SCCM Resource Library-Pro/Con Debate: Should We Allow Spontaneous Breathing in ARDS?

Video Transcription

I am Rahul Manchal. I am a practicing intensivist at the Medical College of Wisconsin. And we have two wonderful speakers for a great session,  Should We Allow Spontaneous Breathing in ARDS? So without much delay, I am going to get started. Our first speaker, who is going to take the pro side, is Dr. John Marini.  Dr. Marini is professor of medicine at the University of Minnesota and probably needs no introduction, just a giant in the field of mechanical ventilation.  So Dr. Marini, we are happy to have you and welcome. And I look forward to listening to your talk.  Thank you, Rahul. It's a pleasure to be here this morning. It's a beautiful theater. Not too many people in the audience so far, but that's fine.  The topic is allow spontaneous breathing, basically, as we talk about ARDS, and my stance is absolutely.  I have no disclosures to really occupy your attention.  I do have a new book that some of the illustrations are taken from.  I'm gonna tell you one half of a picture, because I think as we all can agree, it depends on the status of the patient, whether or not spontaneous breathing should be encouraged.  My side of it is yes.  Now, the goals of mechanical ventilation  are to reduce the work of breathing, to relieve discomfort, offer cardiovascular support, and compensate for metabolic acidosis, and to improve gas exchange.  Allowing anesthesia, pain relief, and deep sedation is another goal, especially for procedures.  Adjust PA, CO2, and pH in special circumstances, for example, head injury,  and avoid the consequences of ventilation. These are our goals, and you have to ask whether or not spontaneous breathing is the way we should get to them.  There are two options that we're discussing. Allow effort, we'll talk about that as active, not necessarily fully spontaneous, but active,  and prohibit effort, passive circumstances.  Now, I think you must understand that I am not a fool, although sometimes I sound like one,  but sustained dyspnea, pain, or agitation is not something to be tolerated, and it's not under discussion this morning.  I'm assuming we're not gonna allow a patient to suffer.  Now, first of all, the physiology of passive versus spontaneous inflation is quite noticeably different. For the same lung size, you can inflate it  with a high airway pressure and minimal pleural pressure, so that the trans-lung pressure is 20 centimeters of water,  and during active inspiration, only 10 centimeters of water airway pressure and a minus 10 outside would do the same thing.  The main point here is that you can't look at airway pressure alone and determine what the effects are going to be,  and that is an important theme throughout critical care, not just for this topic of discussion this morning.  This is a little bit busy, but I think it's worth going through. As we proceed from active breathing to passive breathing, here fully spontaneous, partial ventilatory support,  fully controlled ventilatory support, the magnitude of effect on the vertical axis, the work of breathing, of course, decreases as we assume fully controlled ventilation.  The pleural pressure gently rises, the venous return gently falls, ventilation perfusion matching, suffers a mild deterioration, and left ventricular afterload in a patient  who has got congestive heart failure may fall impressively under fully controlled circumstances.  One important difference between passive breathing and active breathing is that in active breathing, we simulate what is depicted here in a paper by Much a number of years ago.  When you're looking at the fluctuation of frequency and tidal volume,  in spontaneous breathing, it's quite chaotic. It's not chaotic necessarily, but it is variable, whereas in passive control of ventilation, we typically use a monotonous pattern,  and that has consequences that many of us don't concern ourselves with.  Biologically variable or naturally noisy mechanical ventilation recruits at a lictatic lung, both in normal individuals and in ARDS and can improve gas exchange.  We've known for many years, almost 50 years since the mid 1970s, that spontaneous breathing better matches ventilation to perfusion. In the horizontal position, this being supine,  this being posterior, this being anterior, the ventilation perfusion matching is directed toward the dependent zones during spontaneous breathing.  The opposite tends to occur as we paralyze the patient.  The ventilation and distension increase, the perfusion goes elsewhere, and the ventilation to perfusion ratio deteriorates. The point is that natural spontaneous breathing  better matches ventilation to perfusion.  Spontaneous breathing evens the distribution of regional strains.  On this depiction by EIT, non-dependent regions and dependent regions distribute differently whether we're under volume-controlled ventilation or spontaneous breathing.  So the distribution of regional strains, if you're interested in a ventilator-induced lung injury, is probably better in spontaneous breathing,  and we'll come back to that as I proceed.  So then you might say, with all these advantages, why not allow spontaneous breathing?  Well, now we're all obsessed about applying lung protective ventilation and even ultra-protective ventilation.  We want to avoid dyssynchrony. Oh, my God, that's a bad thing.  Improve oxygenation and ventilation perfusion matching, reduce oxygen demand, and avoid the dreaded ventilator-induced diaphragmatic dysfunction because that can happen  during spontaneous breathing, as we've been told, and I'll argue against that in a moment.  We have been dealing with, I think, imaginary problems.  Some of you may know the Stay Puft Marshmallow Man from Ghostbusters. I love that movie.  I consider the concepts of pendaluft,  PCILI, patient self-induced lung injury, and dyssynchrony to be a tempest in a teacup.  Now, it's been shown in beautiful studies that spontaneous effort does cause occult pendaluft during mechanical ventilation. In green here is the pendaluft  proceeding from spontaneous vigorous breathing to more passive breathing, and you see it diminishes. In other words, the shift of gas from one section of the lung to another.  This was initially linked to all kinds of trouble during mechanical ventilation. In other words, lung injury related to the dreaded pendaluft effect,  and it's been said that there has been regional trouble in the dependent zone as dependent zones take in gas that's distributed from over-distended zones  in the process of pendaluft.  The pleural pressure regionally in spontaneous breathing may be quite different than is the global esophageal pressure  determined pleural pressure measurement.  So, that's thought to be the potential wrecking ball  of spontaneous breathing during vigorous efforts.  Well, I have to say as an old man, why don't we ever learn? You know, we have these concepts come up and then get smashed down as we learn more about the process,  and I think this is one of them.  High positive index pleural pressure renders spontaneous effort non-injurious from the same authors who said that it was injurious,  and so if we use higher levels of PEEP, we don't see the yellow inflammation in the zones that were affected before,  high PEEP being in the range of 12 to 15 centimeters of water.  The idea here is that you have more inflammation by PET scanning with spontaneous effort at lower PEEP, and when you increase the PEEP to a certain level,  you generally more than 10 centimeters of water, then you have less inflammation. I don't want to go into the details, but the idea here is that it can be overcome,  and in fact, in this very recent paper, February 2020,  the same authors say high levels of PEEP seem to enable patients with moderate to severe ARDS to preserve safe spontaneous breathing during mechanical ventilation  considering recent experimental studies and clinical studies,  and I think that's a nicely put statement.  There has been a breaking data wave that contradicts the idea of PCILI.  I won't go through these in detail. I'll just read you the titles. Higher levels of spontaneous breathing, higher levels of spontaneous breathing induce lung recruitment  and reduce global stress strain in experimental lung injury.  It's from Gouldner and colleagues about five years ago.  Increased effort during partial ventilatory support is not associated with lung damage in experimental acute lung injury. Here we're going from pressure control  to pressure support at a high level to pressure support at a low level, increasing spontaneous efforts here, and the diffuse lung injury scores  are no different or even improved with spontaneous efforts.  What about the dreaded asynchrony factor?  This is from an old slide of mine.  Well, I get really annoyed when people talk about repetitive double triggering, and that accounts for why this trial didn't work.  If I saw something like this clinically and did not respond or didn't have my alarm set appropriately, I should be thrown out of the ICU.  Seriously, how are the alarms set under these circumstances? Tidal volume can go up, but if pressure doesn't go up, the danger does not go up.  Reverse triggering with making high stretch on the lung? Well, there's no big difference in tidal volume in most of these studies.  And double cycling during mechanical ventilation? It's not rare, but it's not repetitive double cycling is not very usual. And if it's present, it's often reverse triggered,  as Luis Blanc and his colleagues showed before. And reverse triggering happens in rather deeply sedated patients.  So I'm not worried about double triggering nearly so much as many of my published colleagues are.  Moreover, spontaneous modes, certain types do not promote double triggering or excessive stretch. Those include proportional assist ventilation  and neurally adjusted ventilatory assist.  Here's a nice paper from Georgopoulos and colleagues, Vapariti recently published,  that shows basically you let patients have proportional assist ventilation and driving pressure is not a big problem. The presence of sustained high driving pressure  was not associated with high tidal volume, but occurred almost exclusively when compliance was below 30 milliliters per centimeter of water. The driving pressure is high, perhaps,  but it's due to low compliance and not overstretching.  That's an important principle, not only here, but in other considerations of ventilator induced lung injury.  Assisted ventilation in patients with acute respiratory distress syndrome,  basically, in patients with mild to severe ARDS, there is not a severe problem.  And I'll go on to just reiterate the same theme that's coming over and over again. Airway pressure release ventilation prevents ventilator induced lung injury in normal lungs.  It's in comparison to control, it's considerably better.  APRV studied against low tidal volume ventilation. In this study from China by Zhu and colleagues, published in Intensive Care Medicine, shows a striking advantage for APRV,  which allows spontaneous breathing. Low tidal volume ventilation enforced sometimes with deep sedation and paralysis doesn't work nearly so well.  Now, all of us, I think, are familiar with the ROSE trial, which compared paralysis with non-paralysis and found absolutely no difference in the two studied groups,  recently published in the New England Journal of Medicine this year. If strong effort were to cause ventilator induced lung injury and ventilator induced diaphragmatic dysfunction,  we should have seen some signal there.  You don't know, you can rationalize why it did or did not happen.  But if paralysis was good  and vigorous spontaneous breathing was bad, something should have shown up.  Again, we are confronted with the idea that one of the concepts that we've been working with may not be quite so dangerous as the literature has suggested.  This is another one, ventilator induced diaphragmatic dysfunction. On this slide, there are three forms, really.  It's low levels of spontaneous ventilation and vigorous levels of spontaneous ventilation.  Basically saying that these two are theoretically  a problem clinically. They really are not.  This one is. If you don't have spontaneous breathing, you're in trouble.  In fact, the diaphragm being a continuously active muscle, if you stop its action by preventing spontaneous breathing, it deteriorates its function.  The diaphragm requires activation to maintain its bulk, strength and endurance.  Sustained deep sedation and relaxants lead to weakness, sustained critical illness, invasive ventilation, and it's definitely something to be avoided.  Eliminating spontaneous efforts does have predictable consequences in disrupting the sarcomeres of the diaphragm.  This has been shown many, many times in conferences. Reduced fiber bulk with controlled ventilation after 18 hours in controlled study and case studies,  you can see clearly there's a difference in the bulk of the sarcomeres that make up the diaphragmatic muscle.  Now, only a little effort, spontaneous effort, helps assist control mechanical ventilation of tiny weights ventilator-induced diaphragmatic dysfunction.  Catherine Sassoon, 15 years ago in the Blue Journal, demonstrated that to us.  Controlled mechanical ventilation, assist control and control. And you can see that assist control is much closer to control. In other words, spontaneous than is paralyzed control.  Here is a more recent study. Lung protective ventilation that we love to use worsens ventilator-induced diaphragmatic atrophy and weakness. And enforcing lung protection  may be a trade-off with the diaphragm.  This is by Zhu and colleagues published very recently, 2020.  Now, we found out 12 years ago that if patients are allowed to spontaneously breathe and given analgesia only without sedation,  no sedation patients came off the ventilator and were out of the ICU earlier than were patients who received sedation. The point being here that no sedation at all  is necessary in many of our patients.  This is a paper that was published yesterday.  Light and or interrupted sedation may be okay.  This was published in the New England Journal of Medicine. There's a nice editorial by Claude Garan,  basically stating that non-sedation or light sedation is okay. It does not seem to interfere with the progress of the patient in the ICU.  And this, I think, is a very convincing study from the Lung Safe Investigation, concluding that spontaneous breathing in early, early acute respiratory distress syndrome  has the following conclusion. Spontaneous breathing is common in patients with acute respiratory distress syndrome during their first 48 hours of mechanical ventilation.  Spontaneous breathing is not associated with worse outcomes and may hasten liberation from the ventilator  and from the ICU, possibly because of diaphragmatic,  maintained diaphragmatic strength or endurance, maybe for other reasons, but certainly not destructively.  These are from that paper and the familiar Kaplan-Meier curves that were generated. And this was just published in 2019  from a consortium of investigators.  So in conclusion, why not allow spontaneous breathing? We've heard that we should apply lung protective ventilation or even ultra-protective ventilation, which may have a trade-off.  Avoid dyssynchrony. Well, that's never been shown directly to relate to ventilator-induced lung injury. It has been associated with problems,  but maybe problems that relate to intensity of sedation. Improve oxygenation and ventilation perfusion. Well, only if you reduce oxygen demand  and avoid ventilator-induced diaphragmatic dysfunction.  I suspect not. Spontaneous breathing probably doesn't cause ventilator-induced diaphragmatic dysfunction in most patients. So I think we ought to recede from our imagination  on all of those points except one,  reducing oxygen demand, which is a very powerful reason to consider blocking spontaneous effort in the early phase of ARDS. And I'm sure that Ulrich will talk much more about that.  So I must say that there is a right time and a right intensity probably for eliminating spontaneous efforts. And I said so. Early phase of lung protective ventilation,  a place for paralytics, question mark.  And that's the maybe part. And that was in 2006 in an editorial in Critical Care Medicine.  So I know I've gone through things pretty quickly, but I hope convincingly to tell you that spontaneous breathing is really your friend in almost all circumstances  and at almost all the time that we take care of patients in the intensive care unit. But there is another side, and you're going to hear it from Ulrich.  So thank you for your attention.  Thank you John. So we'll move on to our next talk and then we will invite both speakers up to the podium and have a panel discussion and take questions from  the audience. So our next speaker is Dr. Ulrich Schmidt. He is professor of anesthesiology at the University of California at San Diego and he is going  to be taking the con side or maybe we don't know side. Welcome and I look forward to hearing your talk.  So good morning. So I have a hard task right now because you heard all this good stuff about spontaneous breathing and you will hear at the end as most of  the stuff I agree with John. I bring this up right away but let's and I have nothing. I have no conflicts of interest for this talk. Here we go. As a chairman  I bring the ladies of justice up and not Janos. A similar concept should we do it yes or no. So in a trial I asked the experts first and the experts Dr.  Brochard said obviously ventilator-induced lung injury exists in spontaneous breathing patients with acute respiratory failure. He has to say  it because he came up with a term in 2017. So not a really good witness for me right now. Ventilator-induced lung injury does not exist. Okay Dr. Antonelli  and then Dr. Cardinone said I don't know what to say. So this means I don't really have really good experts for myself. We are still where we are before.  So what I want to talk a little bit about evidence for self-induced lung injury. John told you all why it doesn't exist. I will maybe put a little doubt in  your brain today. Showing some animal models and also some patient data and it's a very end suggestion when really to favor spontaneous ventilation. And I  start with a really old paper from 1988. This is a sheep and they put, they got some acid in the foramen magnum and then they started breathing really hard.  Hyperventilation got really high tidal volumes. And then she also thought is it only the acid? So they also gave some CO2. And she and another  group what does CO2 by itself do and looked. And when you look at this on the left sides are our animals who received acid in the foramen. And here you see  that the tidal volumes are much higher than the tidal volumes of the animals on the right side who did not have injections. The PaO2 in the beginning was  also about the same. However, 24 hours later the animals who had from their hyperventilation supposedly had a much lower PaO2. And I don't have  slides on the pathology in this paper but there were much higher severe lesions in the lung in these animals who were hyperventilated with much higher tidal  volumes. So spontaneous ventilation in the sheep seems to exist. Now when we go in a rabbit model it's by Dr. Yoshida. He did lung lavage models and he saw that  when you spontaneously ventilate it's you get in a severe ARDS. On the left side much higher transpulmonary pressures than in less severe ARDS. We will get to  this at the end of my discussions again. This point is important. Similarly in severe lung injury when you don't spontaneously breathe it looks not very  nice your histology. However, if you spontaneously breathe it becomes much worse, much more infiltrates. The lung looks like an ARDS lung. It becomes worse  and worse over time. So this is in rabbits. So what is really the effect of spontaneous and mechanical breathing? Does it really make a difference? Sean  showed us some data but for what we measure whether it's a positive pressure or negative pressure measured by an esophageal balloon makes not much  difference in our transalveolar pressure or transpulmonary pressure. It's about the same and obviously our tidal volumes don't care so much whether you breathe in,  whether you get a breath delivered or whether you spontaneously breathe. This is shown here in this paper. When in full mechanical ventilation you get positive  esophageal pressures. John showed this also. And when you spontaneously breathe harder and harder your esophageal pressures become more and more negative.  However, the overall pressures seen on the right side stay about the same. You replace one with the other, all things being equal. Similar here. This gives us  in a different paper but it's the same message. Flow stays the same. Here you get on the, if you have positive pressures, you have a negative pressure and here like  in pressure support where you have both, so you have some negative components and positive component. So esophageal pressures go up in positive, in positive  pressure ventilation. Here when you spontaneously breathe, very deep negative inspiratory pressures, a little bit less pressure support. However, our plural,  our transpulmonary pressures are about the same in all three methods of ventilating. And the volume is about the same as well. So what's a big fuss? Now  back to patients. So this was a patient, this is a case report in a respiratory care 2010. And this shows how big my evidence is that I have to start with  case reports. John will point out quite rightfully, however, this patient was on relatively high pressure support ventilation for a while, got better, and  suddenly this pneumo-medistanium came up and the patient died. And a very similar diagram that John showed us before, but again it brings up the point.  We have positive pressure down in the middle and here you have a little bit of a peep here. So what comes in here is you give about a pressure of 20, you have a peep of 5,  you lose a little bit down there to resistance. At the end you may have some 20 centimeters down here, around plus the peep, so maybe 28  centimeters of water. And now you breathe, this is added here, your plural pressures, and then you can get a relatively high transpulmonary pressures  by spontaneously breathing on top of it. And you have to be careful with it. And this has been shown in a patient on ECMO again, where the volume was about 4 cc's  in the volume control group, in the pressure support group 6.5 cc's, so not really high. The peep was the same, the peep pressure obviously on our ventilator  was much lower here in pressure support than here on volume control. However, the delta esophageal pressure was here positive 2 and here  minus 35, which gives us this really high, most likely injurious, transpulmonary pressures. So we can tear our lungs apart by ourselves quite  successfully. But one thing is our measurements at the bad side are not really great. Because where do we measure? We measure either at the end of the  endotracheal tube and then we look at our ventilator and see some graphs here and look how our peep pressures are, our plateau pressures are, and we can  measure our peep and we can look at some flow diagrams. Or in addition, if we are mainly in Boston, we put esophageal balloons down and we will measure  how our esophageal balloon settings are. But, and it's important, our lung is not the same all over and it's millions of millions of little alveoli. And to  make the idea that our pressure, which we measure at the end of our endotracheal tube, is exactly the same for all of our alveoli, it's a little naive.  We don't have much better. It's the same as our esophageal balloon. When you look at an ARDS lung here, it's not homogeneous. It's very heterogeneous. So when my  balloon sits here, I measure pressure at some point in my lung, not in all points. So how do I know that my dependent areas is the same transpulmonary pressure  than in some other part of the lungs? I think we have no idea what we are measuring and we hope that the sum of what we are seeing gives us a non-insurious  model. Unfortunately, it's only a hope. So now, John showed you this one. There are two outcomes studies on paralysis in ARDS with absolutely totally different  outcomes. One was in an internal paper in 2010, which shows us that cis-sarticulum infusion improves outcome of patients. And John talked about cis-trials. There  was absolutely no difference. Now, this is unfortunately in curative care medicine  and this is not the first time that we have this, that one trial is positive, one trial is negative. My looking at it is out of the commons and I tend to agree with this. One  problem is that in the later trial, more patients have to be excluded based on paralysis already given, since they were included in the studies. And this  makes me wonder whether there are patients who really benefit from paralysis and they were just excluded because they were already on. There are over 500  patients who have been excluded from the study.  And one, and John brought this up, but is breast stacking a mechanism for harm? And when you have breast stacking, your tidal volumes go up significantly. John  showed this and I agree. Your tidal volumes reach 10 cc's for very short times. However, is this really such a rare phenomenon? And this is out of a paper  here in intensive care medicine and it seems to happen very, very often. So I do not know how it is. We don't measure it regularly in our ICU. Investigators  measured and found high frequency of breast stacking in the ICU patient. So now you can ask you, is breast stacking really so bad? It might be, but here you see  it goes up and obviously when you give a neuromuscular blockade, there's no more breast stacking. So your tidal volumes will be what you put in your ventilator  and not be higher. So the consequence, it could become out of a JAMA paper, and here you see that there was a low PEEP versus lung repeatment and high PEEP. And  interestingly, the low PEEP group had a lower mortality and supposedly also less ventilator dyssynchrony. And again, this is indirect and people may argue  this might not be the reason. It is one of the supposed reasons because there was a little more dyssynchrony, more people died from it. So now, spontaneous  breathing in ARDS, it's a double-edged sword. We talked about the blessing, John brought it up, and I agree, there's really good evidence to do it. I showed you some data  said it might be a curse when you tear your own lung apart. It's not really good for you. So, and ARDS is a continuum, a continuum of time and a continuum of  severity. And I bring here the Berlin criteria up. It starts from mild to moderate to severe with P2F ratios under 300, under 200, under 100 for severe ARDS.  Now we can go back to our little rabbit study. So when I have my, when I, when these rabbits which got, which were made into ARDS, into an ARDS model, in  mild ARDS, and they breathe spontaneously, over time, the PaO2 goes, becomes significantly better. So they are much better than when they were not  allowed to breathe spontaneously. This goes along with the data John showed you and his, and his explanations. However, in the really severe ARDS, it gets better  for a little while, and then our PaO2 goes down significantly. So this severely injured animals get worse when they are allowed to breathe  spontaneously. One thing is, for the same reason here, in the mild lung injury, when they breathe spontaneously, the transpulmonary pressures at here,  alveolar pressure to here, our negative inspiratory pressure, don't become much bigger, whereas in severe ARDS, here's a big decrease in negative  inspiratory pressure, and therefore much higher transpulmonary pressure. And therefore, also seen in the show, this is in the beginning a little bit, our peak  transpulmonary pressure goes significantly up when our animals, our sick animals, are allowed to breathe spontaneously, compared to the rest, where  does not make much difference in the, in the misled ARDS, or very moderate ARDS. Physiology shows you the same, and John showed you very similar slides. In mild  ARDS, spontaneously breathing, allowing permission of spontaneous breathing, improves our histology, improves physiologic parameters. I've shown you  this before, PA2 goes up significantly, whereas in severe ARDS, it gets much worse. So, and on the same token, since the animals, so in patient data, non-invasive  ventilation patient with ARDS, when your P to F ratio is low, under 150, you have a worse outcome. In patients who are, who breathe spontaneously and fully  mechanically ventilated, whereas when the ratio, when the ratio is over 150, it seems not to have made a difference in outcome. So, what, how would I summarize  in mild ARDS? I absolutely agree with John, spontaneous ventilation is the way to go. Patients will do better, most likely. They get all the benefits from  mechanical ventilation, including the diaphragmatic strength doesn't go down, so far, so on. Moderate ARDS, I would say, potentially, yes, spontaneous breathing,  monitor transpulmonary pressures, is knowing that it's hard to do, and to have exact transpulmonary pressures in each of these, in each of the segments is  impossible, but I would say, I would rather say, maybe yes, it's clearly not wrong, but in really severe ARDS, I believe it's a no, and spontaneous  breathing is not the best idea, and patients may really tear their own lungs apart. So, we have a continuum, so where do I stand? Since most of patients have  mild to moderate ARDS, I think for 95% of our patients, spontaneous breathing is fine, and there's this small evidence where it's not, and I think John come up,  and we can discuss this all together, what we, what we have to say. Thank you.  Thanks, Willie. John, if you could join me on the podium, and we, I will open this up for questions from the audience. Sure. Maybe I will start off with a question  to both of you folks. So, some, at the bedside, sometimes you see patients who have these really, they're not doing it very often, but they have these  really vigorous, spontaneous efforts. It's not mild, it's really, really vigorous, and you look at them, and you're like, man, the trans, or the intrathoracic  pressure is, there are really very, very big swings in intrathoracic pressure. Would you allow that? As I tried to explain early on, I would not. I think  early on, when you're trying to get control of the basic physiology, and the patient is at risk of continuous deterioration, or in pain, or agitated,  I certainly would not allow that to go on without sedating them adequately, and even paralyzing them. But, but I have to say one thing. We have been concentrating  so much on the linkage, or presumed linkage, between lung stretch, ventilator induced lung injury, and outcome, and we're forgetting the hemodynamic side,  which is every bit as important. Those same conditions generate big blood flows through the, through the lung, and that increases edema, and actually  worsens the potential for deteriorating oxygen exchange. And, you know, these big swings have marked, even in, I sometimes find that these big swings, even in  people who do not have overt left ventricular dysfunction, have actually marked, markedly increased afterload, and you, on the ventilator waveform, you  actually see big swings of pressure once these, concomitant, these efforts. I agree. The question is, and there's a, some people, and I'm not in this camp, try  partial paralysis in this patient, and it's a, it's a very difficult concept of partial, some people do it, and are adamant supporting this. I would first go  with sedation before I go to paralysis, drive as a narcotics, do it, narcotics plus propofol, plus a benzo, do it, and try to decrease it before I go to  paralysis. I would not jump to paralysis right away, if it's not so bad. We're not talking about a P to F ratio of 50 right now, and I think to jump to  paralysis right away, and even considering early ECMO is a very different, it's very different than a patient who is, looks okay, P to F ratio  is just breathing very hard. I think sedation is my, is the first step, and not paralysis. Very nice. Go ahead. Loretta. Good morning. Thank you very much. I have a  question. When you go to, from full mechanical ventilation to, and you allow patients to breathe spontaneously, do you ever consider to drop the respiratory  rate and have a rate of one, or go directly to pressure support? I find some of my respiratory therapists find somewhere in the middle, they don't  switch them to pressure support directly, but they allow them to breathe and drop the respiratory rate, as I said, to one. Let me repeat the question for the  audience. I think your question, if I got it right, was once you are coming off full support, do you actually decrease the respiratory rate and allow  spontaneous efforts on controlled mechanical ventilation versus go directly to pressure support? Correct.  Well, you know, if a patient is more comfortable on pressure support and is taking deep breaths and becomes more comfortable, and the level of pressure  support is appropriately just adjusted, I think that's a fine way to support the patient, even if they're fairly ill. And, in fact, the new data suggests  that that's true. It does not have a major detrimental effect. Just because a patient has severe ARDS and a minute ventilation of 12 to 14 liters  per minute does not necessarily mean that you need to paralyze them or deeply sedate them. I'm not sure I'm answering your question perfectly. My hearing is  not the greatest. Can you speak up a little? Yes. Do you ever consider to drop on mechanical ventilation the respiratory rate to one, which will allow  the patient to initiate breath, but the volumes at least will be the same? Because the danger in these patients sometimes, when switching directly to  pressure support, is that they will not be able to maintain an adequate minute ventilation throughout. I personally, I try to put them on an AC model. As I  said, they are allowed to spontaneously, and either pressure control, if you're a pressure control fan, if you're a volume control fan, and I'm saying this  because at the end it's not so much difference between the two, so they can spontaneously breathe and afterwards, and if this is comfortable for  them, it's fine. If it becomes uncomfortable to them, because then they can initiate as much breath as they want, and as you know, the difference in their  diaphragmatic stimulation is no difference whether they are spontaneously breathing or whether they are on AC pressure control, because  dropping the rate, my diaphragm and my nerves don't know about this. So personally, I will not drop the rate to a point that they are over-breathing or do  something. I would just keep it where it is and see what and try to do with an AC pressure control mode or AC volume control mode and let them  spontaneously trigger, but after the triggering, whether it's pressure support or pressure control, depends whether they are more comfortable.  I don't know what your opinion is, but if the patient is comfortable and is not getting into progressive distress, I think you can stay with that level of  support and monitor what's happening. I think one of the big mistakes we make is to take snapshots of the situation and then walk away. Something that's  perfectly adequate, as you're pointing out, at one point in time may cause deterioration of problems. Good point. Thank you, that was the question.  Mike Lanspa, Intermountain Medical Center, Utah. Those were both great talks, thank you very much. I had a question about how to, I guess, interpret the discussion about APRV  with spontaneous breathing in the sense that the release breaths that have been studied in APRV often exceed 12 cc's per kilogram of predicted body  weight, even when trying to target low tidal volumes. And so, when you brought up the mention of APRV, I'm curious, do you think that this is still an  acceptable mode of spontaneous breathing when our best quality data suggests that we should try to maintain low tidal volumes?  So, I am biased, and I admit this, I'm biased against APRV for this reason. But, again, in mild ARDS to moderate ARDS, it is a fine mode. But in severe ARDS,  where we try to limit where driving pressures and volumes look like have been shown to improve outcomes, but these are only a few patients, and these  patients I would not do it. But it's only, I don't, we don't see severe ARDS very often anymore. So, when there are a couple of patients a year, you know, a  handful, two handfuls of patients a year, it's not many patients. So, the majority of patients, it is a fine mode, because we don't get into this  problems, and it might help these patients even. However, in the severe ARDS, I personally would not do it. I don't know what, John.  Advocates, Dr. Hrabashi and colleagues have been advocating for decades that APRV used in the right hands can be used in every patient, and can be, can be, you  know, gradually weaned appropriately. And, you know, if you need that high distending pressure to keep the lungs recruited and open, which is usually  true only for the first phase of ARDS, if you're carefully monitoring your fluids, then APRV makes good sense to me. But, you know, but pay attention to the  mean airway pressure that's associated, and people forget that if the patient stops taking those spontaneous breaths, then the dynamics change according to  the physiology that I presented before. I have nothing against APRV. I think in mild to moderate disease, it's actually an attractive mode if the patient is  comfortable on it, and because it allows spontaneous breathing unfettered, I think is, is an advantage. So you would feel comfortable using it even if the release  breaths are... I'm sorry, I really can't hear you. So you would feel comfortable using it even if the release breaths are greater than 12 centimeter, or cc's per kilogram,  even if you had large release breaths with the APRV? You know, you, this sounds crazy for a guy who's been spending his life looking at ventilator-induced lung  injury, but we get so obsessed by numbers. I mean, as Ulrich was pointing out, the baby lung of ARDS can be relatively big or relatively small. The hypoxic vasoconstriction  can be strong or weak, and I think we're making a big, big deal out of ventilator-induced lung injury when it's actually a small component of what  actually causes mortality in these patients, and I think we've been ignoring, for example, the hemodynamic side for a long time, which does contribute to many  of our problems, including the lung edema and the need for sedation and, and the worsened gas exchange. Thank you. So another question for you guys. Would you  allow spontaneous breathing in the prone position, or spontaneous efforts in the prone position, especially when the intensity of the effort cannot be very  easily monitored? You're kidding me, right? I love the prone position, and many of us sleep in the prone position, so if you can allow spontaneous breathing in the  prone position, the patient looks comfortable on it, go for it, man. Yes, of course. Actually, our experience is that people actually require less sedation on  in the prone position many, many times. Ulrich, what do you think? So the question is, are you an early proner or a late proner? So if you have, in other  words, if you have somebody with severe ARDS where you would not allow spontaneous ventilation in the first, in the first place because of what we  brought up, then I wouldn't let them, then I wouldn't let them be spontaneously in the prone position. This is your last rescue effort in  your hospital, but when you are in a group which prones early, I think it's fine to do it, but if, so if you're a late proner, no, as an absolute  rescue therapy shortly before ECMO or so, then I wouldn't do it. If this comes, if this comes earlier in your, in your treatment cascade, then I would say go  for it. Very nice. Go ahead, please. Hi, thank you. In severe ARDS, if a patient is taking huge tidal volumes like 10, 12 mils per kig, but they're  comfortable, are you, would you allow that to happen in the era of 6 mils per kig? I'll hear what Olberg has to say, but you know, we need to pay  attention if the patient is alert to their comfort level. If they're, if they're comfortable, as the data from Valparaiso that I showed with  proportional assist indicate, if they're comfortable, they can probably set an appropriate, you know, level of lung stretch and, and, and yes, I would allow  that to happen. If the patient is uncomfortable, then I think you need to take steps to minimize the level of spontaneous breathing. That would be my  impression. I've, I've found personally you can reduce a lot of sedation use just by making them more comfortable, but allowing larger tidal volumes. Yes, so the  question is, you, this becomes obviously a very gray zone and I cannot give you an, I cannot give you an overall yes or no. These are the patients who are awake,  are interacting with you and get this high tidal volumes and are comfortable, then maybe their transpulmonary pressures are also not so high. They're often okay.  However, when you have somebody who is struggling to breathe and you don't see it because you gave them already so much fentanyl and, and you don't know whether  they're comfortable or not because they're totally snored, but they are still, then maybe I would not do this. I would rather play conservative because  again, we don't know whether we do anything bad and we don't know whether we do anything good. It's like, it's like this gray zones and this is why, and  John and I talked before about it, standing at the bedside instead of, instead of looking and instead of looking at the, instead of looking at  your flow chart and it's a bedside looking at your patient might really change it and looking in the room at the monitors and at your ventilator, what's  going on and it's better than snapshots because short snapshots give bad care. I would like to also point out that if you're going to use paralytics, I think  the whole audience knows this, but intermittent paralytics make good sense because you can then determine the level of sedation that you're using and  whether or not the patient can actively move their diaphragm and actually that small amount of intermittent activity may be very beneficial in preserving  the diaphragmatic function. So I'm not totally against the cautious use of paralytics when they're needed, but usually they're accompanied by deep  sedation, which is probably not a good thing, as opposed to mild sedation. Again, it also, I agree, a day of deep sedation and paralysis is often not a  problem. The problem comes when it goes for, when we say, oh, the patient was sick yesterday, the patient must be bad today as well, so the patient will be bad  tomorrow as well. It's Friday, let's wait till Monday. This makes it harder when we don't look. So short periods of seeing what's going on, the synchrony I  described can be often be interrupted with one dose of any paralytics and there's nothing special about cesarpercurium, in my opinion. You can use,  when you have rocoronium, use rocoronium for one dose. It really doesn't change much. It will be okay. It will go away.  Thank you both for your talks. My question is, if you have an awake patient in front of you who has sat, say, 90% on full support and you're, do you  care at all what their PIP or FiO2 are to start an SBT? I guess. Did you understand? So could you repeat the question, please? I'm a little...  Yep. So right now in our ICU, for example, our SBT is usually, if you have an improving oxygen requirement, and even if I'm not planning to extubate the patient,  we will do an SBT, but we currently are not doing SBT for someone who, for example, has 60-70% FiO2. But, you know, I think in favor of spontaneous  breathing would suggest that that is outdated and perhaps we should be more liberal in who we do an SBT with. So the question is what you see... So this  becomes a little bit semantics. And why it becomes semantics? Because spontaneous breathing trial has been shown in all these algorithms to do it as a bridge to  extubation, whereas spontaneous breathing, as John pointed it out, allowing to spontaneous breathe is not a bridge to extubation. It's giving the adequate  support the patient needs, but allowing them to take spontaneous breaths. So one, these are not... I wouldn't extubate on this, but you can, when somebody has an  FiO2 of 60-70%, there's no... I don't believe that you say, oh, I need full mechanical ventilation for this patient just because of the FiO2. Obviously,  the other parameters you mentioned, like PEEP, should be adjusted to the optimal PEEP, whatever this means for your given patient. So what do you do in  your practice? Who goes on a spontaneous breathing trial? So, again, pressure support is fine for me when the patient looks awake enough and is fine.  And some of these patients stay on spontaneous breathing with a pressure support, with enough pressure support for  ventilation, pressure support to give them an adequate tidal volume and enough PEEP to keep the lung expanded regardless of the FiO2 for me. But that's our  practice or my practice, which may not be everybody's practice. I don't know, John. Yeah, I think those comments are very appropriate. I don't think we really  know what a dangerous FiO2 is. And post-extubation in many of these patients, they're going to need some level of CPAP. Most important thing we  haven't said anything about except for Rahul's statement about proning is that position is so key important, especially in these patients. We often just  let them sit at 45 degrees or 30 degrees. Sit them up and make them comfortable sitting up, and that will descend the diaphragm. It'll improve the  oxygenation usually. And as far as if the patient looks comfortable, usually the FiO2 doesn't need to be any more than 60% to maintain an oxygen saturation of  92%. And I think it's quite reasonable to consider extubation in those patients, even if your FiO2 is more than what you think is desirable. Because once  the patient starts coughing, breathing on their own, taking deep breaths on their own, I think that the oxygenation problem will dissipate if the patient's getting  better for other reasons. Thank you. Last question. We have time for one more. Okay, so both of you have kind of cautioned against taking snapshots of, you know, a  clinical situation, trying to generalize too much from that. And yet, in a lot of these talks, you see snapshots of histology that looks both favorable and  pretty bad to support various sides of the argument. So I'm just kind of interested in the insider's view about, you know, how objective that histology  really is as it reflects a ventilation strategy in such a heterogeneous process.  I'll start by saying that's an intelligent comment, very, very well stated. And you have to understand that academic physicians are trying to  publish papers with new, novel, and interesting things to say. And as I try to point out, like, you know, Lucy and the football, we have made very stuttering  progress with that kind of an approach. People need to be targeting the truth.  When you look at the Kolobov paper, and Ted Kolobov is an esteemed mentor of Gattinoni and a friend of mine who passed away a number of years ago,  brilliant man. In that experiment, if you look at the details, the title volumes between the two groups, in sheep, 160 ml of title volumes in one group versus the  large title volumes in the other. So, you know, I think there are many factors intersecting in that particular study to help explain the results that we have.  Mechanical power, which I'm very interested in right now, was also much higher in those patients with high minute ventilation requirements. But the point  that you're making is that a lot of these illustrative slides and so on are meant to support data. And, you know, the typical example from the experiment is often not  a typical example at all, but the best one you can see. The principle, if it's good, will hold up. The truth will hold up. And I think one of those truths is that the  diaphragm needs to activate itself at least every, you know, 12 to 18 hours or you start to lose sarcomeres. And probably nutrition we haven't talked about is a factor as well.  Yeah, I agree. And obviously, in patients, we cannot get histology slides every 10 minutes. So our problem, and I try to allude to this a little bit in my talk,  our problem is that we do not have good insight in the lung during what we are doing. So both of us, besides what you said, you have to continuously look, are  very well aware of the limitations of it. But that's the best what we have at this point. I cannot send somebody every hour to a CT scan either to see what's going  on. So what do we have is a clinician standing at the bedside and looking at the patient and see how good it is. And as you pointed out, the histology slides only  makes a point and are obviously not there. I do not know how this rabbit looked an hour before and how this rabbit will look an hour later. Obviously, the rabbit is  dead when we get an histology slide. But that's about it. And hopefully our patient do better than my rabbits here.  All right. Thank you both for the wonderful session. Thank you all for coming and listening to us.

Video Summary

In this video, two speakers discuss whether or not spontaneous breathing should be allowed in acute respiratory distress syndrome (ARDS) patients. The first speaker argues that spontaneous breathing should be allowed, as it can help reduce the work of breathing, relieve discomfort, and improve gas exchange. They also point out that allowing spontaneous breathing better matches ventilation to perfusion and can improve distribution of regional strains in the lungs. The speaker acknowledges that sustained dyspnea, pain, or agitation should not be tolerated and that patient comfort should be a priority. The second speaker takes a more cautious stance, suggesting that spontaneous breathing may not be ideal in severe ARDS cases. They argue that spontaneous breathing can lead to high transpulmonary pressures, potentially causing lung injury. The speaker suggests that in mild to moderate ARDS, spontaneous breathing is likely beneficial, but in severe cases, it may be best to limit or control spontaneous efforts. The speakers also discuss factors such as patient comfort, sedation, and paralysis in the context of allowing spontaneous breathing. They emphasize the importance of individualized care and monitoring the patient's condition closely to determine the best approach.

Asset Subtitle

Pulmonary, Procedures, 2020

Asset Caption

The Society of Critical Care Medicine's Critical Care Congress features internationally renowned faculty and content sessions highlighting the most up-to-date, evidence-based developments in critical care medicine. This is a presentation from the 2020 Critical Care Congress held February 16-19, 2020, in Orlando, Florida, USA.

Learning Objectives

Analyze the pros and cons of spontaneous breathing during acute respiratory distress syndrome (ARDS)
Decide when and when not to allow spontaneous breathing during ARDS
Recognize the importance of spontaneous breathing to prevent ventilator-associated diaphragm dysfunction and muscle wasting
Discuss the phenomenon of pendelluft in regional overdistention and contribution to ventilator-associated lung injury in ARDS
Assess early versus late prevention or allowance of spontaneous breathing in ARDS

Spontaneous Breathing in ARDS: Absolutely

John J. Marini

Spontaneous Breathing in ARDS: Absolutely Not . . . or Does It Depend?

Ulrich Schmidt

Meta Tag

Content Type Presentation

Knowledge Area Pulmonary

Knowledge Area Procedures

Knowledge Level Intermediate

Knowledge Level Advanced

Membership Level Select

Membership Level Professional

Membership Level Associate

Tag Acute Respiratory Distress Syndrome ARDS

Tag Ventilation

Tag Cardiothoracic Critical Care

Year 2020

Keywords

spontaneous breathing

acute respiratory distress syndrome

ARDS

work of breathing

gas exchange

ventilation to perfusion

patient comfort

severe ARDS

lung injury

individualized care

	Society of Critical Care Medicine 500 Midway Drive Mount Prospect, IL 60056 USA Phone: +1 847 827-6888 Fax: +1 847 439-7226 Email: support@sccm.org	Contact Us About SCCM Newsroom Advertising & Sponsorship DONATE	MySCCM LearnICU Patients & Families Surviving Sepsis Campaign Critical Care Societies Collaborative	GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. \| Privacy Statement \| Terms & Conditions The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.