Subscribe
Share
Share
Embed
The EMS Lighthouse Project Podcast exists to foster knowledge translation from peer-reviewed scientific journals to the street. Join Mike Verkest and Dr. Jeff Jarvis as they shine the bright light of science on EMS practice in an informative and fun way.
LHP - E90
===
[00:00:00] I'm sure y'all remember every single episode of this podcast. And because of that, I know y'all are going to remember the great, the infamous. episode 23. That's the one where we talked about a secondary analysis of the ALP study. That's the one, the primary ALP study that compared amiodarone to lidocaine to placebo in cardiac arrest.
Now the, in cardiac arrest with a shock algorithm, obviously the secondary analysis in episode 23 was the one where they did an analysis of the effect of the drug when given by IV. compared to IO. Now remember that the parent study shocked, see what I did there, shocked, many of us when it showed no difference in survival between amiodarone, lidocaine, or placebo.
The secondary analysis shed a bit of light on that question by showing improved survival with both amiodarone or lidocaine when given [00:01:00] by IV but not IO. Maybe that was the issue, we thought. Maybe, for some reason, drugs didn't work as well when given by I. O. Now, being a big fan of the podcast, you are, you will also, I'm sure, remember episode 28, where we looked at a similar secondary analysis of the data in the ROC epistery that showed root in cardiac arrest made no difference.
That was primarily looking at epinephrine, of course, because that's. the drug that's given most commonly. Now, maybe, just maybe, it's the characteristics of the drug, like whether it was fat soluble or not, that might make a difference in whether IV or IO worked better. What's that, you say? You don't remember those episodes.
You barely remember what you had for breakfast this morning. Yeah, that's all right. I'm with you there. I didn't remember him either. I had to go back and review them to figure out what we talked about. So don't [00:02:00] feel bad. Not that you were, you know, feeling bad. I was just worried about you, and I wanted to reassure you anyway.
I'm seeing, for some strange reason, a lot of tweets from different Sesame Street characters telling everybody to be nice to anybody who seems anxious or despondent or in despair. So, that was me being nice, because who am I not to follow the sage advice of Oscar the Grouch? So the point I'm actually trying to make here, despite desperately burying it in all my ramblings, is that there have been a bunch of speculation over the past few years as to whether we should start an IV or an IO first in cardiac arrest.
There's even been discussion about whether the location of the IO makes any difference. Tanner Smita was nice enough, maybe because he's a fan of Oscar the [00:03:00] Grouch, he seems like an Oscar the Grouch. Sesame Street sort of fan to include me on a paper he did with the ESO data set. Now that paper showed improved ROSC and survival to hospital discharge in cardiac arrest with humeral head IO compared to lower extremity IO.
Better survival in that observational study with upper extremity versus lower extremity. Now, there's even been a review paper from 2022 that tried to sum up all of the available evidence on this topic. The results, all of those papers, was that we really don't know which was better. Now, most resuscitation guidelines call for an IV first approach, Compared to an IO and reserving IO only for failures.
Now, because of the uncertainty around this recommendation, ill core, the international liaison committee on resuscitation has called for [00:04:00] RCTs to help answer this question. And man, did the research community come through. I have three papers from 2024 to review today, two of which were published in the same creepy Halloween episode of the New England Journal.
I'm not sure what to make of that Halloween bit, but I'm still excited. We got these papers stick around for the details. What is a lighthouse? It is a tower with a bright light at the top Located at an important or dangerous place. The main purpose of a lighthouse is to serve as a navigational aid
Welcome to the EMS Lighthouse Project Podcast. Illuminating the darkness of current EMS clinical practice with the bright light of science. Here's your host, Dr. [00:05:00] Jeff Jarvis. Yes. Howdy y'all. I'm Dr. Jeff Jarvis. Welcome to episode 90 of the EMS lighthouse project podcast. You know, the one it's the one where we shine the bright light of science on the darkness of modern EMS practice episode 90.
Damn, that seems like a big number for some reason. So to summarize where we were coming into this year from, specifically around IV, IO, and cardiac arrest, we had multiple observational papers, including secondary analyses of data collected from RCTs that offered conflicting evidence about whether IV or IO approaches to vascular access We're better in cardiac arrest.
Remember the guidelines recommended IV first with IO reserved only for difficult access. Now, many agencies though, looked at those guidelines and said, and I'm roughly paraphrasing here,[00:06:00]
guidelines. Those aren't based on evidence. You know what? We're going to look pragmatically what we do in our system. And we're going to go for an IO approach first. Why? Well, because it's more successful and it's. More faster. It's faster and it allows cognitive offloading, meaning clinicians get to think more about other things that may make more of a difference to the patient.
You know, things like high quality, well choreographed chest compressions, Rapid defibrillation. But if root of drug administration really matters, well, we'd like to maybe reevaluate that approach. Maybe the guidelines are onto something. So to get into the first of three papers, let's head over to Taiwan.
This was the VICTOR trial and it was by Dr. Ko and colleagues. Now VICTOR is a rather tortured acronym as most trial names with cool names really are. What does it stand for? For some silly reason, I like to [00:07:00] dive into this and try to figure out what it stands for because it's not apparent. In this case, it is Venus Injection compared to intra osseous injection during resuscitation of patients in out of hospital cardiac arrest.
Sorry for that, but like I said, you gotta work for it to get the acronym. The paper, should you go looking on PebMed for it, first off, don't. In the show notes, there's the citation. You'll get right down to it. But if you do want to search for it, don't look for Victor. Look for intra osseous versus intravenous vascular access in upper extremity.
Um, in upper extremities among adults in, out of hospital, cardiac arrest, cluster randomized clinical trial, man, what a name. So where was it published July this year, British medical journal. So it was an open label cluster, randomized trial conducted with all four EMS agencies [00:08:00] in Taipei city, Taiwan. Open label means it was not blinded.
The clinician. Starting the IV or doing the IO knew what they were doing. It's going to be pretty hard to blind that I know what you're saying about. Even a blind paramedic gets an IV every once in a while, different concept. So four EMS agencies in Taipei city, which is the best I can tell all of them.
And they included adult patients with non traumatic cardiac arrest. And they did it. Most people will do greater than 18 or greater than 20. Well, in their case, they also kept it at less than 80. So it's patients between 20 and 80. They excluded the normal patients, trauma, pregnancies, DNRs, those who already had vascular access before EMS got there and patients with ROSC before you ever needed.
Primary outcome here, survival to hospital discharge. Secondary outcomes included sustained [00:09:00] ROSC and functional survival. They did an intention to treat analysis and they did the whole multivariable logistic regression, adjusting for the things, the normal, uh, steam variables, the intervention and control were either humoral head IO or an upper extremity.
IV. So remember, upper extremity only here. Now, the interesting thing about this is their medics were allowed one attempt at a humoral IO or two attempts at IV access. If they were unsuccessful with that assigned group, so if they missed the IO once, they missed the IV twice, no further attempts at access, just go to the hospital.
In other words, no crossover. Also, I might add, unless you're in Taiwan, that's probably not the approach. So just keep that in mind. I don't know if it matters here, probably doesn't, but just keep it in mind. Now overall, they enrolled 741 patients in [00:10:00] the IO group and 991 patients in the IV group. Success with IO was 94 percent compared with 58 percent in the IV group.
Easier with IO. They found no significant difference in any of their outcomes, regardless of how they stratified the results or in any of their sensitivity analyses. Now, because you come to this podcast for numbers, here come the numbers. 10. 7 percent of the IO group. Versus 10. 3 percent in the IV group survive to discharge adjusted odds ratio.
They're 1. 04 with a 95 percent confidence interval from 0. 76 to 1. 42. Now, as a reminder, an odds ratio of one means there's no difference between the groups. Now, this is the part where we have to do some mandatory stats re education. Why is that, that one means there's no [00:11:00] difference between the groups?
Well, that's because an odds ratio, it's just the ratio of the odds. The success with one divided by the success with the other, where success is survival, or Rosk, or whatever. So, it's a ratio of the odds, and any number divided by itself, according to my third grade math, is one. Now, how do we tell if a result is statistically significant or not?
And to be clear, statistical significance is different than clinical significance, because I kind of doubt in this study anybody would think 10. 7 versus 10. 3 is particularly clinically significant. All right, so how do we tell if it's statistically significant? Well, you look at the 95 percent confidence interval.
If that range includes 1, no significant difference. As a reminder, in this case, the confidence interval went from 0. 76 to 1. 42, which most [00:12:00] certainly includes 1. Now, they did look at EMS, ROSC and Neurological Intact Survival, no difference, no difference in either one of them. Now, because of that secondary analysis of the ALPS trial, they wanted to look specifically at the impact of IV versus IO among patients who got amiodarone for shockable rhythms.
Remember that secondary analysis, it showed improved functional survival when amiodarone was given by IV, but not by IO. What did they find here? When they looked at it, no difference. Remember this was a randomized control trial, not a secondary analysis, even though that amiodarone thing, IVIO and shockables wasn't what the trial was randomized for, it definitely was a pre specified, um, outcome.
So. I think that is giving us some evidence here that maybe it's [00:13:00] not making a difference. Next, let's move on to the spooky spooky Halloween edition of the New England Journal of Medicine. We're going to start with paramedic 3. Now, if there's a paramedic 3, that sort of implies there is a paramedic 2 and a paramedic 3.
Trial, maybe a paramedic one trial, right? Well, as it turns out, yes, paramedic two, as we know very clearly from this podcast was the British randomized control trial of epi versus placebo. I have talked more about that damn trial than I have with anything. So I'm not going to bore you with any of those details significantly.
Itty bitty, significantly higher survival, no difference in functional survival and almost twice the amount of neurologic devastation among survivors. Sorry about that. Got a little cough. There's some like, uh, literature based mold in the air, I guess. So paramedic two. [00:14:00] What was the original paramedic one?
Go way back. That was an RCT of the Lucas device versus manual compressions. What did they find? No difference in survival. Something that has been found in every follow up randomized control trial of mechanical compressions, pumpy or squeezy, either one. And we've even done systematic reviews on them. No difference.
Just thought I'd put that out there in the ether paramedic three, that's the name of the trial. The paper on the other hand is titled a randomized trial of drug root and out of hospital cardiac arrest. Pretty simple. Lead author is Dr. Cooper and the senior author is Dr. Perkins, who incidentally was the lead author of Paramedic Two.
So, same group. This was another pragmatic, open label, randomized trial conducted among 11 NHS systems in the UK, National Health [00:15:00] System, or service. Not sure which their health system that 11 included 10 ground units one helicopter EMS Now they included adult patients 18 or over with non traumatic cardiac arrest who again needed vascular access For drug administration.
Now that means they excluded those with early ROSC just like I did Victor trial, the one shock wonders, and those are the patients most likely to survive. I bring that up, not because it particularly impacts this trial because they randomized. Both groups were the same relating to that. But because when you're looking at their survival and comparing it to other like Cares Registry or other people's survival, it's not all comers because they exclude the ones most likely to survive.
So their overall numbers are going to be a little bit lower. Now they also excluded the patients with pregnancy, early termination, DNRs, all the things. They [00:16:00] randomized patients to either IV or IO groups, location of the IV or the IO. Completely at the attending paramedic's discretion. That's the pragmatic part.
The randomized group was about the initial access route. If they failed at the assigned route for the first two attempts, paramedic could default to whatever they felt like. In other words, rescue access site was not controlled for in the study. Again, pragmatic. Now, once they started, the assigned route was used for all medications as long as that line was patent, even if they got an additional access.
For example, they're in the IO group, they go ahead and drill, it works, it continues to work, but they started an IV just in case, as is good practice. They're not using that secondary line unless. That initial line kicked out or something, not that that would ever happen, just in case. [00:17:00] So, primary outcome in this trial, 30 day survival.
And they did include the normal things as secondary outcomes, including sustained ROSC for at least 20 minutes. Functional survival and long term survival at three months and six months. They did the same thing. Logistic regression adjusted for the usual things and reported with confidence intervals.
Now their power calculation here assumed that there was going to be a 1 percent absolute difference between groups. And they figured out based on that calculation that they needed to enroll 15, 000 students. patients. That's a massive study. As a reminder, Paramedic 2 enrolled just over 8, 000 patients. 15, 000 is ambitious.
So I'm going to cut to the chase about this. They didn't get 15, 000. They ended up screening maybe a little over 10 grand to get 6, [00:18:00] 096 patients enrolled before the trial was stopped prematurely. Why did they stop it? But it wasn't for futility, it was because the funding ran out and enrollment was taking way longer than they thought.
Again, this is a pragmatic thing, and pragmatically, no money equals no trial. So what's the bottom line on Paramedic 3? No difference. Again, just like Victor, no difference between IV and IO. again, supporting the findings in the Victor trial. Now, just cause y'all expect the details from this podcast. Let's get into the numbers.
4. 5 IO 5. 1 IV non significant absolute difference of 1. 6 adjusted odds ratio, 0. 94. And that goes from 0. 68 to 1. 32. Also, no significant difference in functional survival. 2. 7 versus 2. 8. Odds ratio 0. 91 from 0. [00:19:00] 57 to 1. 47. No difference. Now, they did, however, find significantly higher ROSC rate with the IV.
That's 21. 7 with an IO versus 24. 6 with IV. Now that adjusted odds ratio that is IO to IV was 0. 85 with a range from 0. 74. to 0. 98. That does not include one that is significant. And what that means is they found 15 percent lower odds of ROSC with IO compared to IV. So higher ROSC with IV, less ROSC with IO.
Now, is that a big deal? Honestly, I wouldn't read too much into this improved ROSC with IV thing. Why? Well, they measured the thing we care about way more, which is survival. And the VICTOR trial [00:20:00] also looked at ROSC again as a secondary outcome and found no difference. And, Because, just a little bit of foreshadowing here, the third trial we're about to talk about, well, Rosk was the primary outcome and they found, well, I'm gonna make y'all wait a little bit, um, but I'm guessing you've already picked up on the trend of this paper.
So, speaking of that third trial, let's jump into it. This was another randomized control trial. This was in Denmark, and it was the IVIO trial. Lead author is Valentine, and the paper is titled, Intraosseous or Intravenous Vascular Access for Out of Hospital Cardiac Arrest, published same episode of New England Journal as Paramedic 3.
Halloween edition, October 31st, this was again, open label because good luck not doing an open label just like Victor paramedic three, it involved all five regions of Denmark [00:21:00] and their EMS system includes EMTs and paramedics on ambulances, as well as physician field response. Now unlike Victor which had survival to discharge and unlike Paramedic 3 which had 30 days survival as the primary outcome, the IVIO trial had ROSC, sustained ROSC at for at least 20 minutes as their primary outcome.
Now they do report survival and functional outcome. They powered their study to detect an 8 percent difference between IV and IO, with a sample size calculated of around 1, 470 patients. And remember, the power calculation is based on your primary outcome. outcome. In this case, ROSC. They included adult patients 18 and over, non traumatic arrest who needed vascular access, sentences across the board.
All these trials did that. And they excluded all of the same patients. Pre existing IV access, one shock [00:22:00] wonders, pregnant women, DNRs, early termination, the things. The randomization was kind of cool on this. They randomized patients like others to get IV or I. O. But! The thing they did that was cool is if they got into the IO group, they underwent a secondary randomization to either humoral head or proximal tibia.
So now, in this way, we can get a randomized look at humoral versus proximal tibia. And I really like that because it should give us a much better answer to this question than the observational ones we've done in the past. So they enrolled 731 patients in the I. O. group and 748 patients in the I. V. group, and that met their sample size goal.
So what was the bottom line? No difference. You guessed it. No difference in sustained ROSC, 30 day survival or functional survival. All right. I hear the demands for numbers. Here they come. Primary [00:23:00] outcome ROSC, 30 percent with IO versus 29 percent with IV, adjusted risk ratio 1. 06 from 0. 9 to 1. 24. No difference.
This result about ROSC, it really helps us put that improved ROSC seen in the underpowered Paramedic 3 study into context since it was not seen in this trial that was adequately powered to look specifically at ROSC. That is, PARAMEDIC 3. Assuming that you would see an 8 percent difference. There is a deep, dark rabbit hole we can jump into about power size calculations and how we should really look at that premature stoppage on Paramedic 3.
Honestly, I think it's probably not worth it here. We now have three RCTs saying the same thing. We should probably accept it and move on. Now I mentioned that secondary randomization. So let's get into that. Of those that got initially randomized to IO, [00:24:00] 316 got randomized to a humeral head, 370 to the proximal tibia.
Bottom line, no difference in ROSC between IO and humoral head, 30 percent versus 31 percent adjusted risk ratio, 0. 98 from 0. 79 to 1. 22. Now I can only report on the primary outcome because when they're doing that, they, that secondary, um, Outcome the secondary randomization. That's not what it was powered to do, so they didn't really dive into a functional survival or anything like that.
It may be buried in a sense in a supplemental that I just haven't seen. I'll go back and take a look, but I think the bottom line no difference in Rosk. So where does that leave us? We have three well done RCTs specifically designed to look for differences between IV and IO routes of drug administration, and all three are consistent.
No survival [00:25:00] advantage to either route. Now, I think this is a great example of why we need RCTs. And why I think the arguments saying they are no longer needed in the age of regression and propensity matched analyses, I think those arguments are unwise. In this case, we had several observational trials, largely with conflicting results, but some showing benefit in IV over IO.
One well designed RCT generally is better able to answer questions than even well done observational studies. Now I wanna be clear here, not all CT RCTs are well done and the well done R CT part is important there. It's hard. It's expensive to do them, but when you do them well, we need to pay attention to the results.
They're important so. All three of these trials agree no survival advantage with IOs. Now, two papers show [00:26:00] no better ROSC with IV, including the study where ROSC was the primary outcome. Now, there was a very nicely written editorial by Dr. Nielsen in the same edition of the journal that Paramedic 3 and IV IO appeared in.
It concludes that those two trials do not support a change in the current guideline recommending IV first strategy. He says that these trials actually increase the certainty of evidence supporting an IV first strategy. Now, It's possible that I misread Dr. Nielsen's editorial, I hope that I do, because I completely disagree with his take.
I think these two papers, the Paramedic 3 and the IVIO, and the Victor trial actually conflict with the guidelines. I don't think they support them at all. I don't understand how we can say three RCTs showing [00:27:00] no survival advantage with an IV first approach approach. increases our certainty in an IV first strategy.
It just boggles my mind. I don't get it. I think these trials argue for a guideline that says there's no difference in outcomes. The approach chosen needs to depend on the patient or the system. Other things, pragmatic things, things like higher, faster success with IO versus IV, particularly in obese patients or pediatric patients, not because, and this is key, not because that faster and more successful access improves outcomes.
These papers clearly indicate they don't. But it does allow you to focus on other things. When there's no difference in the outcome patients care about, there may be still an outcome that the clinicians care about, and that makes it reasonable to consider that after you've looked at the patient oriented outcomes.
I think the [00:28:00] key pragmatic difference that we talked about is cognitive offloading. It allows the crews to worry about one less thing. I suspect all of the listeners to this podcast have worked many, many codes in the field, in the ER, and recognize that by definition, Chaos. It is chaos. There's so many moving pieces.
It is hard to choreograph. You can do it, but it takes a lot of work. You're thinking about a lot of things. Cognitive offloading in this situation, I think, is valuable. Particularly when the thing that you are, are using to buy the benefit of cognitive offloading has no survival difference. I think that's important to know.
It prevents task saturation and tunnel vision on a procedure that doesn't offer a survival advantage. And it allows more attention on those things that do make a difference. That is the highly choreographed, well executed chest [00:29:00] compressions with adequate ventilations and rapid defibrillation. Now you could argue that IO comes with increased risk of adverse events, and that would be a valid criticism.
Except for one thing, these studies looked for it and didn't see it. So, I reject the adverse events argument based on evidence. Now, instead of the conclusion from Dr. Nielsen that it supports the guidelines, I rather like the one from the VICTOR trial. I like it so much, in fact, I'm going to quote it here.
Quote, the decision between IV and IO access should be tailored to the specific characteristics and needs of each local EMS system. I like that. It's a much better fit for the evidence provided by all three studies, and it offers a more pragmatic approach to operationalizing evidence into practice. So guys, that's what I have for you in this episode, the [00:30:00] 90th episode of the EMS Lighthouse Project podcast.
Three RCTs that help us get yet more information into how to best run a cardiac arrest. This is also the first time I've had an opportunity here To say that a conclusion from a paper I was involved in is wrong upon getting better evidence. Again, I mentioned Tanner Smita, student doctor, future Dr.
Tanner Smita, showed that there was higher survival. with humeral head versus lower extremity IO. Again, Tanner let me and some of my colleagues be on that paper, and we concluded humeral head is better than tibial. The problem here though is that the IVIO trial directly looked at this in a randomized way, and I think a better way, and it gives us better answers to the question.
I now have to discard my prior conclusions. That's what science [00:31:00] is about. Making the best estimate of what the truth is, and then revising that with better evidence. I now have better evidence. My priors have been revised. And again, I want to point out, that's why we need science. RCTs. All right, y'all.
Thank you for listening and for any Dodgers fans out there, congratulations on winning the World Series. Now, while I'm disappointed my Astros didn't even make it into the series, that game won walk off Grand Slam. That was exciting as hell. Baseball just doesn't get better than that. Even if it wasn't the Astros with the win.
So congratulations, Dodgers fans, Yankees fans, at least you got to the series and you didn't embarrass yourselves too bad. So congrats. Thank you all for listening again. And thank you as always for what you do every single day. Take care of y'all. You've been listening to the EMS [00:32:00] lighthouse project podcast, a proud member of the flight bridge, ed podcast family and a fire.
Visit flightbridgeed. com for more information.