Transcript
Speaker: Welcome to Critical Matters, a sound critical care podcast covering a broad range of topics related to the practice of intensive care medicine.
Speaker: And now, your host, Dr. Sergio Zanotti.
Speaker: Lung protective mechanical ventilation, including low tidal volumes and limitation of plateau pressures,
Speaker: has been the basis for treatment of patients with ARDS.
Speaker: The use of early neuromuscular blockers has been proposed in severe cases as an adjunct treatment to improve outcomes.
Speaker: The adoption of this strategy has been inconsistent despite a positive clinical trial published in 2010.
Speaker: In this episode of Critical Matters, we will discuss this topic in detail, considering the results of the recently published clinical trial, the re-evaluation of systemic early neuromuscular blockade, or the ROSE trial.
Speaker: Our guest is Dr. Arthur Slutsky, a true world-class expert in ARDS.
Speaker: Dr. Slutsky is a scientist at the Kenan Research Center for Biomedical Research at St.
Speaker: Michael's Hospital in Toronto.
Speaker: He's the Kenan Chair in Medicine at St.
Speaker: Michael's Hospital and a professor in the Departments of Medicine, Biomedical Engineering, and Surgery in the University of Toronto.
Speaker: Dr. Slutsky is a prolific investigator with an interest in ARDS.
Speaker: Specifically, he has focused on the area of ventilator-induced lung injury.
Speaker: His work established the concept of biotrauma and has helped advance our overall understanding of ventilator-induced lung injury and ARDS.
Speaker: It's a true honor and pleasure to have him on the podcast.
Speaker: Art, welcome to Critical Matters.
Speaker: Thanks very much, Sergio.
Speaker: Pleasure to be here.
Speaker: Looking forward to the next little time while we chat.
Speaker: Excellent.
Speaker: So you recently wrote an editorial for the New England Journal of Medicine entitled Early Paralytic Agents for ARDS.
Speaker: Yes, no, and sometimes...
Speaker: And I thought that maybe we could use this as a framework for our discussion and maybe start with the yes, which was, I mean, back in the 2010, when the ICURASIS trial came out of France, you also had the opportunity to talk, to write an editorial for this study at the New England Journal of Medicine.
Speaker: And maybe at that time, you were thinking that there might be potential benefits and outcomes of using early namascar blockers.
Speaker: Could you kind of tell us what was the thought back then and maybe give us a synopsis of that study?
Speaker: Super.
Speaker: Maybe I'll start with a bit of background because I think that's what you're getting at.
Speaker: In other words, what's the rationale for why the use of neuromuscular blocking agents might be beneficial?
Speaker: And I think it relates to what you said in the introduction, and that is there's a hypothesis that it could potentially decrease ventilator-induced lung injury.
Speaker: So, you know, I think ventilator-induced lung injury has sort of grown in importance, if you like, over the last 20-plus years.
Speaker: I would say largely related to the study from the ARDSN investigators published in 2000 in New England Journal, showing that 6 ml per kilo is better than 12 ml per kilo.
Speaker: And again, you sort of hinted at that.
Speaker: So we know that mechanical ventilation can cause iatrogenic lung injury.
Speaker: You know, and the forms are barotrauma, volutrauma, adlec trauma.
Speaker: and biotrauma with release of mediators.
Speaker: But despite that trial, I was published in 2000, mortality from ARDS remains high.
Speaker: And it's thought that some of that mortality, not all of it, but some of it may still be related to ongoing ventilator-induced lung injury, despite the use of so-called protective ventilation strategies.
Speaker: So that's sort of the rationale behind why one might use neuromuscular blockers.
Speaker: So nerve muscle blockers could make it easier to help ventilate patients.
Speaker: Sort of an obvious thing they could do is decrease the oxygen cost of breathing.
Speaker: When people have got respiratory failure, they're working very hard, their respiratory muscles actually can use a fair bit of the cardiac output and oxygen consumption.
Speaker: It can decrease excessive muscular activity and then have less patient ventilator synchrony, sort of the patient fighting activity.
Speaker: the ventilator, so less the synchrony would be a major potential mechanism, can help increase oxygenation.
Speaker: There's some studies that suggest that if you use neuromuscular blockers, oxygenation goes up a little bit.
Speaker: And there's a little bit of evidence that cis-aticurium, which is the neuromuscular blocking agent that was used in both these studies, has some anti-inflammatory effects.
Speaker: I don't think this last one is likely to be the case, even though we published a paper on this a couple of few years ago.
Speaker: But, you know, if it was an anti-inflammatory effect, you'd think things like steroids, et cetera, would be, you know, better than cis-aticurium.
Speaker: So in terms of the background for the ACURIS study,
Speaker: The study was, the first author was Papazian from France, and his group has done a lot of work on neuromuscular blocking agents and had done a lot of work on neuromuscular blocking agents.
Speaker: They published an early study before they started the one that was published in 2010, and they showed that neuromuscular blocking agents could improve oxygenation.
Speaker: The first author, I think, was Gagné, also from that group.
Speaker: And they also had a very interesting study.
Speaker: The first author was Forel, F-O-R-E-L.
Speaker: This was a small randomized control study showing that if they gave 48 hours neuromuscular blocking agents, it decreased biotrauma.
Speaker: There was a decrease in bronchoalveolar lavage levels of a couple of cytokines, I think IL-8, and also in a number of serum cytokines.
Speaker: I think it was IL-6 and IL-1 beta.
Speaker: So that suggests that maybe, because that's what you'd expect if the agent decreased biotrauma, that you'd have a decrease or a change in these mediators.
Speaker: So that was, I think, the background to why Papazian and colleagues sort of did the study called Acurasis.
Speaker: And that's the one that was published in 2010.
Speaker: So just to briefly summarize it, it was a randomized controlled trial.
Speaker: comparing 48 hours of cysadecurium to placebo in patients with ARDS who had a PF ratio less than 150.
Speaker: That's pretty important.
Speaker: So they're dealing with a sicker spectrum of ARDS patients.
Speaker: Their primary endpoint was 90-day in-hospital mortality, and they said adjust it for some predefined covariates.
Speaker: This was a multicenter trial published in about 20 French intensive care units.
Speaker: An important feature was the sedation practices.
Speaker: All patients were sedated heavily to a Ramsey score of five or six, so six, I believe.
Speaker: And you'll see, I'll come back to that.
Speaker: That becomes potentially a really important point here.
Speaker: The control group and the active group were heavily sedated.
Speaker: Then they used a bolus, a cystatic curium, and 48 hours of a continuous infusion.
Speaker: They used volume assist mechanical ventilation with small tidal volumes, 6 to 8 ml per kilo.
Speaker: And they used the low PEEP protocol, PEEP-FI2 protocol that was used in the original ARMA trial that I mentioned that was published in the New England Journal in 2000.
Speaker: And that actually, I'll come back to that point as well later.
Speaker: And they looked at side effects, including barotrauma.
Speaker: And of course, they looked at whether there was
Speaker: muscle strength problems that occurred after the use of cystetic curing, because that's one of the major concerns that clinicians had in using it.
Speaker: They had a sample size of about 400, somewhat under 400 patients.
Speaker: And as I said, the primary analysis was 90-day mortality after adjusting for a couple of baseline covariates.
Speaker: So that was the sort of the outline of the study.
Speaker: It was, I think, a well-carried-out study.
Speaker: Interestingly, in all studies like this, they excluded a lot of patients, but 1,000 patients were excluded.
Speaker: That's not unexpected.
Speaker: There's all sorts of reasons why patients might be excluded.
Speaker: In the end, they ended up randomizing 300 to 400 patients.
Speaker: The median time from inclusion to inclusion
Speaker: The randomization was 16 hours.
Speaker: And the main result was they found a significant decrease in mortality, 90-day mortality, of about absolute decrease of about 8%.
Speaker: And that was important.
Speaker: That's a big difference in mortality, especially in patients who are receiving lung protective ventilation to start with.
Speaker: They had a couple of other interesting things I think that are worth pointing out.
Speaker: One was when they looked a little more detail, and this is a little more detail, the beneficial effect was limited to patients with PF ratios less than 120.
Speaker: So between 120 and 150, absolutely no difference.
Speaker: Less than 120, there was a big decrease in mortality.
Speaker: They also had more ventilator-free days, more organ failure-free days.
Speaker: Interestingly, in the control group, they had a higher incidence of pneumothorax.
Speaker: which again suggests that the underlying mechanism was related to decreased ventilator-induced lung injury.
Speaker: And importantly, they didn't find any difference in ICU-acquired paresis.
Speaker: So it didn't look like just 48 hours of paralytic aging was sort of really bad in terms of muscle weakness.
Speaker: Two other important things that I'll come to later on, and that is almost half the patients were treated with a prone position.
Speaker: You know, the prone position, there was a study published in the New England Journal a few years ago showing that the prone position markedly decreased mortality.
Speaker: And in Europe and in France in particular, the prone position is used a fair bit.
Speaker: That's different, interestingly, than what is done in North America.
Speaker: And it may be something we want to chat about, Sergio, that how little is used in the United States.
Speaker: The second interesting part, I thought, in the results,
Speaker: was if one looks at the Kaplan-Meier curves, if you look at the Kaplan-Meier curves in terms of mortality from that 2010 study, here you have an intervention that's being given for 48 hours and then the treatment is the same for both groups, but the difference in mortality doesn't occur until day about 18.
Speaker: So those Kaplan-Meier curves are spot on.
Speaker: They're just right on top of another, and they don't separate till 18 days, which is really, it's hard to, I find it hard to explain.
Speaker: In the editorial I wrote in 2010, I said, well, it could be related to biotrauma.
Speaker: We know that with biotrauma, you get the release of mediators.
Speaker: Those mediators go into the lung.
Speaker: They can cause injury, but more importantly, the mediators can get into the systemic circulation.
Speaker: and cause end organ failure.
Speaker: And that end organ failure is going to take some days or some time after release to happen.
Speaker: 18 days, quite frankly, seem like a really long time, like really, really hard to explain.
Speaker: So that's a summary of what the study looked like.
Speaker: In terms of mechanisms,
Speaker: I think that I sort of hinted at some of them previously.
Speaker: I thought that, and I think most people thought that the way this worked was by decreasing ventilator-induced lung injury, decreasing the synchrony, and also decreasing the, even though you're on a ventilator, if you're not paralyzed, you can generate large negative pressures that can cause large tidal volumes
Speaker: even though you set the ventilator to give you 6 ml per kilo, but the patient's actually getting a lot more than 6 ml per kilo.
Speaker: So I thought that that was probably the mechanism that, you know, patients can double trigger.
Speaker: So you have the assist control set for 6 ml per kilo.
Speaker: Patient gets a breath.
Speaker: Before the end of the breath, they trigger another breath and they end up
Speaker: what's called breast stocking, and you actually then can have a 12-unleopard per kilo or more, and that might then increase ventilator-induced lung injury.
Speaker: So that was sort of a summary, if you like, of the brief summary of the trial, including some of the rationale for why it may have worked.
Speaker: And it's interesting, Art, that this is obviously almost a decade old, but even though this was positive,
Speaker: the penetration of early neuromuscular blockers in practice is variable.
Speaker: And it's something that we also see with prone positioning, as you mentioned, which is another story of getting people to practice based on the best available evidence.
Speaker: But clearly there were still a lot of concerns regarding safety.
Speaker: And as you mentioned, regarding the targets for sedation that are currently being employed in the ICU, that I think there was enough clinical equipoise for a larger study
Speaker: trying to look into this.
Speaker: And I think that takes us to the ROSE trial, which is the recently published study that you also wrote in an editorial.
Speaker: And could you tell us a little bit about the ROSE study and what they did?
Speaker: And then maybe we can start exploring some of the differences between these two studies.
Speaker: Super.
Speaker: Yeah, I will.
Speaker: And I do think the other reason that people were a bit skeptical was that point I mentioned about the Kaplan-Meier curse separating so late.
Speaker: Like, why did that happen?
Speaker: Could it have been fluke alone?
Speaker: Who knows?
Speaker: So the Rho study was an NIH-sponsored trial, very well-carried-out trial, comparing 48 hours of cysatocurium to placebo.
Speaker: Again, similar to the ACRIS study.
Speaker: It was carried out in 48 to 50 hospitals in the US.
Speaker: It enrolled patients with ARDS who had PF ratios, again, less than 150.
Speaker: So the same cohort that was in the previous trial was a little different.
Speaker: It tried to include patients earlier in the course.
Speaker: So patients were identified in the emergency department.
Speaker: So they were entered into the trial
Speaker: probably a little earlier than patients would have been entered in France in the accuracy study.
Speaker: The important point here, one important point is related to sedation.
Speaker: And in the neuromuscular blocking group study, the cystatic curium study, they again used heavy sedation to a Ramsey score of about six.
Speaker: So it was very similar to the accuracy study.
Speaker: But the big difference was in the control, which was now light sedation, because that's how most of us now practice.
Speaker: We try to keep sedation as low as possible.
Speaker: That was different sort of 10 years ago.
Speaker: So that was one major difference between the studies.
Speaker: And then they used the same drugs, the same dose.
Speaker: They used, in terms of the ventilation strategy, they used volume assist mechanical ventilation.
Speaker: Tidal volume again of six to eight mL per kilo, similar to the French study and to the original ARDSnet study.
Speaker: But another difference they had though in the protocols was they used the high PEEP protocol.
Speaker: If you remember in the ACURAS study, they used the low PEEP protocol.
Speaker: And the difference in PEEP levels is sort of three to four centimeters of water.
Speaker: By the way, that may be actually important as we discuss things a little later on.
Speaker: They powered the study to a sample size of 1,400 patients.
Speaker: That's the advantage of a lot of these American studies, like really large sample size, which is, I think, really important.
Speaker: And they had a stopping rule for futility or efficacy.
Speaker: In other words, they had a rule, they had a data safety monitoring board that would look at the evidence all the way through the trial.
Speaker: And at some point, based on these rules, say, you know what, this is so likely to be positive, we're going to stop early, or it's so likely to be futile,
Speaker: that we're just going to call the study.
Speaker: There's no use entering more patients because we know it's going to be futile.
Speaker: So that was the trial.
Speaker: The results, interestingly, the trial was stopped early for futility after about 1,000 patients were enrolled.
Speaker: Interestingly here, they screened about 5,000 patients to randomize that 1,000.
Speaker: As I said, that's not unusual.
Speaker: But I thought what was interesting was about 650 of the patients
Speaker: were excluded because they were already receiving neuromuscular blocking agents.
Speaker: So despite the fact we talked about neuromuscular blocking agents weren't used routinely by all physicians, a fair number were still using it.
Speaker: 650 were excluded because of that.
Speaker: They also looked at 90-day mortality.
Speaker: And at the end of the study, the mortality in the two groups was virtually identical at about 42%, like
Speaker: Again, the Kaplan-Meier curves are essentially lay on top of one another.
Speaker: And they did some sensitivity analyses looking at PF less than 120, no difference, all sorts of other things.
Speaker: Absolutely could not find no difference between the two groups.
Speaker: They also, when they looked at the long-term outcomes, included health-related outcomes up to one year, they were the same in both groups.
Speaker: Again, no increase in muscle weakness.
Speaker: That wasn't a concern.
Speaker: and basically showed this is a very, very, very negative study, quite frankly.
Speaker: A couple of aspects of the results I should highlight that in this study, the prone position was used only at about 12 or 13% in both groups.
Speaker: That's versus about 45% inaccuracies.
Speaker: So that's a pretty big difference.
Speaker: The peep level in this study on average was about 13 centimeters of water.
Speaker: And that compared to about 10 in Papazian.
Speaker: So the higher PEEP-FI-2 table did lead to higher PEEP levels.
Speaker: In terms of complications, they actually had somewhat higher complications, cardiovascular complications in the cystatic curing group in increased serious cardiovascular events.
Speaker: These were small numbers, but they were statistically significant.
Speaker: So maybe there was something going on.
Speaker: And as I said, ICU-acquired weakness was similar between
Speaker: the two groups.
Speaker: So there you have two, I think very well carried out studies.
Speaker: One positive, one very, very negative, as I said, no matter how they looked, sliced the data, they couldn't get a, they couldn't find a positive result.
Speaker: Actually there's one positive result looking at, I forgot the exact result, but it was, it was likely due to chance alone.
Speaker: I think it had to do with, with race and, but it was by chance alone.
Speaker: So you have two well-done studies that give quite different results.
Speaker: And I think that obviously this is not an uncommon occurrence in medicine and certainly in science.
Speaker: I think one of the big issues that a lot of scientists outside of medicine have also commented recently is the inability to replicate findings of positive studies in other fields as well, which I think raises another question that we can maybe touch at the end.
Speaker: But going back to the neuromuscular blockers and ARDS conversation, as a consumer, if I were a patient, and I know you can't really compare like this, but you have four groups, right?
Speaker: The two control groups were different in terms of how they got ventilated and how they got sedated.
Speaker: And the two intervention groups were a little bit different in terms of the use of prone positioning and how much PEEP they got.
Speaker: But the sedation and neuromuscular blockers were very similar.
Speaker: And overall, I think that in general, we could argue that even though they use different definitions, that they were all patients on the sicker spectrum of ARDS.
Speaker: But when you look at the raw mortalities, it seems that the best group to be in was heavy sedation, lower PEEP, and early neuromuscular blockers, which is the Papazian intervention trial with a 30% mortality.
Speaker: Any comments on that?
Speaker: Yeah, I think it's very, very difficult.
Speaker: to compare mortality among studies.
Speaker: Because subtle differences in the entry criteria can really make a big difference.
Speaker: And in this case, for example, the ROWS study, as I said, included patients that came from the emergency department very early.
Speaker: They included patients very early.
Speaker: If you think about it, a number of those patients were included early.
Speaker: Potentially, if they had been in the accuracy study, they could have died before even being randomized because they weren't included as early.
Speaker: So I just think you have to be very careful in terms of comparing the data, the mortality, the absolute mortality data between studies, even in the same center at times.
Speaker: But now we're talking across two different continents.
Speaker: We're talking a decade apart where all sorts of things could have changed.
Speaker: So I think I'd be very careful about taking the absolute mortality differences and saying that they hold sway.
Speaker: It's interesting to think about what the mechanism might be.
Speaker: For example, in some studies I've seen, the entry criteria, and I don't think that was the case here, but if the age cutoff is just slightly different,
Speaker: that can have a huge impact on outcome.
Speaker: The other reason could be, if you look at the number of patients, for example, who were entered because of trauma versus sepsis, the mortality between ARDS patients with trauma versus sepsis can be a factor of three.
Speaker: So there's lots of things you have to bear in mind.
Speaker: And I think you have to be very, very careful with those kinds of comparisons.
Speaker: And I think that's why I always, I mean,
Speaker: you have to look at all these different aspects, and I think it's not so simple as just looking at what the result was in the abstract.
Speaker: But one of the things that I think is very interesting, and you talked about in the editorial you wrote for the Rho study, was this idea of reverse triggering.
Speaker: And previously, obviously, you've published a lot, and even, I mean, when we were thinking about the accuracy trial, we're thinking of the traditional injuries such as
Speaker: um, atletic trauma, barotrauma and biotrauma and the dyssynchrony driving these.
Speaker: But could you tell us a little bit about how you think that this might be a play a role in terms of what is reverse triggering and why maybe the levels of sedation had an impact on this?
Speaker: Sure.
Speaker: So this is, first of all, just to be clear, this is just a hypothesis.
Speaker: And, um, you know, uh, when the editor asked me to write the editorial, he said, you know, write something controversial that, that,
Speaker: could potentially explain the difference between the trials.
Speaker: And I was, you know, we were given 700 words to do this, Jesus, Filara, and I, when we did it.
Speaker: So you're sort of limited.
Speaker: But I think in terms of reverse trigon, we thought about what is there about the two trials that's different?
Speaker: And in fact, in the editorial, we had a table that summarized some of the things.
Speaker: And we thought,
Speaker: which of these could potentially cause the differences we saw?
Speaker: So, you know, there's a few things that I can talk about besides reverse triggering, but let me start there.
Speaker: So, first of all, reverse triggering is an interesting phenomenon.
Speaker: It was only just described about five or six years ago by a French group.
Speaker: And the idea here is that patients who are on the ventilator,
Speaker: sedated, the ventilator provides a breath and that by whatever the mechanism, the mechanism is not entirely clear, that causes a reflex phenomenon that leads to another, to the patient then initiating another breath.
Speaker: So that's why it's called reverse trigger.
Speaker: Normally the patient goes first, triggers the ventilator.
Speaker: Here the ventilator goes first and triggers the patient to take a breath.
Speaker: And if those breaths occur pretty close in time,
Speaker: before exhalation has occurred, you can get an increase in... You can get breast stacking, basically.
Speaker: So it's an interesting phenomenon.
Speaker: And there's a couple of, you know, I think critical issues about reverse triggering that made us think that this might be part of the explanation.
Speaker: First of all, it's probably...
Speaker: more common than we think it is.
Speaker: It's not easily often picked up at the bedside because the ventilator doesn't often tell you that there's in fact been breath stacking or that this has happened.
Speaker: So it's not picked up clinically, partly because just recently described.
Speaker: Patient ventilator synchrony is sort of a difficult thing to pick up at the bedside often anyway, unless you look for it very carefully, unless you've got an esophageal balloon down or
Speaker: measuring electroactivity of the diaphragm, it can be difficult to pick up.
Speaker: But most importantly, I think, and the thing that made us think about this as being the most likely cause, is that as opposed to other forms of asynchrony, you know, if patients are fighting the ventilator, we try to do a lot of things, make sure they're comfortable, make sure this, that.
Speaker: And if that doesn't work, quite often the reflex is, let's give more sedation.
Speaker: In other words, I guess we're thinking,
Speaker: give more sedation, decrease ventilatory drive of the patient, and that somehow that means that the ventilator and the patient will be better synchronized.
Speaker: It turns out, and the data on this are not strong yet because this is so new, that increased sedation actually appears to increase the probability of reverse triggering.
Speaker: Increase the probability, likelihood of reverse triggering.
Speaker: In fact,
Speaker: Reverse triggering has been described in a brain-dead patient.
Speaker: So it's like you don't, somehow there's a reflex that somehow by increased sedation leads to increased the reverse triggering.
Speaker: So if you take that all together, think about the two studies now.
Speaker: First study, the ECRASIS study.
Speaker: You've got both groups treated with heavy sedation.
Speaker: The control group, because of that heavy sedation, let's say gets a lot of reverse triggering.
Speaker: A lot of, therefore, get a lot of over-distention, ventilating-induced lung injury, increased mortality.
Speaker: The treatment group in that study gets neuromuscular blocking agents.
Speaker: So even though the patient is trying to take a breath because they're paralyzed, they don't take another breath.
Speaker: So the paralytic agents decreases ventilating-induced lung, decreases over-distention, decreases ventilating-induced lung injury, decreases biotrauma, decreases death.
Speaker: That's the ACRIS study.
Speaker: The Rho study, however, they used heavy sedation in the treated group, but the control group had low level of sedation.
Speaker: So if you think about this, if this mechanism, if this hypothesis is correct, that group wouldn't have nearly as much reverse triggering.
Speaker: So that patient, that group would not have much of an increase in mortality.
Speaker: And the group that had the potential for, had the higher sedation were given neuromuscular blocking agents.
Speaker: So you wouldn't see a difference.
Speaker: So you now look at this, these two groups, no difference in mortality.
Speaker: So this could theoretically explain why you have a difference in mortality, why you have a difference in mortality in the first study, but not in the second study.
Speaker: Does that make sense?
Speaker: It does, and I think it's something just like you said, I mean, it's a hypothesis, but I think it's always good to try to understand it and think about this because it also gives us, I think, insight into the pathophysiology of our patients that might be very useful in understanding what we're doing with individual patients at the bedside a lot better.
Speaker: What about the comment or the difference in terms of the use of prone positioning?
Speaker: I think that this is something that is a recurrent theme in ARDS.
Speaker: For example, when people explore studies that look at ECMO, a lot of people would argue, well, if you're not using the things that have already been proven to make a difference routinely, how do we know that's the right patient selection?
Speaker: I think that here, like you mentioned, 40 versus 12% seems like a considerable difference in terms of the utilization of prone positioning.
Speaker: How do you think that impacts the results, and what are your comments on this?
Speaker: Okay.
Speaker: So I guess there's two ways, two comments I want to make.
Speaker: One is I think that North Americans especially do not use the prone position sufficiently.
Speaker: And I feel very strongly about that.
Speaker: I'm not sure why that is.
Speaker: Maybe because there hasn't been an NIH sponsored study that showed it, but the data are pretty strong that the prone position decreases mortality.
Speaker: And I hear quite often that people say, well, it's too hard to do, it's too complicated.
Speaker: You know, I don't buy that.
Speaker: I mean, yes, the first time it's a little complicated.
Speaker: You certainly need logistics.
Speaker: You have to get the nurses, the RTs working together, but it's not so complicated.
Speaker: And if you look at the decrease in mortality, it's spectacular in that study that was published in the New England Journal.
Speaker: So it should be used more.
Speaker: And I'm not sure why that.
Speaker: It's funny how, you know, we talk about evidence-based medicine, but there's many examples of where the evidence is there and yet people don't use it or vice versa.
Speaker: And this is one of those cases where the evidence is pretty strong
Speaker: But for some reason, North Americans don't use prone.
Speaker: So that's a bit of a separate issue, but I think that's really important.
Speaker: And to your point, I think it's almost a crime to put someone on ECMO for ARDS if you're not having tried the prone position first.
Speaker: I mean, if someone says, well, prone is too difficult to do, well, yeah, compared to ECMO, like it's not even close.
Speaker: It's not even close.
Speaker: So I think it has to be in the algorithm
Speaker: to use before one uses ECMO, no question in my mind.
Speaker: So that's sort of my hobby horse.
Speaker: But can it explain the difference in the two studies?
Speaker: Well, in the discussion of the Rho study, the authors suggested that that might be one of the potential mechanisms.
Speaker: But for that to be potentially one mechanism, that would mean that neuromuscular blocking agents
Speaker: would have to be more effective in patients who are prone than supine.
Speaker: I mean, that's what they sort of say.
Speaker: They say, well, there might be some mechanism, but I don't know.
Speaker: It just doesn't, I don't see, I can't think of a really good mechanism by which that would be.
Speaker: You might think it's in the other direction, in fact, because when you go prone, you've got more homogeneous ventilation
Speaker: As far as I know, there's no increase in patient ventilator asynchrony that could make them more susceptible to, therefore, be benefited by paralytic agents.
Speaker: So even though that's a theoretical possibility,
Speaker: I don't, I haven't seen any evidence that would make me think that's a likely possibility for the difference.
Speaker: As I said, that's one of the differences in the study, so you could say that's a possible explanation, can't argue against that.
Speaker: But I could also say, well, this, you know, neuromuscular blocking agents only work in Europeans in Europe as opposed to Americans in North America.
Speaker: I mean, the data fit that perfectly, but that's not a very satisfying hypothesis to me from a physiologic perspective.
Speaker: So I don't think that it's the difference in prone position, although you'd have to say that is a possibility.
Speaker: I don't think that explains the difference between the two studies.
Speaker: But I do think that, like you mentioned, a clear take-home message unrelated to neuromuscular blockers
Speaker: is that we probably should be using prone positioning more and patients have severe RDS because there is data to support it.
Speaker: 100%, 100%.
Speaker: By the way, there are other potential mechanisms that could explain this besides the reverse triggering.
Speaker: So I'm not sort of saying, oh, our hypothesis is correct.
Speaker: I don't know if it is or not.
Speaker: One potential mechanism I think is plausible is
Speaker: relates to the difference in the PEEP used in the two groups.
Speaker: You know, there was an interesting study by Moray in the Blue Journal within the last couple of years where they used rabbits, pigs, and they had some human data.
Speaker: And they hypothesized that in patients with ARDS, you have adalectatic lung regions in the dependent zones, and that that impairs the transmission of the negative pleural pressure swings.
Speaker: So that could lead to greater dependent lung stretch.
Speaker: So down in the bottom, because of this impairment, you have more lung stretch and more ventilator-induced lung injury.
Speaker: So if you use PEEP, you would open up some lung units, and you'd mitigate this injury by decreasing the atelectasis and also by decreasing respiratory injury.
Speaker: by receiving respiratory drive.
Speaker: And they actually showed this in animal studies that when they use low versus high PEEP, the high PEEP group had less lung injury in the dependent zones, which I thought was pretty interesting.
Speaker: Now, they used PEEP levels in the animal studies between five for low and 15 for high, much bigger difference than in the ACRIS study and the ROSE study.
Speaker: In fact, the difference between peep levels and the acarist and rose were about three centimeters of water.
Speaker: Not a huge number, quite frankly.
Speaker: But it's possible that's a mechanism.
Speaker: Having said that, you're familiar, I'm sure, with the ART trial that was published in JAMA two years ago.
Speaker: And I don't want to take credit for that trial, even though it's called the ART trial, especially since it was a negative trial.
Speaker: It was a positive trial.
Speaker: I take credit for it.
Speaker: They named it after me.
Speaker: But it w as a negative trial.
Speaker: And it was, you know, it was one of those recruitment and high peep and then the mortality went in the opposite direction.
Speaker: And I think that actually, interestingly, it's possible that double triggering, the synchrony could explain partially those results.
Speaker: If you look at the ER trial, if you look at the supplements, they have some, they have at least one figure showing double triggering results.
Speaker: that occurred not necessarily reverse triggering, but double triggering.
Speaker: So that can occur.
Speaker: In any event, I think that it's a plausible physiological explanation that higher PEEP might decrease ventilator-neus lung injury.
Speaker: And again, if you decrease ventilator-neus lung injury,
Speaker: in the control of neuromuscular blocking group, maybe that there's nothing to add from neuromuscular blocking agents.
Speaker: So that could explain why the ROSE trial was negative.
Speaker: So I think higher PEEP, I think is a very, to me, is physiologically more plausible than saying the prone position is different between the two, and there might be some interaction in prone and neuromuscular blocking agents.
Speaker: So clearly, I mean, there's a lot of potential explanations, but I think that the permanent role of lung injury induced by the ventilator is always present, right?
Speaker: It's something that I think we finally have understood.
Speaker: It's the only thing that really has moved ARDS care forward significantly and consistently, and there's still a lot of things that we need to learn.
Speaker: But based on these two studies and where we are today, Art, what would you think is the current status of early neuromuscular blockers in ARDS in terms of
Speaker: general recommendations initially, and then maybe how you would apply it to your own clinical practice?
Speaker: So I think I'll give you how I think it should be applied.
Speaker: And so I think that's what's changed for me in thinking about it.
Speaker: You know, prior to the Rose trial being published, I would say that all patients with PF less than, let's say, 120, you should try in the first couple of days, you should use neuromuscular blocking agents.
Speaker: I would have thought that that's, you know, there's sufficient evidence
Speaker: to show that.
Speaker: Now, I don't think that's the case.
Speaker: I don't think, when I say it's not the case, it shouldn't be used routinely in every patient just because their PF is low.
Speaker: I think that it should be used, should be considered for patients where there's a physiologic rationale.
Speaker: So tailor it to the patient.
Speaker: So for example, if you see the patient has, say, dyssynchrony and then it's double triggering.
Speaker: Well, to me, and you've tried other things to change that and it hasn't worked,
Speaker: that to me would be a no-brainer to try there.
Speaker: If the patient has, you know, along with Laurent Burchard and Antonio Pacente, we wrote an article a couple of years ago in the Blue Journal on self-inflicted lung injury, that if patients have high respiratory drive, even breathing spontaneously, they could actually cause lung injury to themselves, you know, sort of ventilator-associated lung injury.
Speaker: It's not ventilation-associated lung injury rather than ventilator-associated lung injury.
Speaker: So if patients have high respiratory drive, which they often have in the first couple days of diagnosis, and they're either got a large P.1 pressure in 100 milliseconds, or looking at accessory muscles, looks like they're very high drive, I think in those patients, it makes sense to use more neuromuscular blocking agents.
Speaker: So I think it's more thinking about the specific patient,
Speaker: thinking about what physiology might be corrected by using neuromuscular blocking agents.
Speaker: As I said, that's different than it was in my mind a couple of years ago.
Speaker: And by the way, the ROSE trial, the other advantage of the very useful bit of the ROSE trial that I think is important is that they show that the long-term consequences in terms of muscle weakness
Speaker: They couldn't pick it up.
Speaker: That doesn't mean it didn't happen, but it wasn't like it's a major, major issue.
Speaker: So that's how I would see using it.
Speaker: I mean, if you look at patients and they had a high dyssynchrony index, there's something called the, I forget what the exact term is, but basically there's been a couple of studies showing that if patients have greater than 10% of breath that are dyssynchronous with the ventilator, they have a higher mortality for whatever reason.
Speaker: Studies haven't sorted out what the mechanism is.
Speaker: So those kind of patients, I think it makes a lot of sense to use neuromuscular blocking agents.
Speaker: By the way, one thing we hadn't talked about for a difference between the two studies, this could be due just to chance alone.
Speaker: One study was less than 400 patients.
Speaker: The other one was 1,000 patients.
Speaker: So the ROSE trial was 1,000 patients.
Speaker: Sometimes if you study enough patients,
Speaker: You get closer to the truth, if you like, and a smaller number of patients can be very, very misleading.
Speaker: I don't know if you remember, there was a trial maybe 10 or 15 years ago for TIFP, for sepsis, and it was published in JAMA.
Speaker: It's actually a very instructive study to pull out because it's something actually you and I talked about just a few minutes ago, Sergio, and that is sample size, et cetera, et cetera.
Speaker: So in this TIFPE trial, they used TIFPE for sepsis, and they had a data safety monitoring board that looked at the trial halfway through.
Speaker: Halfway through the trial, roughly, the mortality rate in the TIFPE group was about 8% less than
Speaker: than the control group.
Speaker: I'm talking about absolute decrease in mortality in septis.
Speaker: I mean, that's a huge difference.
Speaker: Like I'm making up the numbers roughly, but let's say 35 to 27%, something like that.
Speaker: And the P value for that was 0.006.
Speaker: Wasn't enough for them to stop the trial.
Speaker: They then continued the trial and at that they had enrolled about 800 patients.
Speaker: So it wasn't small.
Speaker: They then enrolled another 800 patients, and the trial was completely negative.
Speaker: In fact, mortality went in the opposite direction.
Speaker: So just think, in a sense, how scary that is.
Speaker: If that trial had been powered to 800 patients, we'd all be using TIFP for sepsis in all our patients.
Speaker: So it's possible that, and again, when you think about the fact that the Kaplan-Meier curves sort of separate only at 18 days,
Speaker: Was that just the fluke?
Speaker: Was that study, you know, did it occur by chance alone?
Speaker: So I don't know the answer, but that's a possibility.
Speaker: And I think that this leads me into a question I wanted to ask you.
Speaker: And Danny Kahneman, who is a brilliant psychologist, who really was one of the followers of cognitive psychology, wrote a paper many years ago on the law of small numbers, arguing exactly that same thing, but in psychology, that small numbers tend to overestimate, especially when there are positive results,
Speaker: and that it can lead us down the wrong road.
Speaker: And we see this over and over again in critical care when small studies are very positive and everybody jumps on board.
Speaker: I think we most recently have seen this with a vitamin C and other cocktails for
Speaker: for septic shock.
Speaker: But really we need larger studies to get a better idea.
Speaker: But the question I guess leads to, I mean, you've done so many trials.
Speaker: Do you think that the randomized clinical trial is the right tool or are we gonna just keep seeing negative study after negative study?
Speaker: Well, I know there's a lot of people who are skeptical about randomized control trials.
Speaker: I happen to be a fan up to a point of randomized control trials.
Speaker: And by that, I mean, I think they provide the best level of evidence.
Speaker: And it's so hard to have to provide this level of evidence without doing a randomized control.
Speaker: And the reason is because you get rid of biases by randomizing.
Speaker: You know, if you don't randomize, no matter how many patients you include,
Speaker: you could have still the same bias in all cases.
Speaker: Now, that's not to say everything needs to be randomized controlled trial.
Speaker: But I think for things like this, when it's drugs like this, whether it's steroids or I think that this is the only way to get the correct answer.
Speaker: Now, for example, where it's not necessary potentially is if you think about the study, the observation 35 years ago about auto PEEP in patients
Speaker: you know, with COPD.
Speaker: There's a study by Pepe Marini where they noticed that patients, I believe the two patients have COPD, that if they stopped the ventilator, there was decreased pressures in the chest.
Speaker: The patients who were hypotensive now were no longer hypotensive.
Speaker: Like that's pretty dramatic.
Speaker: So very dramatic.
Speaker: So I think that sometimes the physiology is very blatant, but still,
Speaker: I think that the randomized control trial provides the strongest level of evidence and changes practice.
Speaker: We only have to look at the first ARDSnet trial, how that has changed practice, has made us all believers that ventilator-induced lung injury is important.
Speaker: You know, I chaired a consensus conference in 1993 or so, and we published a paper in CHEST sort of summarizing the conference.
Speaker: And one of the suggestions we had at the time was we said,
Speaker: try to limit the plateau pressure to 35 centimeters of water because you might have lung injury if it goes higher than that.
Speaker: And you can't believe how much grief I got after that.
Speaker: People were not emailing me back then, mailing me, phoning me.
Speaker: How could you suggest that?
Speaker: That's crazy.
Speaker: And now, you know, if you have a randomized control trial, you now have, I think, very good evidence.
Speaker: Now, there are other approaches of using large databases, using artificial intelligence to sort of
Speaker: analyze the data but you're still left with you know if the patient if you're analyzing patients who didn't did not receive a certain therapy you're going to use that in the back of your mind is why did that patient receive the therapy when the comparable patient didn't receive it and could that explain the difference in outcomes so i'm actually a fan to a certain point but i think you have to ask well-asked questions like you have to have a very good hypothesis
Speaker: It has to be done well.
Speaker: And ideally, you have big enough networks that can carry them out quickly, because that's another complaint about randomness control.
Speaker: They can take years, you know, and they only then, you know, include a small subset of potential patients, et cetera.
Speaker: The generalizability may not be there, but I think when done well, and if you're lucky with the results, it can have a huge impact.
Speaker: Excellent.
Speaker: So I think this is a great place to stop with a discussion on ARDS.
Speaker: And one of the things we like to do, Art, and the podcast is close with some questions that try to tap into the wisdom of our guest, not related to the topic.
Speaker: Would that be okay?
Speaker: Sure.
Speaker: Again, I'm not sure about wisdom, but you can tap into my brain anyway.
Speaker: So the first question relates to books.
Speaker: And I wanted to know if there's any book or books that have influenced you significantly or book that you have gifted very often to others.
Speaker: So I would say there isn't sort of one book that stands out.
Speaker: I'll just tell you about maybe one or two that I just read recently.
Speaker: And actually, you triggered one of them with your point about Danny Kahneman.
Speaker: You know, there's a book or a couple of books by a guy named Richard Thaler, T-H-A-L-E-R.
Speaker: He won the Nobel Prize for economics last year, two years, three years ago.
Speaker: One of the books is called Misbehaving.
Speaker: And what he talks about is the development of something called behavioral economics.
Speaker: So it sort of combines, if you like, psychology and economics.
Speaker: You know, traditional economics is based on the concept that we all act rationally, we always do the right thing.
Speaker: That's rational.
Speaker: But as we know, humans aren't rational when it comes to decision making.
Speaker: So the whole field of behavioral economics, I think it's had an impact in a lot of ways, whether it be
Speaker: transplantation and pensions.
Speaker: What do I mean by that?
Speaker: If you ask people to opt in to be an organ donor, you get much different results than if people have to opt out to not be an organ donor.
Speaker: And some countries like Spain have, I think it's Spain, have an opt-out.
Speaker: So if you don't opt out and you die, they can take your organs for transplantation.
Speaker: And the rate of transplantation in these places is much higher.
Speaker: And that's just behavioral economics in a sense.
Speaker: Shouldn't matter whether it's opting in or opting out.
Speaker: It's the same decision, but somehow there's this inertia.
Speaker: And this book gets at that, and I think it's very good.
Speaker: And in fact, in the UK, they started something called a nudge unit that, you know, this guy Taylor, obviously, not obviously, also wrote a book called Nudge.
Speaker: So I thought that was an interesting book.
Speaker: Before you go to your second book, yeah, I just wanted to make a point.
Speaker: which I think is very important on Thaler's whole concept, is that it applies tremendously to what we do in medicine, right?
Speaker: So for example, this whole concept of decision architecture.
Speaker: If you write a protocol and if you ask, if you have to,
Speaker: check something, it's less likely to happen than if you have to uncheck it if you don't want it to happen, right?
Speaker: So if you want something to happen, you pre-check it in your protocol.
Speaker: And that's how you drive compliance.
Speaker: And I think that it also speaks to confirmation bias, right?
Speaker: How can you show the same study to a group of intensivists and half will say it's a positive study and the other half will say it's a negative study?
Speaker: It's the same results, right?
Speaker: It just confirmed whatever they believe before the trial.
Speaker: Exactly.
Speaker: That's where the Bayesian analysis comes out.
Speaker: But that's exactly right.
Speaker: Sorry to interrupt you.
Speaker: What was the second book?
Speaker: The other book I thought I just finished a little while ago was I found a really interesting book called Bad Blood.
Speaker: It's got a longer title, but it's Bad Blood, and it's by someone by the name of John Kerry, who is a journalist for the Wall Street Journal.
Speaker: And it's about a biotech startup called Theranos that was founded by a woman by the name of Elizabeth Holmes, who quit Stanford, I think she was about 20 years.
Speaker: And they supposedly had a technology that allowed one to do about 200 blood tests from a drop of blood, sort of within an hour, like almost instantly.
Speaker: And this was a startup company.
Speaker: The company reached a valuation of about $9 billion.
Speaker: Holmes was worth on paper $4.5 billion roughly.
Speaker: This was in five or seven years.
Speaker: And it turns out the company didn't have anything special technology at all.
Speaker: Essentially, the whole thing was a scam that sort of engulfed Silicon Valley, lots and lots of smart people.
Speaker: Robert Murdoch, Rupert Murdoch lost $125 million.
Speaker: The board of this company had luminaries like Henry Kissinger, George Shultz, who's the Secretary of State, General James Mattis, a couple of senators, including Bill Frist, who was also a heart surgeon.
Speaker: And so just, again, a story about that impacts medicine, if you like, but
Speaker: She's now being sued and I guess goes to court in 2020, they said.
Speaker: It's just fascinating how so many people could be so fooled for so long.
Speaker: This woman who was worth, on paper, $4.5 billion three years ago is now worth zero.
Speaker: It's a fascinating story.
Speaker: I did not read the book.
Speaker: I did see there's a documentary with the same name based on the book.
Speaker: That's right.
Speaker: It's coming out.
Speaker: It's very interesting, the whole concept of
Speaker: fake it till you make it or I guess till you go to jail, right?
Speaker: That's right.
Speaker: Exactly.
Speaker: But I think it again speaks a lot to how we behave in irrational ways and how we think we make rational decisions, but yet we can be deceived.
Speaker: both when we read papers or when people are making these investments.
Speaker: And probably there's a lot of that herd mentality.
Speaker: Well, if somebody else that's very smart is investing, why would I not invest?
Speaker: Right.
Speaker: So if somebody else believes in something we see in medicine, I usually I'm more likely to believe it as well.
Speaker: So I think these are excellent books that we'll definitely put at the show notes in case our readers are interested in reading.
Speaker: The second question relates to something that you believe to be true that most other people don't believe to be true.
Speaker: I must tell you, I don't have such a truth, so I have to disappoint you there.
Speaker: One of my favorite mantras is make sure you love what you're doing.
Speaker: In medicine, we go into medicine because hopefully because we love it.
Speaker: I tell the residents,
Speaker: They ask, you know, I want to I want to be, you know, get into academic medicine.
Speaker: What do I do?
Speaker: I say, well, you know, don't look at it that way.
Speaker: Do what you love.
Speaker: I think that's that's really important.
Speaker: And part of that doing what you love from a leadership perspective, then, is.
Speaker: creating a culture in an organization that helps people to want to work together, that sort of decreases the negative parts of competition.
Speaker: Some competition is very good.
Speaker: And I must tell you, I led the critical care division here for a number of years, and I also was vice president of research at the hospital.
Speaker: When I recruited people, I have sort of a saying that everybody who knows me here knows, and that is,
Speaker: Obviously, I'm looking for the smartest people.
Speaker: So I always go and recruit for brains, but I hire for personality.
Speaker: So I want to make sure that hire people who want to work together.
Speaker: Because you think about it, when you go to work, you spend more waking hours with the people you work with than probably your family members.
Speaker: So you better want to work with them.
Speaker: They better want to be the kind of people who...
Speaker: who help each other, who are collaborative, leaving into disciplinary teams and the like.
Speaker: So that's, you know, but I can't certainly say that that's not what other people believe.
Speaker: A lot of people believe that not necessarily, they don't necessarily practice it, but I think, you know, building the culture of an organization, making people want to happy and then making sure you hire the right people, I think is really important.
Speaker: Well, and I think that it's a very timely comment, especially, and I don't know, I mean,
Speaker: what the microcosmos in Canada right now is for medicine, but clearly in the United States, a lot of burnout, it's a rampant problem, not only with young physicians, but also older physicians.
Speaker: And I do believe that it's not universal.
Speaker: There's plenty of people who really enjoy what they're doing, but maybe it goes art to really what people don't believe in is that they don't believe in being honest with themselves and really identifying what are the things that really cause them pleasure, that really cause them pride.
Speaker: And focus on that.
Speaker: There's always things that you can complain about.
Speaker: But if you are focusing on the things that you really enjoy and that really make a difference for others and for yourself, I think you're much more likely to love what you do.
Speaker: 100%.
Speaker: 100%.
Speaker: So the last question and the closing question would be, what would you want every listener to know could be a fact or a quote after listening to this podcast?
Speaker: So I would say not related to the content, the specific content we talked about, but I think the point that you just sort of brought up and that is, you know, intensivists work hard there and the nurses and the RTs work really hard.
Speaker: It's high stress, but I think it's really important.
Speaker: We want to take care of patients, but it's also important to take care of yourself.
Speaker: And so burnout, as you pointed out, I think is a big issue.
Speaker: And I think that physicians,
Speaker: and all the staff in the ICU and in the hospital should be thinking about, thinking a little bit about themselves as well, you know, making sure they don't burn out because it's obviously much better for them, but quite frankly, it's going to be much better for the patients as well.
Speaker: So I think that enjoy what you're doing, get help if you need it, you know, think about burnout, think about making sure that you're healthy physically and psychologically so you can then help your patients as well.
Speaker: Excellent.
Speaker: I think this is a great place to stop.
Speaker: Art, I really want to thank you for your time and be so generous with your expertise.
Speaker: It was a real pleasure to talk with you.
Speaker: And hopefully we'll have you again on the podcast to talk about other topics related to critical care.
Speaker: My pleasure.
Speaker: Take care, Sergio.
Speaker: Thanks again for listening to Critical Matters.
Speaker: Make sure to subscribe to this podcast on iTunes or Google Play.
Speaker: You can also listen at www.soundphysicians.com backslash podcast.
