SCCM Resource Library-ARDS: Medical and Ventilator Management

Video Transcription

Hello, my name is Sofietu Rogers and I'm the moderator for the session ARDS, Mechanical and Ventilator Management. I have no financial disclosure.  ARDS is a common complication seen in the COVID-19 as well as septic patients in the ICU.  In this session, we are going to discuss medications used in the treatment of ARDS, various ventilator  modes for the treatment of ARDS, and using case-based learning, we will discuss some clinical scenarios with the management of ARDS.  I hope you find this session helpful in your practice. Thank you very much for your attention. Good afternoon. My name is Jonathan Van Cleve.  I'm a critical care pharmacist that practices in a surgical and medical mixed ICU setting.  Today I have the privilege of talking to you about the pharmacotherapy of acute respiratory distress syndrome, or ARDS.  This is a topic that's become much more popular over the past year with the current COVID-19 pandemic.  Today's presentation will only cover pre-COVID-19 data, seeing that data is always coming out according to COVID-19 ARDS.  So for disclosures for the presentation today, I have no disclosures. Going into the objectives for today's presentation, first of all, I want to examine key trials  for the pharmacological treatment of ARDS, and second, I want to discuss practical issues  involved with these medications used to treat ARDS, as there can be some complicating factors when using these.  The first drug class that I want to discuss is paralytics, and I want to start off with  a guideline statement from the SECM neuromuscular blockade guidelines from 2016, and they state  that we suggest that a neuromuscular blocking agent be administered via continuous IV infusion  early in the course of acute respiratory distress syndrome for patients with a PDF ratio of less than 150.  Now, that was a weak recommendation with moderate quality evidence, and so going into the evidence, the first trial that I want to look at today is the ACQUIRISIS trial.  Now this was published by Dr. Papazian and colleagues back in 2010, and this is the foundation for that recommendation for the guidelines.  Now in this trial, they were looking at moderate to severe ARDS with a PDF ratio less than 150.  If you remember, moderate is less than 200, and severe is less than 100, so right in the middle of that range, and they were looking at 340 different patients in this trial.  Now the goal of this trial was to get patients early on in diagnosis, so they randomized the patients within 48 hours, and there was no difference between the two groups.  The intervention in this trial was cis-accurium or NIMBEX, 15 mg bolus, followed by 37.5 mg per hour for 48 hours.  Now the control group received heavy sedation and did not receive paralysis. Now the outcomes that they looked at in this study were primary outcome of 90-day mortality.  Now as you can see, I have a little asterisk beside that. They had a pre-planned Cox regression analysis, which they planned to adjust for baseline  PDF ratio, SAPS-2 score, and plateau pressure. Now the secondary outcome they were looking at, or the main secondary outcome they were looking at was 28-day mortality.  Now looking at a visual survival curve, you can see there was this nice separation between the two curves.  However, as you will note, this separation did not occur until around 15 to 20 days. This is very interesting that a 48-hour intervention would not take effect until later on in the  disease course. Now if we were looking at their crude 90-day mortality, it was 31% versus 40%, with a p-value of .08, so not significant.  However, with the pre-planned adjustment, they had a hazard ratio for death at 90 days of 0.68, which was significant.  Now their secondary outcome of 28-day mortality was 23% versus 33%. Now that was a significant outcome.  Now going into this, this looked like an adjustment that they pre-planned, and their outcomes showed that it was significant when they made that adjustment.  Now this mortality benefit was seen with the group that was with a PDF ratio less than 120, so not the overall group of less than 150.  Some criticisms of this trial was they were not able to completely bind the two groups as one was paralyzed and one was not.  They did not use TRANF-4 to measure the level of paralysis, and ventilator dyssynchrony was not characterized or monitored in any way.  Now for more recent evidence, the ROSE trial came out in 2019. Now Dr. Moss and colleagues wanted to mimic the trial that was previously done, so they  also looked at monitor to severe ARDS. However, their trial was much larger with almost three times as many patients in this group.  They were randomized early within 48 hours of diagnosis. They had the same intervention. However, there was a couple notable changes with this group.  Now the control group in the ACQUIRSYS trial was deeply sedated. However, the control group in this trial was lightly sedated.  Now this is more in line with the most recent PADIS guidelines to lightly sedate ICU patients.  Now their primary outcomes were very similar with a 90-day mortality as their primary outcome and the 28-day mortality as a secondary outcome.  Now for their primary outcome, they found absolutely no difference, 42% in each groups with not even a trend towards significance.  Now an interesting safety outcome was they had higher cardiovascular SOFA scores on days one and two.  As you will note, those are the days where the patients were being treated with cis-atricurium.  Now this is very interesting as most clinicians are not looking for cardiovascular toxicity when they're thinking of cis-atricurium.  Cardiovascular toxicity has more commonly been seen with pancuronium and rocuronium through their vagolitic activity.  However, this could have been due to deeper levels of sedation as agents such as propofol and dexmedetomidine can cause both bradycardia and hypotension.  Now looking at a visual of their results, you can see absolutely no separation in their survival curve. Also, other secondary outcomes had no difference at 28 days.  Those include hospital mortality, days free of ventilation, days out of the ICU, or days out of the hospital.  Now if you look at patients discharged to home, there was also no difference. So two conflicting trials that we have evidence for now. So why the difference in outcomes?  Well first of all, I want to look at proning. Now we know from the PROSIVA trial that proning has become standard of care, and it's one  of the most effective ways to decrease mortality in this disease state. If you look at the ACQUIRIS-CIS trial, 50% of the patients were proned.  However, if you look at the ROSE trial, only 16% of patients were proned. Now sometimes paralysis is needed to facilitate proning.  However, during this current pandemic, our practice have actually seen you can sedate patients deep enough where you can prone them without paralysis.  The second area I want to look at is sedation. Like I said earlier, the ACQUIRIS-CIS trial had deeply sedated patients. However, this is not the standard of care now.  Now of note, a lot of the original sedation trials where it looked at deep versus light sedation excluded severe ARDS patients.  Now if we're looking at the ROSE trial, they were lightly sedated. As an effect, this could have caused them to have higher cardiovascular or SOFA scores  in the paralysis group, which was heavily sedated. The third point I want to look at is ventilator to synchrony.  Sometimes paralysis is needed to make a patient more synchronous with the ventilator settings. This was not characterized in the ACQUIRIS-CIS trial or the ROSE trial.  So overall, should we use paralysis in ARDS? I think routinely we should not be using paralysis. However, there are a few select groups where paralysis might be necessary.  These groups would include those with a high plateau pressure, those who would need paralysis to facilitate proning, or those with significant ventilator to synchrony.  Some practical points on paralytics. Does the choice of paralytic agent matter? Do they all have the same efficacy? So far, the majority of studies have used cis-astrocurium.  However, any paralytic should help a patient be more synchronous with the ventilator.  I think most of the studies have used cis-astrocurium due to its unique pharmacokinetic properties  as a quick onset, a quick offset, and the clearance is not limited by hepatic or renal dysfunction.  Some of the longer paralytics might cause more muscle weakness later on in the trial as the patient would be paralyzed for longer due to accumulation of the drug.  Should the intensity of paralysis be measured? Neither trial measured the intensity of paralysis. They did not use a train of four or anything like that.  So I think if we were to use paralytics, as long as it's used for a short amount of time, paralysis should not be monitored.  One of the reasons why we monitor paralysis is to limit the amount of neuromuscular weakness down the road. However, this was not seen in either one of the trials.  Now, what dose of paralytic should be used? In these trials, they used a bolus with a continuous infusion. However, now with the ROSE trial showing no difference,  I think we need a trial looking at ventilator synchrony and paralyzing those patients. Fourth, how do you appropriately measure the level of sedation in a patient who is paralyzed?  Now, there are things out there such as the bispectral index score or BIS. Now, this has been more validated in the OR than the ICU, and it's not currently recommended.  In our practice, we deeply sedate patients to a RAS of negative five prior to paralysis and then put them on a set rate of sedation while they are being paralyzed.  However, I think there's some questions still out there on appropriate levels of sedation while paralyzing these patients.  The next drug class I want to look at is corticosteroids.  I included a guideline recommendation from the 2017 CERCI guidelines. In this, they recommended the use of corticosteroids in patients with early moderate  to severe ARDS with a P to F ratio less than 200 and within 14 days of onset.  So, let's look at the data behind that.  Some of the older data that I wanted to talk about today, Dr. Madhuri and colleagues, all the way back in 1998 in JAMA, looked at a 32-day steroid taper regimen.  They started with methylprednisolone, two milligrams per kilogram per day, and tapered it down over 32 days.  What they found was improved oxygenation and reduced lung injury severity score. They also found reduced ICU and in-hospital mortality.  Now, Dr. Steinberg did a larger study published in 2006.  What they found was improved ventilator-free days, however, no improvement in overall mortality. There was also higher mortality when used late in the disease state.  There was also more neuromyopathy.  So, Dr. Madhuri also published a study in 2007 that showed reduced duration of ventilator days and ICU length of stay.  Improved survival to discharge from ICU, but not from hospital. So, we have some conflicting results here from these early trials.  So, when they put four of the randomized studies with early steroids into a meta-analysis, what they found was a hazard ratio of 0.5 that was significant.  So, when you put all these studies together, you did have a significant decrease in mortality.  But this left us wanting for a better study that was larger and one where we had a firm outcome on whether to use steroids in these patients.  Now, looking at the DEXA ARDS study that came out in early 2020, right before the COVID-19 pandemic,  these patients had moderate to severe ARDS with a P to F ratio less than 200. They looked at 277 patients.  Now, the intervention they used was not methylprednisolone like the earlier trials. They used dexamethasone 20 milligrams daily times five days,  followed by 10 milligrams daily times five days. But the patient was extubated, then the treatment course was stopped.  Now, dexamethasone does have some unique properties, such as having a longer half-life of 30 to 40 days. But it's not a very effective treatment.  So, it's not a very effective treatment. So, it's not a very effective treatment. Dexamethasone does have some unique properties,  such as having a longer half-life of 36 to 48 hours.  And it does not have any mineralocorticoid effect. So, it does not cause fluid retention, which could make ARDS worse.  The outcomes they looked at in this trial were ventilator-free days at 28 days. And the secondary outcome they looked at was 60-day mortality.  Now, they did find a significant difference  in their primary outcome with 4.8 more ventilator-free days.  Looking at their secondary outcome of 60-day mortality,  they found a 21% versus 36% difference in mortality.  Now, looking at their Kaplan-Meier curve, you see this nice separation in survival.  And they also had secondary outcomes that were all statistically significant as well. So, a positive trial. Now, looking at some practical points, timing is very important.  Like I said earlier, you do not want to use these steroids late in the disease course. It can actually be worse.  Plus, we want to be using these steroids early in the disease state when the inflammatory state is going on. Second, the dose.  You want to make sure you're using an appropriate dose. Now, I would prefer the dexamethasone due to the recent DEXA-ARDS study.  However, there are many mechanistic reasons to use methylprednisolone, and that was the study used early on. So, an equivalent dose of dexamethasone 20 milligrams  would be methylprednisolone 100 milligrams. So, I think that's a reasonable option as well.  And third, I want to look at adverse effects. Steroids, as we all know, have many adverse effects.  And I don't think they are always well characterized in clinical studies. For example, in the DEXA-ARDS study was stopped early  and did not meet power to look at some of these adverse effects.  So, I think you always want to be watching out for these. Now, they did not find any differences in adverse effects in that trial,  including barotrauma, hyperglycemia, or infections.  The last drug class I want to talk about today is inhaled pulmonary vasodilators. Now, this is a drug class that we've looked at  over the past few decades to improve oxygenation and potentially mortality.  Now, what all the studies have seen is what you see over here on the left, which is a temporary improvement in oxygenation for about 36 to 48 hours.  However, there has been no decrease in mortality seen across these studies. Now, there have been eight randomized studies in over 1,000 patients with ARDS,  with at least 10 patients treated with inhaled nitric oxide, and they have all shown a temporary improvement in oxygenation, but no improvement in mortality.  So, we have a drug that works temporarily, but does not improve overall outcomes.  Now, another important factor that you should look at when using these drugs is they are very expensive.  Inhaled nitric oxide, it can be one of the most expensive drugs to use in the ICU. So, inhaled epiprostanol has been shown to be non-inferior in efficacy to nitric oxide,  but considerably less expensive, with one study showing a cost difference of $14,000 for inhaled nitric oxide versus $800 for inhaled epiprostanol per patient.  So, if you are using inhaled nitric oxide frequently in your ICU, a huge cost savings would be significant.  Frequently in your ICU, a huge cost savings would be switching over to inhaled epiprostanol,  even though it does require special equipment and special considerations when using that drug. Now, some of the adverse effects that have been seen are listed below,  including hypotension, methemoglobinemia, inhibition of platelet aggregation, pulmonary edema, and renal dysfunction.  Now, some of these are theoretical and have not been seen very frequently in clinical trials. With the exception of renal dysfunction,  so a meta-analysis of all the trials using inhaled nitric oxide,  renal dysfunction was seen as an increased risk factor when using inhaled nitric oxide.  So, overall, looking at inhaled vasodilators, these should probably not be used frequently in clinical practice to treat ARDS patients.  However, they possibly could be used to back a patient back from a cliff when they are severely hypoxic, or as a bridge to therapies such as ECMO  to give you more time to get on to different therapies.  So, with that, I'm going to finish my presentation today. Thank you for having me, and we will have a Q&A session at the end of all of our speakers.  However, if you have any further questions, feel free to email me at the email address listed below. Thank you.  Hi. I want to thank the Society for asking me to give this talk today, and I've been asked to talk about ventilator modes and the management of ARDS.  These are my disclosures, companies that I either consult for and have financial relationships with or have research support. I wanted to do a little bit of basic to begin with,  and I don't want to spend too much time on it, but to make sure we're talking the same language. So, a mode of ventilation describes the sequence of breaths that occur  and the ventilator's response to the patient effort. So, modes of ventilation are terms we use like assist control,  continuous mandatory ventilation, intermittent mandatory ventilation, pressure support, APRV, NAVA.  These are all modes of ventilation.  A lot of times we think of, well, what mode do you want? Pressure or volume control? And pressure and volume control are breath types.  So, for instance, you can do pressure control assist control or you can do volume control assist control. So, it's important that we make that distinction  as we go through some of these discussions. Now, again, 40 years into my career, there's no uniform terminology, and unfortunately, on five different ventilators,  doing the exact same technique, I'll call it something different.  I've resigned myself to the fact that that's probably never going to change. But again, a breath type describes what variable is controlled  and will always be maintained at the sacrifice of other variables and are usually either pressure, volume, or now we have the adaptive breaths.  And remember, mode has nothing to do with peep usually. That's a totally separate setting on the expiratory side.  So, within the modes of ventilation, then we can have modes that provide only mandatory breaths, CMV assist control,  a combination of mandatory and spontaneous breaths, SIMV, APRV are good examples, that are only spontaneous breathing modes, pressure support, proportional assist, NAVA.  And then there are closed loop modes, which are interesting because, for instance, adaptive support ventilation is simply SIMV  with a set of rules that control the output of the ventilator based on inputs from the measurements that the ventilator makes.  This is from a paper in the Cleveland Clinic Journal of Medicine that I think is from Ehab Daoud and Rob Chapern. And just to give you an idea of this,  what's the difference between volume and pressure control? So, as you look at this, the black line is volume, the red line is airway pressure, and the green line is,  or the blue line is patient effort.  So, in volume control and pressure control,  in the absence of respiratory, any inspiratory effort, this is what the waveforms look like.  If there's a small respiratory effort, in volume control, there's no change to the flow pattern at all. It stays the same, the flow's in green,  whereas the flow might be altered significantly differently here in pressure control. And then, importantly, a big deep breath,  look what happens to volume control to the airway pressure. In red, it gets drawn underneath this pressure because the flow stays constant regardless of the patient effort.  So, we see the big patient effort, a reduced flow, a fall in airway pressure.  Pressure control maintains the same pressure waveform, prevents that deformation of the waveform, but will give a larger tidal volume than you set.  So, pressure control can't guarantee tidal volume. And so, you see what happens here over a series of breaths. And this is important for understanding,  as we talk about caring for patients and lung protective approaches.  So, I put these three different kinds of breaths, the volume control, the pressure control, and adaptive pressure here, and show you what happens when the compliance gets stiffer,  what happens when it gets better, what happens if patient effort increases, and what happens in the presence of intrinsic PEEP. And I think most people are probably very comfortable  with what happens during volume control and pressure control. But I want to draw your attention to the fact that adaptive pressure breaths, PRVC, volume control plus,  autoflow, lots of names, that the response is different. So, in particularly, in an increase in patient effort, the airway pressure delivered by the ventilator will go down,  while the tidal volume will go up or may stay the same. And that's really important because it helps us describe what happens to the patient  as the ventilator responds to patient effort.  Now, again, I've spent enough time now,  you know, where we used to, the goal of mechanical ventilation was to normalize gas exchange. And I think early in the, or the middle 1990s, Art Celesti probably should get credit  for a lot of this approach towards, it's not normal gas exchange that's important, but preventing ventilator-induced lung injury. And I'm sure you all have seen this picture  from the Blue Journal, where Art goes through the, there are both biochemical injuries and biophysical injuries.  And I'm not gonna go through all this, but what our goal then is, if with our CT scan over here to the right of the patient with consolidation in the dorsal part of the lung  is as we move down to the bottom, we wanna not over-descend the lung that's open, and we want to recruit the lung that's collapsed. Not, maybe not all of it,  but enough of it to improve gas exchange. So, it's a balance of ventilatory support.  I really liked this paper by Gallagher that was published in Intensive Care Medicine this year, where they combined the two new concerns.  One, not only minimizing ventilator-induced lung injury by preventing breast stacking due to dyssynchrony, so double triggering,  decreasing the patient's effort, the excessive inspiratory assist, resulting in high tidal volumes, providing too much PEEP, and again, too much PEEP in total  or too much PEEP regionally, insufficient PEEP then, of course, regionally, can result in derecruitment and atelectasis.  The P-CILI, so pressure self-induced lung injury is kind of a new concept that honestly doesn't have all that much evidence, and it's kind of more of a hypothesis,  in my opinion right now. It may be there, it may be part of the lung injury cycle, and maybe it's not.  But on this side, we're worrying about diaphragmatic injury, which of course has to do with how much effort or little effort that is, because like everything,  you can ask the diaphragm to do too much and you can paralyze patients or overdrive them with the respiratory rate and it can do too little,  and both have problems from disuse atrophy  to eccentric injury caused by expiratory dyssynchrony. So this concept of ventilators don't save people's lives, they provide support,  and we want to deliver the ventilation that's safe and effective while the patient improves and responds to medical therapy.  Here we have, again, from the same paper, looking at ways of accomplishing these goals, the first being to try to get the patient to trigger the breaths,  and we're not talking about spontaneous breathing, just the active triggering of the breaths, monitoring the patient effort, so you can use P.1 to see if the effort is excessive.  So in the boxes below, you can see the things that you can change, increase inspiratory flow, increase the pressure, increase the rise time, give more sedation,  monitor lung stress and strain, which has been very popular and not all that well understood, things like driving pressure that we talk about now or power,  you can reduce the inspiratory pressure, adjust the PEEP to improve compliance, and again, monitor for asynchrony. Then if all these, if they go well,  lead to weaning and liberation, and if they don't, often lead to ECMO.  So Eddie Phan published this paper in JAMA, and again, with the time that I have today, I'm just gonna kind of present it to you.  Of course, you're gonna be able to watch this online,  but there's lung injury mechanisms, so we're worried about volutrauma, so overdistension, atelectrauma, the repetitive opening and closing,  but we're also looking at other kinds of lung injury, this ergotrauma from excessive mechanical power,  and this shows you then what's the clinical response, the potential tools and monitoring, things like EIT, which I think will become more popular  and kind of gives you a real unique look into the lung and the regional differences in the distribution of ventilation. Myotrauma, again, using esophageal pressure, looking at P.1,  looking at the signal from the diaphragm.  Patient self-inflicted lung injury, again, I think is something that needs further study, and then patient ventilator to synchrony,  more and more, it's always been associated with poor outcome, now we see that it often is, might be the cause of poor outcome.  And as you go down on here, and I think this is really important, these are the recommendations by the Joint Committee on Guidelines about mechanical ventilation, prone positioning,  high frequency, higher versus lower PEEP, the use of recruitment maneuvers, and ECMO, and again, I don't wanna go through all this, but what I wanna point out to you here is,  nowhere in here is the consideration of the mode of ventilation, so it's really the goals that we're looking for that are important, not the way that we achieve those goals,  you can do it with volume control, you can do it with pressure control, as long as you understand the principles of ventilation.  So if you're gonna use a mode or a new mode, and I thought this being a big society meeting coming off the pandemic, what are the things that you would want a new mode to do?  Well, you'd like for it to avoid excessive ventilation or volume trauma, you'd like for it to be able to avoid the adelectrama,  you wanna avoid the maldistribution of stress and strain  resulting from lung inhomogeneity, and of course, you wanna be able to have a mode that would reduce the intensity, frequency, and the duration of the tidal cycling  or the mechanical power.  So real quick to go through a couple, and again, NOV is not a new mode, it's been around a long time, but I wanted to at least share that there's a new paper out,  so NOV, you know, uses a NG tube that's placed and has an array of sensors that can determine the diaphragmatic contraction, and so instead of triggering  from airway pressure, volume, and flow, you're triggering from the neural signal, which from a standpoint of speed of response, has to be faster than waiting  for it to happen at the airway. NAVA may has been suggested as allowing early spontaneous ventilation. It might avoid the myotrauma that Gallagher has talked  about extensively, and it prevents overdistension. Patients under control of NAVA rarely get too big a tidal volume. So Bob Kastmeric and the group from  Spain published this paper, Intensive Care Medicine, just recently. It's an EPUB, and I'll just draw your attention to this. So 153 patients under randomized  control trial in NAVA, and then the group patients in the control group getting assist control and then pressure support. No real difference in mortality, and I  don't know that you'd expect it, but there were significantly difference in the total duration of mechanical ventilation in days. The NAVA group  averaging about eight days, and the control group about 12. So this is one of the first times we've had some evidence for evidence for NAVA that it has an  out positive effect on an outcome of interest. It clearly improved synchrony. The question has been, is improving synchrony worth the fact the extra cost  of the catheter, the fact that only one ventilator can do it? And you know, there are some criticism of this trial, but again, new evidence suggesting maybe it's  useful, and just to show you here is probability of survival wasn't any different. APRV gets talked about a lot as if it's new. It's not, and I show you  this picture. This is the very first APRV device I put a patient on, which was a Downs flow generator, basically a big solenoid valve, and it switched back and  forth between 30 and 20 centimeter peep valves and created this waveform, and John Downs and Chris Stock said, hey, APRV is gonna maximize recruitment, improve  oxygenation, improve hemodynamics because of the spontaneous breathing, result in reduced sedation because you don't have to worry about synchrony, and uncouples  the patient and the ventilator. And there have been a lot of studies, and again, I'm just gonna show you these really quickly. So here are the bunch of crossover  studies through the early 90s showing that compared to conventional ventilation, you had a reduced peak pressure, but similar gas exchange, but nothing really  important. Then you have some randomized controlled trials, and again, I'll just call your attention to the more recent data, this publication by Zao and  Intensive Care Medicine, which has gotten a lot of criticism, but is a randomized controlled trial of a lung protective approach with a very short eye time,  initial, or expiratory time of one and a half seconds, followed by setting the expiratory time, a release time based on peak expiratory flow, and they showed  improved oxygenation and compliance, lower plateau pressures, and significant reductions in mortality. But as a couple of people have pointed out, some of that  comes at the cost of not doing the CMV portion of the study very well. And just to show you some of the data, they showed a difference in mortality of 20 to 34  percent, but it was not a statistically significant difference, no real difference in barotrauma, which really isn't much of a problem anymore, but big  difference in ventilator-free days. And this is one of those studies where the the results are actually sometimes almost seem too good to be true, but at  least we have a randomized controlled trial that people have been calling for. More recently, the group that's been a real proponent of APRV has switched from  the APRV styles on the left of progressively shortening the release time to what they now call TCAV. And TCAV is an adaptive form of ventilation  that uses really high inspiratory pressure and really short release times. And again, in the interest of getting through this in a reasonable amount of  time, so here's a traditional ARJNET method of delivering volume control ventilation. Here's, you know, constant flow, so you know it's volume control,  airway pressure, the set PEEP, the mean airway pressure in black, and the peak airway pressure. The TCAV method here uses an airway pressure release  ventilation with a really short release time and control of that release time by the slope of the expiratory flow. And I just bring this out again, I'm not a  proponent of TCAV, in fact I'm going to tell you a little bit what I think here in a second, but here's from the paper by Neiman looking at a patient who's on  conventional ventilation and then on TCAV with the changes in driving pressure associated with improved compliance because of full recruitment. And you know  these pictures are kind of impressive, but I'll remind you that another, the same group from New York years ago was saying high-frequency oscillation  is the answer for the same reason, this maximum recruitment reduces ventilator-induced lung injury. And TCAV I think is still, has a lot of  experimental models with few clinical applications, very high mean airway pressures that are going to require increased fluid requirements and perhaps  vasopressor support. And again, so many of these techniques, high-frequency  oscillation, super-PEEP, have failed when it comes to clinical trials because the high airway pressure requires sedation and paralysis, the sedation and paralysis  requires greater fluid requirements or vasopressors, and all those things lead to worse outcomes in patients. And here's the data from the OSCILLATE trial, again  run by a group of people who are very, in the past, were proponents of high-frequency oscillation, I think still are as a rescue therapy, but demonstrating  that in this case, there was a lot greater use of vasoactive drugs and days of vasoactive drugs and an increase in mortality. So the last thing  I'll talk about real quick in a mode is, so it's not available in the U.S. yet, but it's called INTELLiVENT, and INTELLiVENT is a closed-loop control method of  adaptive support ventilation, which has been available in the U.S. for 25 years, with a controller to control PEEP and FiO2, based on the two ARDSNET, ARMA, and  alveoli trials. So the PEEP and FiO2 increase in blue using the ARDSNET alveoli, or the ARDSNET ARMA trial, and fall using the ARDSNET alveoli trial. So  here you have rapid increase in FiO2, here you have slower decrease, and a slower decrease in PEEP. And there has been some data from the studies from the  European groups showing that it's feasible to use something like INTELLiVENT in closed-loop ventilation, but no real important differences in  length of ICU stay, length of hospital stay. Again, this study is too small to really look at mortality, but is a mode that I think, again, in the midst of a  pandemic where we have to wear PPE to see patients, having the ventilator make some mild changes so we don't have to interact with it every time might be  useful. But I will warn you, everything has a caveat, the number of ventilator changes, and this always gets reported. Look, with INTELLiVENT, these closed-loop  modes, we don't have to interact with the ventilator as often. The ventilator is making these changes for us, and I would ask you, is that an  advantage? And if it's an advantage, is it an advantage for who? And there is some concern that some closed-loop methods might actually result in a lack of  vigilance, clearly something we need to be aware of. And then the last thing I'll tell you, it has nothing to do with modes, but at least has a discussion  of, you know, we keep saying there's going to be precision medicine, and this is a paper out of the British Journal of Anesthesia that Marcelo Amato is a  senior author on, I believe, doing a reanalysis of the ART trial, which was using recruitment maneuvers, and what they found, and I think not  surprisingly, but in an analysis far more complicated than I could ever consider, is that a technique or a mode for any given patient isn't good for  everybody. And again, you can read this as easily as I could read it to you, but here's the look at the different clusters of patients based on their  illness and injury, and that in some cases, ARDSnet techniques without recruitment result in better outcome, whereas there are other  patients where it doesn't result in a better outcome. So patients are different in this precision medicine I think we're going to look for, but clearly in the ART  trial, pneumonia and sepsis in patients with hypotension didn't respond very well to recruitment maneuvers. So I think going forward, we really need rigorous  clinical trials before we start using this TCAV idea based on what we learned about high-frequency oscillation to understand where we should use them if  ever in ARDS, and again, I've seen presentations where the TCAV group talked about the number, you know, the liters of fluid required to maintain  blood pressure in these patients, and I think that's going to be a problem in the long run. In the interim, we shouldn't be using these unless they're in a clinical  trial. Again, I don't know that I've used NAVA all that often. I'm impressed about how it improves synchrony and comfort, and we really need  better and more trials to suggest that it does something positive, and I think closed-loop control may have advantages in select environments. And this is from  editorial that Rotano and Phan did in respiratory care, where you look at mild, moderate, and severe ARDS, and blue is where we have proof for we ought to do  these things, red is where we have proof that we probably shouldn't do these things, and yellow is where we need more information, and again, that customized  ventilation is at the top. So I don't want to go on too long, so I appreciate your attention, appreciate the Society asking me. I'd be happy to send you the  slides if you email me, and hope you have a good day and stay safe. Thank you. Hello, my name is Eric Magana. I'm a pulmonary critical care physician from  Albuquerque, New Mexico. I'm here to speak on ARDS management with a mechanical ventilator. I have no disclosures. There are many great topics  to discuss regarding ventilator management. I'm going to focus just on PEEP and tidal volume. So we start off with a case as a 55-year-old female. She  presents with acute pancreatitis, and she develops hypoxemic respiratory failure and needs to be intubated. Her echocardiogram shows an ejection  fraction of 60%, and there are no valvular or diastolic abnormalities. She's placed on mechanical ventilation at volume AC, 6 cc's per kilo, with a rate  of 30, PEEP set at 14, and the FiO2 is 80%. Her PF ratio is 75. She meets all the criteria from Berlin definition for ARDS. She meets it based upon timing, because  she's been sick less than a week. Her chest imaging certainly shows bilateral infiltrates. Her echocardiogram shows this is not cardiogenic in origin, and  her PF ratio puts her in the severe category of ARDS. So the ventilator management goals we'll have for this patient, and for most patients, is we want  to provide adequate oxygenation while providing adequate ventilation. In order to do this, we need to recruit alveoli. We also want to avoid ventilator-induced  lung injury, such as atelicta trauma and barotrauma or volutrauma. This overall is called a lung protective strategy of mechanical ventilation. The slide on the  right portion shows, the top panel shows that there's quite a bit of atelictasis that represents alveoli which are not open. As we apply PEEP, there's  significant alveolar recruitment. The more alveoli that we have open, the better gas exchange that we're going to have. In this slide, on the panel on the  left, shows a tidal volume being delivered to the patient. With inspiration, you can see that the lung is inflating, all the way to the point  where now we switch to expiration. And as the patient expires, the alveolar units that were recruited now deflate. As the lung is inflating, we're going up the  compliance curve, and then we go back down the compliance curve. This is an important concept to think about when we're performing mechanical ventilation.  So PEEP is the tool that we use to keep alveoli open between breaths. PEEP alters the patient's position on their compliance curve. We can have PEEP of, say,  a 5, or PEEP of 8, which results in a little bit more inflation. If we increase the PEEP to 12, we have even more inflation and we're in a higher position  on the compliance curve. When a patient is breathing, ordinarily the alveoli are  open in between breaths, and then the tidal volume goes in and it distends these alveoli to return back to an open alveolus. In people who have diseased  lungs, their alveoli are collapsed in between breaths, and as a tidal volume is given, the alveoli will open back up and then collapse down again between breaths.  For very severe disease, these alveoli are never open. We're trying to use PEEP to take all the alveoli from these two groups and convert it into this group.  When the alveolus opens and closes between breaths frequently, that results in shearing trauma within the alveolus that increases the inflammation and  worsens lung disease. PEEP will mitigate this. Sometimes people have talked about using recruitment maneuvers, and we use these to varying degrees. There was a  recent study where they used recruitment maneuvers where inspiratory pressures of between 40 and 60 centimeters of water were used. This significantly improved  oxygenation in these patients. Recruiting more alveoli certainly improves oxygenation, but it also came at a price of increased mortality. So we're trying  to open up this alveolus rather than having this, if when this alveolus is collapsed, it results in overdistension of this alveolus. If we can apply  pressure to this alveolus and recruit it, now the air can be distributed between these two, and it results in better distribution of ventilation. There's a  concept called barotrauma or volutrauma. People sometimes have arguments which is present, barotrauma or volutrauma, but you can see from the pressure-volume  curve that you can't have a pressure without a volume, and you can't have a volume without a pressure. So pressure and volume coexist, and hence we use the  term more specifically ventilator-induced lung injury. And it's really, it can be from adalectotrauma or it can be overdistension of alveoli.  The top portion of the compliance curve is representative of overdistending alveoli. There's an upper inflection point that shows beyond this you have  large changes in pressure with very small changes in volume. Not all patients are going to benefit from PEEP, however. If you look at the panel on the  bottom, the patient has atelectasis at 5 centimeters of water PEEP, and when we increase them to 45, we've recruited a significant amount of lung. You can see  there's improved aeration, but in the top panel the aeration really doesn't change, the atelectasis doesn't change much between 5 and 45 of PEEP. So not all  patients are recruitable, not all lung is recruitable, but even if they're not recruitable, PEEP oftentimes will improve oxygenation, but this often comes at the  cost of overdistending some alveoli in those patients who have non-responding lungs. This is an example of a poorly recruitable lung. This is a COVID patient  on ECMO, otherwise they wouldn't be able to be alive. You can't really see the lungs very well, so I've outlined them in red. They're completely atelectatic. We  probably actually could recruit these alveoli. It's just going to take a lot higher pressures than we're normally used to using. We're not going to  limit our plateau pressures to 30. If we gave him a pressure of 50 or 60, we could probably recruit some significant lung, but the alveoli that are already  open may suffer the consequences and we may get pneumothoraces associated with that. So there are lots of ways to manage PEEP. We can use PEEP tables where  oxygenation is the signal, and for a certain FiO2 level we're going to use a certain PEEP. Consequence of this is when we use PEEP tables on patients who  have less extensive lung disease, oftentimes they're overdistended. We can use compliance, but compliance doesn't necessarily equal optimal oxygenation. We  can use transpulmonary pressures, which involves utilizing an esophageal balloon, but this is not standard of care at this time. Plateau pressures is a target. CT  can be used to evaluate for recruitment. Whichever method is used, none of them have shown a mortality benefit. It begs the question, are we looking at the right  signal? Is oxygenation the best signal? Is compliance the best signal? There are many things that affect oxygenation. Shunt's one of them. Instead of  oxygenation, should we be looking at oxygen delivery? Certainly PEEP can affect your cardiac output by decreasing your right ventricular preload and  increasing your right ventricular afterload. These are questions that we need to have answered. So the optimal PEEP titration strategy is unknown as it  stands right now. Let's go to a different topic, at least title volume. Of title volume we have some pretty good answers. So years ago we used to give patients  15 or 20 mLs per kilo, and the theory was developed at that time that since they had a lot of lung that wasn't functional because of debris in the airways or  atelectasis, that maybe we should give them smaller lung volumes, perhaps treat them like they were a baby. We wouldn't give a baby a liter of title volume like  we would do adults. So that got us really thinking about using smaller title volumes. The ARMA trial compared 6 cc's versus 12 cc's and it found a 9%  reduction in absolute mortality. So this has pretty much been the standard of care for about 20 years. Let's look at what that's doing with our compliance  curve. If we're giving the patient on the left 6 cc's and the patient 12 cc's per kilo, they're both starting at the same PEEP level. And as they get their  inspiration, a patient at 6 cc's per kilo may go up to this position on the compliance curve, where a person at 12 may be getting over distended. This is  where we think the mortality difference comes into play. In the ARMA trial, we also found out that looking at the plateau pressure could give us some  information. The plateau pressure is a surrogate for alveolar distension and patients in the ARMA trial, if they had a plateau pressure less than 30, we found  this was associated with less volume trauma or barotrauma. If their plateau is 30 or greater, we're suspecting that this is representative of the patient being  at a higher position on the compliance curve. But then in 2015, Amado did a meta-analysis of the data from the ARMA trial, as well as a couple of other  trials, and he found out that the tidal volume didn't necessarily impact mortality. The signal that he found was that if the delta P was able to drop by  7 centimeters of water, this was associated with improved mortality, with the delta P being the plateau pressure minus our PEEP. Let's look at what  that really means. So if a patient had 12 mLs per kilo, and they had a plateau pressure of 34, you can see there the high position of the  compliance curve. The plateau greater than 30 suggests that we're over distending the patients, which the curve represents. As we decrease the tidal  volume to 6 mLs per kilo, we can see that we drop our position on the compliance curve, and this is reflected in our plateau pressure as well. The  pressure effects on the lung are going to vary based upon the particular compliance of the lung. If you have a patient, for example, who is on an  airway pressure of 16, in this patient that may be with ARDS that has very stiff lungs, we're still in a compliant portion of the compliance curve where  somebody who has more compliant lungs, they may be over distended. This is because chest wall has different stiffnesses in different patients. Some  patients are morbidly obese, or they may have an ileus, which causes their chest wall to be stiffer. This is why we're really interested in looking at  transpulmonary pressures. Right now we're using the esophageal balloon, but the research is still ongoing to see the best way to measure those transpulmonary  pressures. So not only does the tidal volume matter, but the location on the compliance curve matters as far as changing your delta P. In the patient on  the left, he still has 12 mLs per kilo of tidal volume, but his plateau pressure is 29, which is in the safe zone. You can see on his compliance curves, tidal volume  doesn't quite yet get to the stiff portion. If we were to decrease this patient to 6 mLs per kilo, he doesn't get a 7 centimeter water drop in his  airway pressures. This can just show how come it's not necessarily tidal volume that is telling us the answer. It's kind of the patient's position on the  compliance curve is the theory that Amato is essentially proposing. PEEP can also change the patient's position on the compliance curve, and it can change  the required driving pressure. If both patients on the left and right are getting the same tidal volume at 6 mLs per kilo, but we have a plateau  pressure of 34 on the left, it's because we started at a much higher PEEP, let's say 15. The patient on the right, if we decrease the PEEP from 15 down to 5,  keeping the exact same tidal volume, our plateau pressures will drop. So we can keep the tidal volume of the change the same, but change our PEEP, and that's  going to change our position on the compliance curve. It's going to change overall alveolar sustension as well. So two things are involved in the  mortality benefit, both the PEEP and the tidal volume, because they interact with each other as far as positioning the patient on the compliance curve. So if  we're going to titrate PEEP for compliance, we need to do a PEEP titration, and we can do it on inflation or deflation. The pressure required to  open an alveolus is more than it requires to keep an alveolus open. This is called tissue hysteresis, and it's largely representative of the forces  that are necessary to overcome surface tension in a closed alveoli to get it to open. Since we're primarily interested in the pressure that it  takes to keep an alveolus from closing, we're really wanting to do our PEEP titrations on the deflation limb. So you can do a best PEEP trial. This can be  done in pressure or volume ventilation. Classically, you're going to check a plateau pressure and check your PEEP levels and measure compliance at each of  these different PEEP levels. A way you can do it at bedside without necessarily calculating compliance, but it can give you a functional assay of this as well,  is if you put the patient, this is what I do, you put them into pressure AC, you keep the breath the same for each breath. So they're going to  get a tidal volume of 15, this instance applied for one second, and we'll do this 15 times a minute. We'll start at a PEEP of 23, and we get a tidal volume of 280.  I decreased my PEEP to 20 after a few minutes, and my tidal volume went up to 290. I further decreased my tidal volume to 17, and my tidal volume went up. At a PEEP  of 14, my tidal volume went up even more. I decreased my PEEP to 11, now my tidal volume goes down. So what this shows us is that when we were in a PEEP of 23,  we were in the over-extended portion of the compliance curve, and as we were decreasing our PEEP, we were coming down the compliance curve. When we got to 14,  this was our optimal PEEP, which we describe as being the lower inflection point. When we went to 11, we started derecruiting even more alveoli, and our  tidal volume started to drop. So a lot of people would try to target the PEEP for just at this point of the lower inflection point, or a little bit higher.  It used to be thought that the lower inflection point was where all the alveoli were recruited. This is not true. Alveolar recruitment occurs throughout  the inspiration, throughout your compliance curve. The lung is heterogeneous, however, and you'll be over-extending some patients in this  portion of the curve, and just opening other alveoli in the same portion of the curve. At this portion of the curve, you're going to get some recruitment, and  certainly some over-extension. So each alveolar unit has its own compliance curve, and then the entire lung has a compliance curve of its own as well. And  so we can't find the perfect PEEP, or we can't find the best PEEP, because we're always over-extending some alveoli and not recruiting others. It may be better  to think of having the best, rather than the best PEEP, let's talk about having a better PEEP. We're trying to ventilate the patient, however, within the middle of  the compliance curve, where they have the best compliance. So if we look at the tidal volume based up on plateau pressures, we want to keep our plateau  pressure less than 30 centimeters of water. Well, if our tidal volume is at 6 mLs per kilo, as in this instance, we can increase it to 8 and still be  below a plateau of 30. We can increase our tidal volume up to 10 mLs per kilo, as long as we're staying below this portion of the compliance  curve. One patient may take 10 mLs per kilo, another patient may only allow you to have 8. We need to look at these patients' positions on the compliance  curve to try to determine what's best for that particular patient. If you start at a higher person position on the compliance curve, that can limit your  tidal volume. In this instance, we have a PEEP of 15, and we can't go any higher in our tidal volume, whatever is set here, because we would be exceeding 30 mLs  per kilo. If we decrease our PEEP, we're able to increase the size of our tidal volume. If we increase our PEEP some more, our tidal volume can  increase even more, because we're still staying below that 30 centimeters of water. What we're doing is using lung mechanics to evaluate PEEP and tidal  volume. We're setting the PEEP above the lower inflection point. This is where we're starting to have a higher proportion of recruitable lung units.  We'll still have some that are getting over distended, and we're setting our tidal volume to limit our plateau pressures to less than 30. Well, let's  look at a scenario where we're using this clinically on our patient with ARDS. Let's say she had a PEEP of 20 and an FIO2 of 80 percent. Her plateau  pressure is 28, so she's just right at that border where we don't want to overgrow that. She's at 6 mLs per kilo. Her pH is 7.1, PCO2 is 45, and the PO2 is  70 with a bicarb of 14. We have adequate oxygenation in this situation, but we need more ventilation, both because we have VQ mismatch, which we overcome,  but we also need to compensate for a metabolic acidosis. So if we drop our PEEP, it will give us more room for tidal volume to increase our minute  ventilation. And you can still see with a lower PEEP, we're still maintaining the safe range of being less than 30 mLs per kilo. And a different scenario  on the same patient, if we had a PEEP of 10 and FIO2 of 100 percent, and we're running the patient at 8 mLs per kilo, plateau is 28, we have a  pH of 7.38, PCO2 of 38, PO2 of 55, and a bicarb of 22. Here, we're not adequately oxygenating the patient, we are ventilating adequately.  So we could increase our PEEP, and if we increase our PEEP higher on the compliance curve, it'll be necessary for us to drop our tidal volume  to stay less than our plateau goals of 30. But really what we're doing here is we're realizing that there's no best PEEP, it doesn't exist.  There's no best tidal volume as well. We're always looking to see where our patients are at on the compliance curve, what position they are based  upon using PEEP and tidal volume. The numbers that are out there are not absolute, however. If a patient has a plateau of 40, that may  be okay if they weigh 400 pounds, as oftentimes we exist or we see, or if they have an ileus. You don't have to be set with 6 mLs  per kilo, adjust it up or down based upon your particular patient.  We want to choose the best combination to achieve the goals of lung protective strategy of ventilation. We wanna oxygenate, provide adequate  ventilation, and avoid ventilator induced lung injury.  The PEEP tables, compliance curves, or transformatory pressures are all just starting points. Whichever one you wanna use is fine. I'm not gonna say  that one's better to another. But then think of the compliance curve  and figuring out where your PEEP and your tidal volume are on there, and that'll let you help. That'll help you make some clinical decisions  in managing your patient throughout the day. Thank you.

Video Summary

The video transcript discusses various aspects of ARDS, mechanical ventilation, and ventilator management. The moderator introduces the session and highlights the importance of understanding and treating ARDS, a common complication in COVID-19 and septic patients. The first speaker discusses the pharmacotherapy of ARDS, focusing on the use of neuromuscular blocking agents and corticosteroids. The second speaker talks about ventilator modes and management, highlighting the importance of lung protective strategies such as using low tidal volumes and appropriate levels of PEEP. The speaker also discusses the use of recruitment maneuvers and the benefits and limitations of different ventilation modes. The third speaker focuses on the titration of PEEP and tidal volume for optimal ventilator management, taking into account factors such as compliance, plateau pressure, and transpulmonary pressures. Overall, the session provides insights into the medications and ventilation strategies used for ARDS treatment and management. The speakers emphasize the importance of individualized treatment based on patient-specific factors and the goals of lung protective ventilation.

Asset Subtitle

Pulmonary, 2021

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 2021 Critical Care Congress held virtually from January 31-February 12, 2021.

Learning Objectives

Describe medications used in the treatment of acute respiratory distress syndrome (ARDS)
Explain ventilatory strategies in the treatment of ARDS
Discuss management of a patient with ARDS through case management

Medications Used in the Treatment of ARDS
Jonathan Van Cleve

Ventilator Modes for Treatment of ARDS
Richard D. Branson

Clinical Scenarios and Their Management of ARDS: Case-Based Learning
Eric Magana

Meta Tag

Content Type Presentation

Knowledge Area Pulmonary

Knowledge Level Foundational

Knowledge Level Intermediate

Knowledge Level Advanced

Membership Level Select

Tag Acute Respiratory Distress Syndrome ARDS

Tag Ventilation

Year 2021

Keywords

ARDS

mechanical ventilation

ventilator management

COVID-19

neuromuscular blocking agents

corticosteroids

lung protective strategies

PEEP

ventilation modes

	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.