65. Journal Club: The RSI Trial

If I'm a patient and you induce me and you have to put me on some pressors, I don't care about that. Whereas if you induce me and I die, I care about that or I should care. I would care about that hypothetically, right? And so I feel like the equivalence between these things is you got to be careful. Welcome to Critical Care Time, the podcast for everyone who cares for the critically tale. I'm your host, Dr. Nick Mark, joined as always by my friend and co-host, Dr. Cyrus Askin. Oh, thank you for that kind introduction, Nick. I appreciate that. So this is kind of a special episode. We weren't going to record this, but we just couldn't resist. This is like a breaking news story. The RSI trial was just presented at Critical Care Reviews down under with simultaneous publication in the New England Journal. Yeah, it's a big, big week for Critical Care, big week for induction medications. This trial has shaken things up. I think, to an extent, it's got a lot of people talking, a lot of people excited. And so we figured we might as well jump on the bandwagon. Right. Or a bunch of wrecked bank drongoos chucking a wobbly over it as they say down under. I'm trying to learn, you know, as I watch CCR down under God. Maybe we should just have a moratorium on the Australian accent. I think we're already accruing haters by the second here. All right, mate. Oh, no. Okay. Yeah. So I'm trying to get beyond that. Guys, and if you're still listening, and I understand if you're not, the viewership is dropping off in Australia, as we speak, precipitously. Yes. We realized we did just do a journal club episode. But this is kind of a little bonus. And we thought that this content, this particular publication is one that really should be highlighted. So here we are. Okay. So this trial, the RSI trial was done by the pragmatic critical care research group. In my opinion, these guys are like the goat of US critical care trials. They did this trial at 14 sites, six EDs, eight ICUs across six medical centers in the US. They excluded patients who were pregnant prisoners, trauma patients, or had an immediate need for intubation that precluded randomization. They also let the clinician exclude people if they determined that one drug or the other was necessary. They enrolled 2,365 critically ill adults. They randomized them to ketamine versus automataid. The patients were about 60% male. And the median weight was 78 kilograms with a median BMI of 27. More on why I'm sharing that in a second. About 45 to 50% of them had hypoxemic respiratory failure, 15% hypercarbic respiratory failure. The patients were intubated in the ED and the ICU. About 60% were done by ED physicians and about 40% by ICU physicians or anesthesiologists. Roughly 55% were done by residents and 35% by fellows. So only about 4 to 5% of the intubations in the trial were done by attendings. That's interesting. That's an interesting kind of point here is that the majority of folks that were the airway operators were in some sort of trading status. Which raises important questions about like who was making choices about this, like who was deciding on what other meds to give and how to plan the intubation. So put a pin in that as well along with the weight. Yeah, sure. Kind of moving forward with what they did next. So about 69% of patients receive rockuronium, 31% received succinyl colon. About 45% were preoxygenated using a non-rebreather mask. Only about 25% were preoxygenated using non-invasive ventilation. So recall, you know, we've had some podcasts about this in the past breaking news over the preoxytrial. Yeah, which is the same group, right? I mean, it's kind of funny. Exactly. So it's interesting. You kind of see that the best practice that was proven in one study didn't really make it into the next study. Yes. It's also kind of interesting that in the year of our Lord 2025 that there's that many people that are getting succinyl colon, like one in three. Yeah, that's kind of wild. This is like, you know, that meme from a rest of development. Like you still doing that. That's how I feel about that 31% still using succs. It's interesting to me. And I know some people have their reasons for using it. I don't quite buy into the logic there. I don't think you'd be unique. No, I don't. I'll just just send them to our prior episodes about this. I think our airway controversies episode, our conclusion was, you know, that Indiana Jones. It belongs in a museum. Right. Yeah. Yeah. Oh, well. Yeah. And then it was going to look not everyone's going to see everything, you know, from our perspective. But that's why we're here to share our our wisdom, whatever. That's right. Hopefully change a few minds. Maybe. Anyway, so that this is the patient's day enrolled. Now let's get on to the kind of meat of it. So patients were randomized to ketamine or automodate. The median dose of ketamine was 140 140 milligrams or 1.6 mix per gig. The median dose of automodate was 20 to zero milligrams or 0.28 mix per gig. And I just note that both of these were dosed according to actual body weight, not ideal body weight or predicted body weight. Furthermore, that was the median dose. A quarter of patients in the ketamine group got more than 2 milligrams per gig. And if you look at figure S2 and the supplement, you can see that there were some patients who got some pretty, pretty high doses. Yeah. But yeah, higher than I think, you know, you and I typically use. I'm sorry. Is that for mix per gig of ketamine? Yeah. Yeah. Yeah. It looks like there's on the histogram, there's like one dot at the end where somebody got like a monster. So go. Yeah. That is a deal if I ever saw one. So yeah. So I mean, okay. So maybe we'll put a pin in that too. We'll talk more about dosing, I think. Let's move on to the intubation itself. So video langoscopy was used in 95% of cases, clap, clap, clap, good, good practice. We like that. There was a reasonable split between bougie and style at use. I think that's kind of a style choice. First pass success was 85%. It is a little on the low side, but I think it is important again to recall that these are mostly trainees that are intubated. Exactly. So I think it's lower than I think you and I typically see, but it's not wildly lower. Right. Now, here's where it gets interesting. So if we go back to this supplement, time from induction to laranoscopy was 70 versus 62 seconds with ketamine and automatae respectively. Time from induction to intubation was 112 seconds with ketamine and 103 with automatae. This was a statistically significant difference, meaning it took about eight to nine seconds longer to intubate using ketamine for some reason. Okay. But I'm sure, as you always do, Nick, there's some wild theory that you have behind why that's the case. Exactly. My wild conspiracy theory is that they probably use the same size syringe that we do in my ICU. We typically draw opamids in its 10 cc syringe. And if they're using the standard 10-meg per mil concentration of ketamine, that means they would have to use two syringes if they wanted to give more than 100 milligrams of the drug, which they did in most cases. So presumably, it took longer because they pushed one syringe and then they took it off and they put another syringe on and they pushed that syringe. So that probably, you know, probably the timing reflects the high dosing. I think that's a reasonable theory, actually. Not as crazy as some of your other theories. No, don't give me started on the grass, you know. Okay. You want to take a turn and tell us about the outcomes? Yes. We'll pivot away from the specific grassy null. We'll talk about the outcomes. Okay. So the primary endpoint here was hospital death by day 28. So a reasonable primary endpoint, I think, to select. Very patient centered. The secondary endpoint, and this is where it's a little interesting, was quote unquote cardiovascular collapse during intubation. And so I don't know much unique, but when I think of cardiovascular collapse, I think of like a serious decompensation. Who's doing CPR? Yeah, or like, you know, peri arrest situation that's like really, really kind of sphincter tightening, right? So in this paper, they define a cardiovascular collapse during intubation as the occurrence of any of the following events during the interval between induction of anesthesia and two minutes after tracheal intubation. The events were a systolic blood pressure below 65 millimeters of mercury, receipt of a new or increased dose of vasopressor. So like a tiny little ratcheting up of the of the more epinephrine would killed. And then of course, cardiac arrest. So this is the classic composite endpoint where we combine together three completely dissimilar things that I may care about or may not care about. So I care about hypotension, that's bad. I don't really care about vasopressor dose, but I definitely do care about cardiac arrest. So whenever I see a composite endpoint like this, I'm like, okay, what is driving it? Those are not the same. Yeah, absolutely. With that, having laid out the red carpet, drum roll, please, and the results were. So the results of the study, RSI, in hospital death by day 28 occurred in 28.1% of those in the ketamine group and 29.1% in the atomic rate group. So that was a 1% difference, which was not statistically significant. Now, a little brief side known about the power of the study. So they expected a 5.2 percentage point difference between the groups and incidents of in hospital death by day 28. So that is a wildly high estimate for effect size. Like if you think that your choice of induction med is changing outcomes by 5%. I'm not sure it's ethical to do the study if that's your, if that's your estimate. So, you know, on the other hand, I could also understand that if they had predicted a 1% difference, which is what they got, there was numerically 1% lower 28 mortality with ketamine, you would have to do like a 10,000 person study to see that. So I understand why they chose what was probably a statistically achievable power. But I also think this study was kind of designed to fail because that's an overly large estimate of what the effect size would be. Right. That's yeah. Yes. I think that's a fair point. Now, getting back to the composite endpoint. So the composite of cardiovascular collapse during intubation occurred in 22.1% in the ketamine group and in about 17% in the automating group. So that 5.1 percentage point difference was significant. All right. Well, we're done. No more ketamine ever. It's done. No, no more ketamine. Throw it away, put it in the, put it in the vault along with the succinyl colline. All right. Thanks for listening folks. See you next time. But wait a second. What was the breakdown here? Yes. That's the key question. Whenever you see a composite endpoint, what's the breakdown? So the first place I go to when I'm reading a paper like this is the supplement. That's where people hide the truth. So if we go to table S16 on page 72 of the supplement, you have to dig for it, they find that a systolic blood pressure less than 65 occurred in 6.4% versus 5.5% that was not significant. New or increased vasopressures occurred in 21.3% versus 15.9% that was significant. And cardiac arrest occurred in 1.0 versus 0.8% that was not significant. So just to summarize that the cardiovascular collapse in the ketamine group, the 5% that was basically driven almost entirely by more vasopressures in the ketamine group. But if we look at the same table, it gets a little bit more interesting. So numerically, more patients in the atomic group were started on vasopressures prior to intubation, 23 versus 20.9%. And more received prophylactic vasopressures with induction, 19.7 versus 17.6%. So you're saying that they were more likely to start vasopressures preemptively in the atomic group and more likely to start them reactively in the ketamine group. That's exactly right. Which would, by definition, impact that secondary outcome. Right. So theoretically, right, an increase in, you know, you could increase in starting or the same thing. But I have a feeling that if you already started them, you may not need to adjust it very much. You may have a little bit of cushion there. Whereas if you haven't started them, now starting them is going to be your secondary endpoint has been hit. We hope you guys are enjoying the episode so far. I know we are. Briefly, we'd like to take a moment to thank our returning sponsor C-Star Medical. We are fortunate and grateful that this episode of Critical Care Time is sponsored by C-Star Medical. Returning once again to support us in our quest to bring high quality critical care education straight to your eyes and ears. C-Star Medical is advancing the science of cell-directed extracurricular therapy to help restore the balance of a dysregulated immune system in acute kidney injury and sepsis. Check out their website, www.cstarmedical.com to learn more. That is S-E-A-S-T-A-R medical.com. They were recently awarded the Corporate Innovator Award by the National Kidney Foundation for their Polymium Therapy. We wish them all the best as they continue working to bring this therapy to more and more patients across the globe. Congrats to C-Star. Thanks for the support and with all that, let's get back to the show. So then, why don't we take the rest of this episode to talk about some takeaways and whether this really is going to impact our practice? So, I think that some people are interestingly taking the position of like, well, parties over for ketamine, no need to use it anymore. I think for the last decade or so, ketamine has been kind of this like wonder drug in the ICU, along with like TXA, right? It's the shiny new thing. It's got a great reputation for safety. And I think it, on one hand, it's almost inevitable that studies would come along that would deflate that sort of like that bubble around the drug, right? I mean, the shinyness is going to have to, you know, fade eventually. But I do think that there is a real risk that if we draw vast conclusions from half vast data, we might throw out the baby with the bath water. I think probably the first really big issue with the study that I have is a reminder of what parasolsists said 500 years ago. The dose makes the poison. Yeah. So presumably, four migs per kilogram is maybe a high dose. Well, maybe pushing us in the poison direction. Touch, yeah. Touch, touch. But is, you know, 1.6 migs per kilogram too high of a dose? Right. So the median dose in this study objectively is not higher than what's been used in many prior trials. No, it's a reasonable dose. But I do think it's higher than what a lot of people myself included actually use in the ICU. Yeah. I tend to agree. And then I'd also say, you know, should we be using ideal body weight in these cases? Right. So in the prior studies, they used actual body weight, but I'm not sure that's right. Yeah. So, you know, Nick, you'd mentioned this before, but for our listeners, remember that the median BMI in this study was 27 with an intercortial range of 23 to 32. So that means that more than half of the patients were overweight in a quarter were obese. And this can have big implications for drug dosing. Right. Like we said, the median dose in the trial was 140 milligrams. I rarely give more than 100 milligrams. And as we pointed out, if you look at the histogram in s9, there were, there were a lot of patients who got way higher doses than that. I think that could be driving a lot of the adverse hemodynamic events. Yeah. So really there are two issues here. Right. So should we be using IBW or actual body weight? And then what should the dose be? Yeah. So I like theoretically, I can make an argument for why it should be ideal body weight. You know, I think people would say that ketamine and atomic are both lipid soluble drugs. And so, hey, it's soluble. It goes to your whole body. So if you are bigger, you need a higher dose. But I think you have to remember a couple of things. The place where the drug is working is your brain. A 150 kilogram person's brain is the same size as a 75 kilogram person's brain. So you don't necessarily need more there. Lipophilic drugs will eventually distribute to the whole body. But the bolus pharmacokinetics are much more determined by that central volume. Right. If you're twice as big, you don't have twice as much blood volume. Right. So you may not need twice as much drug. And then also remember that the hemodynamic effects are largely going to increase with the dose. This is like prices right rules. Right. So you want the closest without going over. You know, that person who got 3.5 bigs per gig or four makes per gig, you probably went over there. And then last point, when somebody's in shock and they're hypoperfused, that central compartment shrinks. So you may actually need a lower dose in patients who are already in shock. Personally, I think 1 to 1.5 milligrams is enough. And I rarely exceed 100, which just so happens to be the amount that comes in the typical size dosage bottle that we have. Yeah. And I think just reacting to that, I'm kind of similarly like in some of the populations I've worked with like young like more kind of robust folks who you know are healthy, a lot of muscle mass like I might be a little more generous with the amount of ketamine that I'll give them and get closer to that 1.6. But I'm rarely unless it's like battlefield medicine where you're really trying to treat a horrific wound or like a blast injury, you know, I'm not really pushing into the types of doses that they're using here. And it's possible, well, I guess they excluded trauma patients, right? Well, I was going to say, yeah. So they didn't treat any blast injuries here. Yeah. So I don't, yeah, I'm really like struggling to understand that like four big brinking dose, like what are they doing? I mean, I mean, in fairness, that's like one dude in a county 300 person study. But I think the fact that the dosing was so wide, tells you that there's something going on here. There's some variability. But yeah, no, I, I tend to agree like I think if I'm using ketamine alone, usually about one big per gig is good, maybe maybe like 1.2 migs per gig. And then if they're, you know, if they need more, I can give more, man, that tends to work pretty well. But so with all that being said, another question would be with respect to the secondary endpoint and what, like let's unpack that a little more. Yeah, so I understand why they chose a composite secondary endpoint. He might end up collapse, make sense as a secondary endpoint. But I think it was the wrong choice. I think starting vasopressors before intubation is a best practice. And I disagree with received vasopressors as a complication, right? I also, I also wonder like if if this were, if they didn't call it a composite endpoint, but they just had like several secondary outcomes listed discreetly, would we be talking about this paper? I mean, I guess we might be just because of the significance and the size of the population, but really like, then your only significant reporting point is this kind of very odd factoid, I guess. Yeah, I didn't, I didn't play around with the stats on this paper, you know, super in depth, but I kind of wonder like in if we, if we read it this study as like a win ratio study, which we've talked about before, as like we love that design. And mortality was number one, and cardiac arrest was number two, and like, you know, human dynamic collapse, like however you define it was number three, I wonder what the outcome of that study would be if you actually combine them together in a win ratio study where you rank them. Because I think it's very clear to me and probably to our listeners that like, if I'm a patient and you induce me and you have to put me on some pressers, I don't care about that. Whereas if you induce me and I die, I care about that, or I shouldn't care if I would care about that hypothetically, right? Yeah, I think that's another great point. And then another thing we probably should mention is the fact that this was an open label study, so there's the limitation of an open label study is that physicians will make choices accordingly. So if you think that it's going to drop the patient's blood pressure, then you might be more careful about starting pressures preemptively. Whereas a lot of us, I think, you know, we recognize that ketamine does pause the release of endogenous cataclysmines, and that can in and of itself act as a sort of like internal presser, quote unquote, or you get a little bit of a pressure effect in many cases. So perhaps we're less likely to start pressures, you know, prior to intubation, and that can totally change the results. So yeah, so I think if this had been a blinded trial where they were pushing a syringe that was, you know, study drug as opposed to knowing what it was, they might have behaved differently. They might have started more pressures preemptively in the ketamine group, you know, who knows. I think another thing to remember too is that actually if you look at the supplement, you can see that the blood pressures after, after induction were actually, or during laryngoscopy were actually a little higher in the mortality group, which may have just been that they were under sedated. They may have felt it more, right? So pain is a vasopressor, you know, discomfort is a vasopressor. So, but it's not a very good one. And so I wonder if that's another thing, which is that a low dose of one drug versus a high dose of another got you more analgesia in one group than the other, and that's why your blood pressures were dissimilar. That's a feature or a bug. Yeah, so I think we've, you know, shared a lot of our insights and thoughts on this, but, you know, where the rubber meets the road is when it comes to your practice and how this might change your practice. So Nick, what do you think? How is this going to change your practice? So I got to admit, my practice, you know, I was, I was very pro ketamine for years, and my practice has been evolving lately. Like we did that airway episode with guest Jared, Jared Mosher, and he put a really good way, which is that like it's probably kind of a Coke versus Pepsi question. Like there's a lot of operator preference here. I think, you know, I used to be much more in the, oh, you know, Coke is way better than Pepsi. Ketamine is way better than a tomaday camp. And now, you know, for the last years, so I've been using them both. And, you know, sometimes logistically, it's easier to get a tomaday because it's in the med bag as opposed to ketamine, which has to be pulled separately. So, you know, I don't, I don't view it as much, much better. But I'm going to keep using ketamine when I think the circumstances are right. I think they both have a role, and this study doesn't necessarily change my mind about that. Yeah, I think I feel pretty similar to that. I mean, I think, even though, even though it was not statistically significant, the 1% mortality difference, I mean, I don't know, to me, that's, that's something. Yeah. And it certainly doesn't move the needle in the opposite direction. I, it doesn't make me feel like I need to stop using ketamine because I'm likely doing harm to my patients, potentially contributing to their mortality. The only harm that it seems I'm potentially contributing to is in, if you break it down like we just did, would be an increased need for vasopressor, which is not really, I mean, that's kind of part and parcel of being in the ICU. It's a good practice to start low-dose pressures before even to bait any way, right? Even if somebody is normal-otensive, I'll often start, nor happy at a low dose of like two or something. Because once it's flowing, it's very easy to just turn it up if you need it, as opposed to like, oh, get it, hang it, spike it, program the pump, you know, waiting for it to get through. So, I mean, if we take anything away from this paper, it's that ketamine may not be as, as humaninamically neutral as we think, and we should be very careful about starting pressures in advance, even if we're not sure we're going to need them. And I do the exact same thing, and I think it's a, yeah, nice little reminder to, you know, our listeners is sort of like measure twice, cut once, prepare, prepare, prepare, before you go into the, into the, the battle, so to speak. It, I think it's, it behooves us to do things like have, you know, a leader of LR on pressure, ready to go, have your pressures running at a really low dose so that they're already in the line, like you said. I often will ask the nurse to have like a fetal lephrine stick so I can push, you know, 40 or 50 mics, a fetal lephrine as my sort of like rescue drug if I need it. And, you know, what you taught me, Nick, is this idea of like, you know, have two sections up and ready to go. I know that's not related to induction, but just kind of the whole planning thing in general. I think it's a good reminder to our listeners to like be thinking through all of the different elements of the intubation, not just like I have to put a tube in this person's airway and be kind of blind to everything else. And doing all of these things every time too, because it's, yeah, it's really the unanticipated challenge, which is going to get you into trouble, you know, it's so it's the time you don't prepare everything that you wish you had. All right, folks, well, we really hope that you guys enjoy this sort of bonus episode talking about the RSI trial, a breaking news in the world of critical care, airway management, Nick, any final thoughts before we conclude the show? Yeah, so I guess I'll I'll throw I'll throw out kind of a fun one, which is, you know, like we were talking about this as a Coke versus Pepsi debate, right, like which is better. And I think what this study tells us is that forcing people to to chug two leaders of Coke is not as good as sipping a normal amount of Pepsi. It doesn't tell us which one is better. It tells us that the quantity matters. And with that, if you guys want to learn more about this topic or any other topic that we've covered before on critical care time, check out our website. That's www.criticalcaretime.com. You can grab show notes from there, subscribe to our mailing list, or leave a comment. Yeah, this is part of the show we like to say thanks. Thanks to our listeners, who have left us reviews comments. We really appreciate that. We read everything you send us, and it gives us ideas and inspiration. If you haven't yet done so, please give us feedback, like five star review. It really helps other people find the podcast. I'd also say thanks to the pragmatic critical care research group because they tend to put out some pretty great studies to discuss and seem to be a definite force for good in the world of critical care. So we appreciate their hard work. We appreciate them too. They do some solid work. Those guys are great. Even if they force people to chug way too much Coke. Now we would love to hear from you guys as we always do. You can tweet X, whatever it is today. Maybe it's Y or Z. I don't know. At Crit Care Time or individually at Nick M. Mark at Askins_Razer. We're on Instagram. We're on threads. You've got our new LinkedIn page that we're trying to promote. We're all those people on LinkedIn. That's right. You know, we got we got a appeal to the masses. So we're trying to get all over the place. Bear with us as we continue to grow. Yeah, big thanks to the people who make this show possible. Our production folks over at PodPaste are awesome theme music by Kurt Bellnap. You guys you guys are amazing. Couldn't do it without you. And finally, some disclaimers. The music expressed within this podcast and any associate media do not necessarily reflect the views of our employers. All references to patients. More encounters have been modified to be hit by compliant and thus a similar use to real world cases are purely coincidental. Planning this podcast is for educational entertainment purposes only. It should only be used in lieu of seeking medical advice. With that, thanks again for listening. I am Dr. Cyrus Askin as per usual. And I'm Dr. Nick Marks. See you next time.