off intubation in that setting.

OGNJEN GAJIC: Wonderful. Any other thoughts before we intubate a patient? So now a patient has not yet been intubated, but is going to be soon. So any other thoughts before that--

[INTERPOSING VOICES]

STEVEN HOLETS: Yeah, I think it's covered well, as far as patient assessment. I think with non-invasive, the use of the [INAUDIBLE] score, and with nasal high-flow, the ROX index may help you have a clearer idea of those patients that are likely to fail.

OGNJEN GAJIC: Wonderful. OK, so let's move forward, because everyone is busy, and we want to get to the end of this.

[INTERPOSING VOICES]

OGNJEN GAJIC: Yes, who is that?

GUSTAVO CORTES: Oh my gosh, I've been trying to speak over the last several minutes, and I can't. This is Gustavo Cortes, how are you?

OGNJEN GAJIC: Hi, Gustavo. Because you were making noise, so we had to mute you. But go ahead.

GUSTAVO CORTES: Noise?

[LAUGHTER]

[INTERPOSING VOICES]

GUSTAVO CORTES: Sorry I am arriving a little bit late to the party, but two comments, one comment about definition of ARDS and respiratory hypoxemia And I think something that is really lacking right now is the presence of pulmonary mechanics into the definition of hypoxemic respiratory failure in ARDS. And I think that will probably help us make a more homogeneous assessment of these patients. And that's why you see so many discrepancies in the individual or personalized definition of refractory hypoxemic respiratory failure.

And to add to the panel about failure of non-invasive-- either positive pressure or high-flow nasal cannula-- the trajectory is important, but also, it's important the interventions that you are doing. So if the patient is not doing well, you're going to wait an hour to see how the therapies you are establishing are going to help the patient or not. If nothing else is done, then you can not expect things to change drastically with high-flow or non-invasive.

The effect of PEEP that you get with high-flow is probably negligible, especially in high respiratory drive, when patients are going to try to breathe through their mouth. And on non-invasive with acute hypoxemic respiratory failure, the augmentation of the tidal volume from [INAUDIBLE] ventilation can be drastically injurious in these patients, especially if they have a significant amount of capillary leak, like we are presuming they have here with COVID-19. So I think being extremely objective and extremely aggressive about moving toward intervention when we are having these-- [INAUDIBLE] strategies is very important, especially in the light of COVID-19 infection.

OGNJEN GAJIC: Wonderful. Thank you very much. I think we could--

GUSTAVO CORTES: I know you were waiting for me to say that.

OGNJEN GAJIC: No, it's wonderful. [INAUDIBLE] to move forward and intubate this patient. So we tried everything, it didn't work. Especially in a constraint with avoiding crash intubation and having to put the protective equipment-- we just gave an informed consent patient that [INAUDIBLE] to take a chance, and we are going with a time limited--

[INTERPOSING VOICES]

OGNJEN GAJIC: --versus mechanical vent--

[INTERPOSING VOICES]

OGNJEN GAJIC: Can you just mute yourself, Gustavo? Just for when we are not talking, try to mute. I will call for the others, so just [INAUDIBLE] OK, so we intubated the patient. The patient is still hypoxemic. Let's go the first [INAUDIBLE]. So what people have done in the past, it was a recruitment maneuver, PEEP and recruitment maneuver that has fallen out of favor with recent studies, like ART and PHARLAP using maximal and prolonged recruitment.

I have some feelings about it, but I want to hear the panel. So Philippe, any thoughts of recruitment? Do you do it? So you just intubated, the patient is still hypoxemic. What are you doing?

PHILIPPE BAUER: Well, I would say it depends on the kind of respiratory failure [INAUDIBLE] with compliance. So it's very easy to basically, adjust to the driving pressure. And that's probably the easiest way at the bedside if [INAUDIBLE] So I never do a full classical [INAUDIBLE] for sure. I think you need to find not the best PEEP, but the best compromise.

Now if you have a patient [INAUDIBLE] and is now intubated, because I've been off the ventilator too late, then you have a stiff lung. It's clear that the lung may not be recruited. But you don't have to do any maneuver. You know very quickly if it's a stiff lung or not. It's very easy. So you can get a lot of information at the bedside by testing the patient. And so that would be it, you know? Is my lung compliant or not, and we need to find the PEEP optimal. So you can get a lot of answers from the bedside.

OGNJEN GAJIC: Steve, how about recruitment?

STEVEN HOLETS: Yeah, I think we all know what the ART trials showed, that super high pressures for super long times-- minutes-- is injurious. Arnaud's study showed that most of the lung can be recruited in 10 seconds. And the idea of going to a high pressure quickly has been shown to be injurious in animal studies. So I think if you're going to do a recruitment maneuver, you need to do it carefully.

So if you're going to do a step-wise recruitment, it should be a small steps, and probably limit total inspiratory pressure 40 or 45. If you're using a pressure control mode, have an inspiratory pressure of 15. That's your driving pressure. But I would be very careful on recruitment maneuvers. Some of these patients seem to respond well to them, but you can't do these big harsh high pressure maneuvers.

OGNJEN GAJIC: Understood. Sarina, PEEP and recruitment [INAUDIBLE]?

SARINA SAHETYA: I feel strongly about this. So I agree with Philippe. We rarely use recruitment maneuvers, or I rarely use recruitment maneuvers in my clinical practice, based on the ART trial, and I think even looking at PHARLAP as well-- the high incidence of cardiac arrhythmias that we're seeing without substantial improvement in other secondary outcomes. So if I am doing a recruitment maneuver, it's primarily in the setting of someone who has had a de-recruitment maneuver-- got disconnected from the ventilator, had a big turn and a long coughing spell, something that I think has triggered them to de-recruit substantially.

And then I'm doing it kind of in the same way that someone just suggested, which is keeping a controlled driving pressure. But I tend to do a very gentle or controlled recruitment maneuver. What I'm primarily doing-- if someone's intubated, and we're thinking about adjusting their PEEP is a test dose of higher PEEP. So rather than doing anything extremely fance, it's just I start them on the ARDSnet table, and then I increased the PEEP by four 4 to 5 centimeters. I watch their oxygenation. I watch their hemodynamics, and I measure their driving pressure.

If everything gets better and they tend to trend in the same direction, then I consider them more PEEP-responsive. And we tend to titrate to driving pressure more than oxygenation in my clinical practice. And you know, there are all these indices out there to measure recruitability, and to measure recruitment to [INAUDIBLE] index. But I think the test dose of PEEP is extremely practical at the bedside, quick, and gets you a lot of the answers that you need to know. Are they PEEP-responsive, or are they not?

OGNJEN GAJIC: OK, so what I'm hearing-- if you do a recruitment maneuver, it's something that certainly is not of high pressures, and it's not for a longer duration, and it has a limited value, and it's relatively equal to a PEEP trial, that after that, you need to adjust it to, let's say, driving pressure, whatever the lung protective inflation is, and the cumulative oxygen saturation response, which would probably capture both the recruitment and a shunt aspect, and any negative effect on hemodynamics, either on RV or the LV. Any other thoughts from those who didn't speak on recruitment before we move to body position, and other ways to improve the hypoxemia? Gustavo, Matt, do you have anything else to add to what has been said about PEEP and recruitment?

MATT SIUBA: No. I think overall, I really-- this is Gustavo. I really do agree with everything that has been said. But I also feel very strong about really using all the tools that you have at the bedside to assess recruitability. And perhaps I'm biased because of the tools I have in my institution here at Mayo. But knowing the best you can at the bedside whether or not your lungs are recruitable is step number one, and should be step number one always before applying PEEP to these patients.

And I do agree. I think in the light of current data about high pressure, high time and recruitment maneuvers, the best protocol will be to see the response of driving pressure to different levels of PEEP, as Dr. Bauer was pointing out. Because I think that's probably the most reliable. And it's basically, the assessment of recruitability.

For a given tidal volume, your resulting driving pressure should reduce if you are recruiting. And that's something that you can do very easily with basically three steps of PEEP. And you can kind of try to adjust to whether or not the patient is PEEP-responsive or not. If you have the possibility to assess a lower inflection point whenever your chest wall is not compromised, that's also a very important tool.

And eventually, I do feel that whenever there is [INAUDIBLE] especially when you have changes in parenchyma and chest wall compliance-- where more advanced tools are still having a role in PEEP titration. And that is the esophageal balloon manometry. In those specific areas of discordance between the chest wall and the lung compliance.

OGNJEN GAJIC: Wonderful. Any other thoughts on best PEEP recruitment before we move forward? OK, so I want to ask the prone queen, as they call her at Mayo. Alice is a fan of proning, and especially, she's doing it [INAUDIBLE] She has developed a really safe protocol with a multidisciplinary team, with nurses and other multi-professional teams. So when we do it, it's very, very [INAUDIBLE] So that is something to take into account.

But Alice, do you believe in some magic effect of proning beyond controlling the oxygenation--

[INTERPOSING VOICES]

OGNJEN GAJIC: Do you prone just because--

[INTERPOSING VOICES]

OGNJEN GAJIC: And can you mute, please, Gustavo? Yes, thank you. So Alice, just to repeat the question. You're doing it, you know how to do it safely. And obviously, if there is no other way to get the improved oxygen saturation and oxygenation in hypoxemic respiratory failure, that's another thing to do.

But let's say you are not there yet. The patient is on 80% oxygen, with prone PEEP that is just right. And the patient has 91% saturation. But the P/F ratio-- if you did the gas-- it is below 150. Do you routinely advocate for using a prone position because it is ARDS and recommendations by these guidelines, or you reserve it for if you cannot get your saturation without that? So what are your thoughts about indications for a prone position?

ALICE GALLO DE MORAES: So my thought will be that you have to make sure that you have all your other things before. So like Gustavo mentioned, that your mechanics are in line, that your vent is adjusted for your patient, not trying to adjust your patient to the vent, that patients are adequately paralyzed and sedated-- let me flip that, sedated, and if needed, paralyzed-- to make sure that they are syncing with the vent before we go to proning, I would say.

And if all of these things are met, then again, like you said, we have a very safe protocol. We don't use the roto-prone. We have been doing it manually since 2015, and we haven't had any complications, no extubations, no line removals. We had one pressure ulcer on the chin in all of these years. So it works for several reasons. The weight of the chest improves V/Q matching, and secretions, also. A lot of times, you prone them, and then the secretions just pour out. So I think it helps in several ways.

OGNJEN GAJIC: I have thoughts on proning with regards to when I was doing it, and when I'm not doing it, and what's the indication.

MATT SIUBA: Yes, I think from my perspective, I agree with what Alice said. I think what sometimes people forget about is PROSEVA and other studies that gave people a 12 to 24 hours of ideal mechanical ventilation before we moved to proning. So I think there at least do that, depending on their body habitus, which Oggie I think will probably support. Body positioning may be more helpful, depending on their sort of body mechanics.

But I agree. In general, I'm pretty quick to prone if they meet the criteria for it. Not only does it have the oxygenation and ventilation benefits, but also, if you're worried about RV function-- I'll talk about the RV for a second-- it may in a way be a right ventricular protection as well. So I think that's another appealing benefit of it, at least in the short term.

OGNJEN GAJIC: Phillipe?

PHILIPPE BAUER: I would say that's my feeling I almost feel that proning is a failure of having missed the ideal timing of intubation, having missed the goal of optimization of mechanical ventilation within two hours, and having missed the rescue within six hours. So this study showing that when you send a patient to ECMO, and they do apply those rules, there's really not too many people that end up getting ECMO.

It seems to me that there are some specific indications for proning. But it's more rare than the French study. I would also agree with the previous speaker that an obese patient can do very well in 25 degrees position, and you don't have to do always proning, but positioning of obese patients, for example, is very important.

OGNJEN GAJIC: So Trendelenburg, reverse Tredelenburg position for obese patients, and postoperative [INAUDIBLE] or abdominal surgery. Steve, any thoughts about proning? When do you suggest? Do you rush to it, or you wait for it? What are the indications for the prone position in your opinion?

STEVEN HOLETS: Well, my personal opinion is the sooner the better. If you've optimized mechanics and PEEP, and you still are in refractory hypoxemia, it's time to prone. Because PEEP are proning work two different ways. And I'll let Gustavo explain that. So go ahead, Gustavo.

OGNJEN GAJIC: Want to unmute, Gustavo?

GUSTAVO CORTES: Oh, thank you, Oggie.

[CHUCKLING]

So yeah, I can talk about proning for hours, but I agree with Steve. I feel that oxygenation is probably 20% of what prone actually does for your patient. So prone is going to homogenize ventilation. And that will basically distribute the energy of ventilation more evenly. And that has been proven by CT scan analysis of aeration.

In addition to that, as I think Matt was pointing out, the production or the promotion of zone 2 and zone 3 were zones that are more homogeneously distributed across the lung produce a reduction in PVR. And that is going to unload your right ventricle. In addition to that, there have been very clear studies, both from the human perspective and animal perspective, that because of the homogeneous ventilation, and kind of homogeneous distribution of energy, you have a lower degree of inflammation for any given mechanical ventilation strategy.

So if any given mechanical ventilation strategy you are applying to your patient is safer when you are in prone position, therefore the sooner, the better it is for your patient to be prone. And I think that has been something that seems to work, based on all the reports that you hear from around the world. And they have both synergic and kind of independent functions, PEEP and proning.

So for the same PEEP, for example, in supine versus front position-- this is something that we have done, both in patients and in the lab with big animals-- the best compliance is achieved when you're in prone position for the same PEEP level. So if you can compare your lung compliance-- and I'm talking about based on esophageal balloon measurements-- your lung compliance is much better when you're in prone position compared to supine. So for all those reasons, I am-- I don't think it's magic. I think it's actually basic physiology, and certainly, a tool that is under-utilized in the ICU.

There are different caveats to it. And there have been different kind of methods to determine at bedside whether or not your patient is going to be prone-responsive. And I think one undescribed method that you can find in a review article from approximately a year ago, is that trying to restrict the upper anterior portion of the chest, and see if under [INAUDIBLE] ventilation or volume control ventilation, your pressure or your plateau decreases. And that is going to basically mimic the effect of prone position, and potentially predict whether or not the pulmonary mechanics and oxygenation will improve in prone position.

So I couldn't advocate more for early proning, especially in severe ARDS. But you can always make the case it should be considered in patients with moderate ARDS. Like, it's recommended by PROSEVA trial and other different studies. And I think something that we can check quickly after proning is the behavior of the CO2. And that has been proven to actually show from the physiologic perspective whether or not your patient is responding to prone position.

OGNJEN GAJIC: Thank you, Gustavo. If you could [INAUDIBLE] and mute yourself while you're not talking, because of the noise that comes from the phone. I want to ask something. So what I've heard is that there are obvious benefits of prone position on both shunt reduction, and potentially, for RV function. And in a case of refractory hypoxemia, that means after your PEEP recruitment and optimizing ventilator strategies, if you are still unable to achieve it, there is no reason to delay prone positioning.

And I have no problem with that. I'm a fan of proning, especially before you get intubated, as Sarina said, and I've used it quite a bit for people with a high-flow nasal cannula and even non-invasive ventilation with some success. But I want to ask Sarina-- there are some side effects of prone position when you're intubated, not when you're not. Because once intubated, prone position often means some some way of heavy sedation. You're not going to be talking.

So how long do we need? So I know we did something. But do we do 16 hours every day for a week, and lose all our ground [INAUDIBLE] that we were getting by what we have achieved in the last 10 years learning [INAUDIBLE] mobilization, less sedation [INAUDIBLE] spontaneous brain [INAUDIBLE]. So when do you-- if you choose [INAUDIBLE] prone and when you stop proning? Because in our institution, our average length of proning is one night. So that's what Alice has found, not like a five, or seven, or two weeks. So what are you guys doing over there at Hopkins?

SARINA SAHETYA: So interesting question. I think, like everyone else on your call, I am a fan of early and aggressive proning. I will say that once we get someone in the prone position, we keep them prone for a prolonged period of time. So our minimum number of hours is 16 hours. But if we haven't made substantial progress on their oxygenation or respiratory mechanics, we will maintain them [INAUDIBLE] based on some of the subanalyses out of the the PROSEVA study suggesting that it wasn't just oxygenation response that predicted a mortality benefit with proning. And so we'll continue to keep them prone for two to three or four days.

We'll do trials of supination. If they de-compensate, we turn them back over. And I think, as you guys know, our ICU, led by Dale Needham, is a big proponent of light sedation or no sedation and early mobility. So we recognize that it's a compromise between the two. But we try to balance that by still having occupational therapy work with splinting positioning, and have the nurses work on that, as well as making sure that we rarely use paralytics with proning now to avoid the double effect of paralytic's heavy sedation and pronation contributing to neuromuscular weakness.

OGNJEN GAJIC: Any other thoughts? So when do you stop proning, Philippe. If you do prone someone, [INAUDIBLE] do you have other ideas of how long you're going to be proning. Not in a day. I know, I understand when you first prone, it's going to be the whole night, so up to 16 hours. But let's say when do you start when it is over, and you go back to your other things?

PHILIPPE BAUER: Well, that's a difficult question. Because I think it depends on the context. And once again, the current experience with COVID, especially the Italians, says that those folks are PEEP-sensitive for a variable period of time. So it really depends on the type of ARDS we are dealing with, and [INAUDIBLE] the risk [INAUDIBLE]. But definitively, at least for four days. But it's a tough question. I think it depends on the patient, I would say.

MATT SIUBA: Yeah, I agree with that. I think it seems like the majority of people who respond, not only respond quickly, but in our experience, at least, don't need more than a couple days of this at the most. I think I've gotten away with one to two days very often. And to say how do you how do you pick your end point for that is challenging. But I think if your mechanics are sufficiently improved, and you don't backslide significantly when you re-supinate them, like Sarina said, then we're a little bit more apt to leave them face up, for a lot of the reasons that Oggie mentioned in terms of sedation and mobility, and things like that.

OGNJEN GAJIC: Alice, Steve, and Gustavo, any other thoughts? Unmute yourself, Alice, sorry. Anything else [INAUDIBLE]?

ALICE GALLO DE MORAES: I agree with everybody. I believe if you're committed to proning, you have to commit to at least 14-16 hours. That would be my only take on all of that.

OGNJEN GAJIC: But I mean, with regards to average number of days that you will continue that, when do you decide to stop proning?

ALICE GALLO DE MORAES: So I agree with everybody. I would say a couple of days, and I would keep proning until the oxygenation improves on prone position.

OGNJEN GAJIC: Gustavo?

ALICE GALLO DE MORAES: But I would talk to my ECMO colleagues as soon as I prone them for the first time, just to make sure that they are aware.

OGNJEN GAJIC: That'll be a good segue to another rescue therapy that we have. Gustavo, can you unmute yourself so we can hear if you have any other thoughts?

GUSTAVO CORTES: Oh? OK, now I can talk.

OGNJEN GAJIC: Yeah, go ahead.

GUSTAVO CORTES: Yeah, definitely. So I think it's a very fluid patient. And so by protocol, I think what we have decided here at Mayo is you prone the patient today. If the patient doesn't show signs of improvement the day of proning, consultation with ECMO should happen. If the patient shows signs of improvement while in prone, and de-recruits or de-oxygenates while in supine while in the first turn, it means that pulmonary mechanics are still viable and fluid, which means the benefit of proning should continue to be established.

And I think it's something that you have to play by ear on a daily basis, and that additional proning-- there is not a lot of downside to it, as far as it's done appropriately. If there is a worsening of other organs, if there is arrhythmias, if there is worsening heart function, the decision to basically [INAUDIBLE] scopes of care has to take place, then I think those decisions affect your decision of continued proning rather than the other way around. I don't see this as an ECMO equivalent, or that you have to necessarily remove this drastic therapy for any given reason without looking at the context of the patient.

But I don't think there is an established number of days. I'm pretty sure the longer the time of proning and the number of days, the more prone the patient is to fail, and to potentially die. But that doesn't mean that the proning itself is failing. It might be for many other reasons.

OGNJEN GAJIC: Thank you. I just have one question before we [INAUDIBLE] prone, and go consult ECMO. What I've found in my experience is that many times, we don't necessarily distinguish between the contribution of potentially shock or venous admixture, or mixed venous oxygen saturation to refractory hypoxemia, where actually simply, norepinephrine or transfusion, or sometimes even a fluid bolus or something that would improve the RV dysfunction is more important. Because it will improve the SvO2 to [INAUDIBLE] rather than something that [INAUDIBLE] shunt. And some of these patients may actually be impaired by staying to prone position because of the risks of not being able to assess, let's say, with the echocardiography of the heart, and other things.

So any thoughts on contribution of venous admixture and shock? How often does it happen? Should we be worried about it? And kind of thinking that there is both heart and lung in the chest. We're not only taking the heart. So any thoughts on that?

GUSTAVO CORTES: Yeah, I do, I do have thoughts on that. Oggie, do you hear me?

OGNJEN GAJIC: Yes, go ahead.

GUSTAVO CORTES: Yeah, so definitely. So I think that the difficult part about that is that there is not basically a very well standardized way to inform that decision or make that assessment. I do suspect in many patients [INAUDIBLE] be hypotensive to have a significant contribution of the shock component. And I would say rather reduce the oxygen delivery due to cardiovascular, either contractivity or volume component. They don't necessarily need to be hypotensive.

And you can use your SvO2, as we have talked before, Oggie. But it has to be well measured. The closer to the right atrium, the better. But I do suspect it when there are lot of fluctuations of oxygenation in these patients with stable steadiness. And I have to say that that fluctuation tends to happen more when you are in supine than when you are in prone. And I think that makes sense from the physiologic perspective. And it has to do basically with the better V/Q matching that you have in prone position.

And therefore, the disparity of either contractivity issues or relative hypovolemia without hypotension may help the oxygenation. But I do think that is an important assessment. It kind of contemplates a complex physiology. I think we don't have a standardized or user-friendly way to determine that at the bedside. But it's something that we all should be aware. And also, not to go in the opposite direction, which is basically, giving fluids, and increasing PEEP, because the oxygenation is getting worse. And it's basically because of pulmonary vascular congestion.

OGNJEN GAJIC: The vicious cycle. Any other thoughts on contribution of venous admixture and shock to refractory hypoxemia?

[INTERPOSING VOICES]

OGNJEN GAJIC: Philippe?

MATT SIUBA: Go ahead.

PHILIPPE BAUER: Yeah, sorry. So I mean, it's really important to realize once again, that if you have a good arterial blood pressure [INAUDIBLE] we may still have some tissue hypoperfusion. So it's really important. So we have, of course, the Svo2, as we discussed, but also the [INAUDIBLE] too-- and I didn't mention the lactate. So don't be fooled by a normal or [INAUDIBLE] blood pressure, where the ongoing inflammation or, let's say, hypoperfusion contributes to the ongoing lung injury, in spite of some good mechanical parameters. So I think it's fallen on them to make sure that there's no evidence that you can detect of ongoing tissue hypertrophy.

OGNJEN GAJIC: So one thing that fits here-- and somehow, we may [INAUDIBLE] we were talking about not trying to prone them. Selena said that they are not paralyzing the patients as much. But obviously, there is a role for paralytic in both of these issues. One is to assure ventilator synchrony, and the other one is to decrease oxygen consumption. That would obviously improve venous admixture. So any thoughts of use of paralytic, intermittent versus continuous, from any one of you?

SARINA SAHETYA: Yeah, so we have moved away-- or, I shouldn't say we moved away, because we never did it routinely for all patients who presented with moderate to severe ARDS, even after ACURASYS. We primarily did it when we had the two suspicions you had, of ventilator dyssynchrony, or a concern that the oxygen consumption from their respiratory muscles was contributing to worsening shock, and the problems we've all discussed previously. We've continued to apply that in the post ROSE era as well.

I think everyone can agree we shouldn't be using it routinely. But when we use it, we should have a goal in mind. And we tend to bolus first, evaluate response to it, and evaluate what happens when the paralytic wears off. And if at some point, they need a drip, we don't hesitate to start it, versus just give repeated bolus dosing, depending on how they're doing.

But I've also found that the more you sit at the bedside, and you are trying to adjust the ventilator based on what type of dyssynchrony that they're having, you can optimize their ventilator-patient interaction without using paralytics, or even deep sedation in many cases. And it alleviates the need for that. So I think our first line is, can you play with the ventilator, optimize it for the patient so that they're not having significant dyssynchrony? And then if you are still in a difficult situation, or you think that their shock is related to the oxygen consumption from how hard they're working, then bolus dosing and reassessment.

OGNJEN GAJIC: I'm sure Steve would agree-- and Gustavo-- about trying to fix the ventilator before fixing the patient. Any thoughts on that from Steve or Gustavo?

STEVEN HOLETS: Yeah, I agree totally that the first step should be to try to fix the ventilator. Try different modes and stuff. We're mainly a volume mode. And I think because we like to keep it simple, it's very easy to see dyssynchrony on a volume mode and harmful spontaneous breathing. It's very sometimes difficult to understand it on a spontaneous mode. So I think that's one thing you need to watch out for.

You can't let them breathe huge volumes and generate huge pressures. And we've seen and experienced-- and sometimes, they don't understand that if you have somebody who had a driving pressure of 15 with a tidal volume of 400, and then you allow them to breathe spontaneously on a pressure support, and tidal volume is 800, well, their driving pressure has to be double of what it was. And that's if they're on the linear portion of the compliance PV curve.

So I think it's very important to watch out for too much spontaneous breathing. And the other thing is when you put them on ECMO, Gattinoni showed that you can't-- in severe hypoxic failure on ECMO-- half the patients with ARDS on ECMO-- they could not control their breathing, even though they can control their blood gases.

OGNJEN GAJIC: Excellent points. I think we should-- because we have a five more minutes. So we just have to move to ECMO [INAUDIBLE] because people [INAUDIBLE]. So VV-ECMO-- I'm not talking about VA-ECMO, I've used that. That can save a life if your heart is not moving. But VV-ECMO I have never yet had to use myself in my practice. Doesn't mean that it's not going to happen, but I haven't. So what would be the indication why to proceed with VV-ECMO after all that we have done?

STEVEN HOLETS: Only after you've done everything we've talked about, proning and everything.

OGNJEN GAJIC: Alice?

ALICE GALLO DE MORAES: Well, I learned everything I know from Steve, so I completely agree with him. I say, if you've tried everything, we have this beautiful refractory hypoxemic protocol, like Gustavo said, I would say go through the thought process in your head several times. And once you've said, yes, I've done everything, then ECMO.

OGNJEN GAJIC: Sarina?

SARINA SAHETYA: So we don't use ECMO frequently for severe hypoxemia or respiratory failure. And again, separating it from VA-ECMO mode, which we do use frequently. I've referred maybe-- I can count on one hand the number of patients I've referred for VV-ECMO. With the steps that we've discussed up until this point, it really hasn't been a strong need to refer people to VV-ECMO. And also stepping back and looking at what else is going on with the patients, in terms of the patient who's already in multi-organ system failure is going to remain in multi-organ system failure, even on ECMO.

And I think we've also instituted kind of early and aggressive conversations-- I shouldn't say aggressive-- but early and honest conversations with the family about what the chances of surviving on ECMO or without ECMO tend to be. So I think [INAUDIBLE] coming out, there probably is a role for ECMO in some situations. Based on some of the sub-analyses I've seen, I would likely tend to refer people earlier for those who can't tolerate lung-protective ventilation. We routinely get down to 4 cc's or even lower on volume control without ECMO, and tolerating some permissive hypocapnia and respiratory acidosis.

So really, ultra lung protective ventilation. If they can't tolerate ultra long protective ventilation, that might be someone I send for ECMO. Otherwise, it really has been the rare patient who comes in with severe refractory hypoxemia, doesn't respond to proning, doesn't respond to any of the maneuvers we've talked about, and has no other organ system dysfunction.

OGNJEN GAJIC: Wonderful. Philippe or Gustavo, any thoughts?

PHILIPPE BAUER: Yes, so I would just simply mention a case that we had a few months ago, where the patient came [INAUDIBLE] hypoxemia ECMO, and we extubated the patient on ECMO the next day. So this is not a prime time, but the technology is here. And I was very fascinated to see a patient go below 50, and really, everything has been [INAUDIBLE] extubated the next day. I know that Steve mentioned that [INAUDIBLE]

But I think that that's really interesting. So that may not be applicable for infectious [INAUDIBLE] so those things, or maybe we can [INAUDIBLE] people have a high P/F exacerbation, for example. But I would mention this technique, because I was really impressed from paralyzed proning [INAUDIBLE] to extubating the next day after ECMO. Something to keep an eye on. That's going to be probably interesting in the future.

OGNJEN GAJIC: Gustavo?

GUSTAVO CORTES: Yeah, yeah, so I think I agree with Steve, first of all, about having done all the systematic approach to refractory hypoxemia. I want to believe that-- so I don't think anybody needs to be convinced that when you have VV-ECMO on board, of course, you can reduce your settings to the minimum on a mechanical ventilator. There have been studies [INAUDIBLE] animals and some randomized clinical trials performed by [INAUDIBLE] and colleagues in the UK looking at ultra protective mechanical ventilation without VA-ECMO.

And it's actually quite deleterious to the RV function of the patient. And it's basically because-- I mean, at 4 mL per kilo, you're already very close to approach your reserve volume. And therefore, you're completely almost to the left of your PVR to lung volume relationship. So the lower the tidal volumes, the harder it's going to be for your right ventricle to work, and to generate the cardiac help you need to have gas exchange. So definitely extending a more aggressive lung protective mechanical ventilation [INAUDIBLE] hypercapnia has been proven to serve well up to a certain point.

But the more you approach the patient with a reserve volume with ultra low tidal volumes, the harder it's going to be for your RV to function. So I think that's something that everybody has to keep in mind. And that's one of the reasons why we activate ECMO call really early on the protocol. As I mentioned, if proning is not effective immediately during the first proning trial, the ECMO team should be notified. Because the chances are this patient is not going to be prone-responsive.