false
Catalog
Multiprofessional Critical Care Review: Adult 2024 ...
4: Acute Respiratory Failure: Difficult Cases (Rob ...
4: Acute Respiratory Failure: Difficult Cases (Robert C. Hyzy, MD, MCCM)
Back to course
[Please upgrade your browser to play this video content]
Video Transcription
This lecture is entitled Acute Respiratory Failure, Difficult Cases. I'm Bob Heisey from University of Michigan. And this year, like last, given that COVID pandemic is still raging, I thought it'd be useful to discuss the following, discuss whether respiratory failure in COVID-19 is unique, and recognize how respiratory effort can contribute to lung injury and ways to assess and mitigate this effect. Well, as I said, the pandemic is now, gosh, a year and a half old, and there are early reports coming out at Lilly who got hit sooner than the U.S., that perhaps these patients are different, perhaps they're qualitatively different. And the issue related to the fact that it was felt these patients had a lower degree of elastins, in the converse of which is to say a higher degree of compliance than our usual ARDS patient. In spite of that, they had a fair amount of shot. And so this led to this notion of two separate sub-phenotypes. Perhaps some had near normal lung compliance with isolated viral pneumonia, and then other, the more traditional ones, had decreased pulmonary compliance, which more potentially reflected the natural history of the disease, or alternatively, and as we'll discuss, might have been made worse by strenuous respiratory effort before intubation, something called patient self-inflicted lung injury, or PCILI. When you do look at patients and you control for variables, you see differences, admittedly, with a wide degree of overlap, suggesting that there's really more of a quantitative difference than a qualitative difference. Here's a panel to the left, looks at ARDS patients of a more pre-COVID era, and then the brown is a COVID ARDS. When you control for oxygenation in terms of PDF ratio, you see a higher compliance in the COVID patients. But again, a lot of overlap. When you conversely match patients on compliance and look at oxygenation, you see lower PDF ratios. So these are different, in fact, but there is still a lot of overlap. And I would argue this is more of a quantitative rather than outright qualitative difference. There is also, when you look at CAT scan, greater aeration, meaning to say less areas of compressive atelectasis. This is the work of Gattinoni looking at the CT scans in this COVID-19 population. So it's important to discuss some of the current concepts of lung injury, this notion of VILI and the VILI vortex of stress and strain of mechanical inhalation, or not mechanical inhalation, the stress and strain produced with spontaneous breathing or ventilation contributing to lung injury. And it's the energy that is distributed into the lung that determines lung injury. The way to think about this, perhaps to look at the two patients driving pressure of 20, and yet the volume of the overstretched lung here would lead to more in a compliant lung conventional static variables of plateau pressure of 25 and driving pressure of 20. So the idea here, the analogy here, if you will, is that static variables, such as plateau pressure and driving pressure, while evidence-based and I think important, may not tell the whole story with regard to lung injury, that more dynamic variables would be important, sort of analogous to fluid resuscitation and substance, if you will, where CVP is less important than straight leg raise, for example. This led to a number of early thoughts regarding this condition as to how one actually should approach them differently. So the notion was that only a minority of patients had ground glass and that hypoxemia was more due to VQ mismatch, perhaps due to active microthrombolemboli, and that high tidal volumes with higher transponder pressures, if utilized, could worsen the lung injury by distributing more mechanical energy to the lungs. And as an adjunct to this, includes a patient's pre-intubation who are stressing and straining. In other words, a strenuous respiratory effort would lead to wide transponder pressure swings, which generate large tidal volumes and worse lung injury. This was a concept that was developed prior to the COVID pandemic, but these differences in compliance were ready-made to have the application of this PCILI notion, this patient self-inflicted lung injury to the COVID pandemic. Now, is there evidence for a PCILI in COVID-19? This observational series found that patients with a high driving pressure, despite having the same P to F ratio at admission, had higher ventilatory ratios, suggesting more microthrombolemboli and dead space, and had a lower survival. In other words, using heated high flow and keeping them off the ventilator, potentially being even detrimental in this subset. I wouldn't say this is compelling, this is not perspective, but at least as a hint of some veracity to these notions. This led to a whole array of notions. This is John Rainey and Luciano Gannoni in JAMA in the height of the pandemic, making various recommendations for how to treat a COVID patient. One was hearkening back to PCILI here, was to option it to, I'm sorry, invasively ventilate them sooner. Don't let them breathe spontaneously or worse lung injury. And to their way of thinking, perhaps take patients with less alveolar filling and make it worse. Again, these are unproven concepts. These are early concepts and I'm not necessarily endorsing them, but I am trying to convey to you how some of these controversial aspects of management were conveyed. And during mechanical ventilation, the idea was to not give high PEEPs, to avoid this ventilator-induced lung injury, to avoid imparting energy by minimizing transpulmonary stresses. After intubation, to use lower tidal volumes and high PEEPs only in a subset, really, to try to sub-phenotype these patient populations. So in the highly compliant patients, you use higher tidal volumes and lower PEEPs than the more conventional alveolar filling, stiff lung patient population, the more conventional population. So it was controversial then to potentially, to not only intubate people early when they may not otherwise meet criteria, but to use larger tidal volumes than one might ordinarily consider with conventional lung protective ventilation. And then the weaning phase, to maybe not wean people as aggressively because all that spontaneous breathing might perpetuate lung injury. So these are untested concepts and this occurred early. I don't endorse them, but it certainly created a huge degree of interest in how to best, everyone wanted to manage a COVID patient the best, and maybe there should be some differences in how we approach them. Now, I'll harken back to only tell you as I did in my earlier lecture that the Arsenic ventilation always allowed you to go up to a tidal volume of eight per kilo, ideal body weight provided plateau was under 30. So I'm not sure that necessarily was an attempt to not allow for some liberalization in tidal volume and 60 cc per kilo is not rigid here. That is true, by the way, that even when we look to ARDS patients prior to the COVID pandemic, we knew that there were subsets here based on compliance. This was an attempt to look in three categories, but there were patients with low elastin site, high compliance here that were perhaps not as sick, but in the pre COVID world, we recognize that there was a spectrum. And again, in conventional Arsenic ventilation, tidal volumes were allowed to go up as high as eight. You can see here, 8.49, perhaps a little higher than eight in order to keep the plateau pressure low. So really it turns out that these, the high elastin and lower elastin site, either the low compliance and high compliance, because that's the converse concept are not mutually exclusive. There is just a whole lot of overlap here. So when we think about it, there's a number of things that are occurring in the ARDS patient that pertains to respiratory effort. Now we know that just blindly paralyzing patients, at least in a non COVID setting by the ROSE trial, that early neuromultiplicate doesn't make a difference, yet we know diaphragmatic dysfunction, transpulmonary pressure, lung overdistension, and imparting energy to lung in terms of mechanical power, having a high and excessive respiratory drive, more than that in just a minute, a high respiratory rate, which conventionally were tolerated in ARDS ventilation. All these things can be ways of potentially imparting more energy in the lung and thereby resulting in perpetuation and exacerbation of the underlying lung injury. In fact, even something so simple as prone ventilation can have less intensity of spontaneous breathing, which can then contribute to less transpulmonary pressure. Maybe one reason prone ventilation works is not only redistribution of lung volume, ventilation in the prone position, but also maybe less mechanical power with less stress and strain due to alterations in spontaneous breathing effort. So let's drill down here a little bit and talk about mechanical power. So here's the formula for it. It's kind of fancy, but the key variables are transpulmonary pressure. We talked about that already with this silly notion of why transpulmonary swings with tidal volume and driving pressure are important respiratory rate, critical variable, flow, and PEEP all contribute to the calculation of mechanical power. The question is how much energy is being imposed on the lung as a key variable for lung injury. And this does suggest, by the way, given the fact that tidal volume benefit is now proven to be more effective with low lung compliance, that there might be some veracity to this notion of getting away with slightly large tidal volumes. Again, this is not a variance with ARDSNet ventilation, but it does emphasize the point that perhaps that the future of management of the heterogeneity of lung recruitment and distribution of gas in ARDS would be on a more personalized basis. So there's some data here, and you can't see the reference all the way. Lancet Respiratory Medicine from I-Corps Registry, a very large database in Toronto. Again, this is not perspective. This is observational, but they looked at mechanical power and various other parameters, and they found that a higher ICU mortality is associated with lower pre-death ratio. That makes sense. The worse suctionation, the worse a patient would be expected to survive. Higher ventilatory ratio, which would be another indication of dead space. We knew this already, right? Dead space in the ARDSNet original trial, different quartiles of dead space, higher mortality. But driving pressure, we've seen that before, over 15. And then finally, mechanical power, more than 17 joules per minute, being associated over time with higher mortality, and that's what the data look like. So we're talking about exposure. We're not talking about one single measurement. We're talking about repetitive measurements over time. Now, this is hypothesis generating. Just as with driving pressure, we don't have a perspective study targeting mechanical power as, an end point for our approach to mechanical ventilation. But I do think in terms of proof of concept that this dynamic variable of mechanical power may certainly be important for the worsening of lung injury in our ARDS patient population. Now, even though driving pressure is a static variable, as I said, mechanical power might be better to measure according to the I-Corps criteria, both really are associated with mortality. And I guess what I wanted to point out now is not only is driving pressure important, but changes in driving pressure with ventilator changes. In other words, I'm fond of saying that driving pressure is a surrogate for recruitability, recognizing that not all ARDS patients will respond to PEEP. You can see here two different trials that we're familiar with. The ARDSnet alveoli trial, the high P versus low P trial, and then the European Express trial. If you look at the left-hand panel, you see the changes in driving pressure. So it's a little confusing, right? Delta P is the abbreviation for driving pressure, but delta delta P means the change in driving pressure. And if driving pressure goes up, God forbid, with higher PEEPs, death will likely go up versus the opposite, right? In other words, if you go up on PEEP, driving pressure goes down. If you're able to get that lower, then mortality will decrease. Meanwhile, oxygenation will improve. And so mechanical power and delta P are predictors of mortality. That's true in I-Corps, that's also true in the series from intensive care medicine. Now it's felt here that they're really looking at two different things. Each of which independently predict mortality in these two different series. As you can see, the number of patients in the sample here, the airway pressures here for delta P versus power, which is measured over here and mortality going up. So perhaps these are two different parameters. But a recent paper, and this just came out in March in the Blue Journal, suggested that you can actually use delta P and respiratory rate in combination as a surrogate for mechanical power. So maybe these two variables are actually linked and certainly the calculation is easier. Four times delta P plus respiratory rate. And this also points out, going back to one of the earlier slides, how higher respiratory rates are more likely to impart energy into the lung and make lung injury worse. Now that's a little bit of an odd concept because when ARMA first came around, the original RSF ventilation, they were allowing patients to breathe 35 times a minute. And that kind of ran in the face of what many people used to do, which is to say sedate or paralyze people to keep them breathing so fast. So maybe that was the right thing to do in hindsight. But the idea here is that both driving pressure and mechanical power are associated with mortality. Again, this is post hoc data analysis, not perspective management based on these variables, that they're perhaps either independent or perhaps even linked in a quick and dirty, if you will, way to estimate mechanical power. Four times delta P plus respiratory rate, such that an increase of one delta P need to be offset by a decrease in respiratory rate of about four breaths per minute in order to not increase mortality. So the idea here is if you are actually losing ground here while managing your patient and driving pressure is actually increasing, you better knock down their respiratory rate through sedation or other means. So it really is a fine balance between minimizing, I should say, stress and strain in the lung and optimizing diaphragmatic effort and synchrony, right? There's a lot also that's written around synchrony. And does asynchrony also contribute to stress and strain through double triggering and other events of that nature? Is diaphragmatic injury, which is now being studied in terms of atrophy with diaphragmatic ultrasound also part of this whole process? So recurrent asynchrony is another thing that has been studied and it's certainly associated with worsening mortality. Now, double triggering is the most common kind of asynchrony seen, and you can increase flow rate or tidal volume or sedation to overcome it, provided you still stay in the lung protective means. But both ARDS and dexamethamine were associated with recurrent asynchrony in this particular observational series. So they looked at patients over time. This is not just one observation. They felt that recurrent asynchrony, if they had two or more episodes, two or more times. And so asynchrony is probably... Now, again, this also flies in the face of some data because if asynchrony is so bad, then why did the ROSE study looked at neuromuscular blockade early on not work? Well, I mean, these are the data and this is what it shows. So double triggering is a bad thing and is associated with worsening outcomes. We just don't know if suppressing it alone prospectively would be associated with better outcomes. Reverse triggering can also lead to breath stacking. So asynchrony is part of the game. Ventilator movements, spontaneous breathing, these are all considerations one needs to make. This was a nice series from earlier this year, pretty much full of diagrams about recognizing asynchrony, which I could recommend to you. In addition, we talked a little bit about PCILI and that transpulmonary pressure. How hard is the diaphragm working? Changes in esophageal pressure. Now, this is not invasive, not invasive mechanovulation, but predict NIV failure at 24. So clearly how hard you're working to breathe can be predictive. So if you're going to target a change in pressure, should you use an esophageal balloon? In other words, this was a non-invasive study. It's easier to put an esophageal balloon in a intubated patient than in a patient on NIV who's on a face mask. But it begs the question of drive to breathe, not just measuring respiratory, what's the drive to breathe? And is there a shortcut way one can get a grip on that? And in fact, there seems to be. Now, P.01 is the respiratory effort made the first 10th of a second. It's felt to correlate quite well with the drive to breathe. So, you know, this was an interesting thing that was looked at as a weaning parameter decades ago and kind of dropped when the frequency tau on RISB index came around and as a weaning parameter. But it seems to have a little bit of resurgence now on ventilated patients or at least patients on receiving ventilators that could be non-invasively as indicative of a respiratory drive. And we talk about P. cili off the vent or drive to breathe on the vent as potentially injurious in terms of transmission of mechanical power through transpulmonary pressure. Maybe there is a way forward, and this is more speculation at this juncture, to look at P.01 as a therapeutic target to mitigate some of the stress and strain imparted into the lung. So do you really need a balloon is the question. We don't have an answer to that, but there have been increasing amounts of literature on P.1. I thought I'd share that with you. I think we're gonna see more of this in the days ahead. And this is just another example of P.1. So what have I told you? That lung low tidal volume ventilation is so important, especially in patients with low lung compliance, which is true of most non-COVID ARDS patients as well as many, many COVID ARDS patients as well, acknowledging that you can go up a smidge on the tidal volume at times up to eight, perhaps. We also think that extreme respiratory effort may be detrimental in patients with transpulmonary pressure, pressure swings, pleural pressure swings, in patients without, which would be the PCILI notion or with endotracheal tubes, asynchrony, for example. We think changes in driving pressure of lung recruitment are important and should be considered along with the respiratory rate in terms of trying to understand issues of how much power is being transmitted to the lung and also whether or not lung recruitment has successfully occurred. And we're waiting to see if some of these fancier or newer measurements and more invasive esophageal pressure or pleural pressure or the non-invasive P.01 if applied as therapeutic targets for management result in improved outcomes. Thank you.
Video Summary
This lecture discusses acute respiratory failure, particularly in the context of COVID-19. The speaker explores whether respiratory failure in COVID-19 is unique and how respiratory effort can contribute to lung injury. The lecture highlights the different sub-phenotypes observed in COVID-19 patients and the potential role of patient self-inflicted lung injury (PCILI). It also discusses the importance of assessing and managing lung injury, including the concept of mechanical power and its association with mortality. The lecture touches on the significance of asynchrony and the impact of respiratory effort on lung injury. It suggests the use of parameters like driving pressure and respiratory rate to estimate mechanical power and optimize ventilation strategies. The lecture concludes with the exploration of potential therapeutic targets to mitigate stress and strain on the lung, such as the measurement of P.01 as an indicator of respiratory drive.
Keywords
acute respiratory failure
COVID-19
respiratory effort
lung injury
mechanical power
patient self-inflicted lung injury
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.
×
Please select your language
1
English