David Dillon, MD, PhD, and Ralph Wang, MD: Narcan OHCA (12/6/24)

All right. Okay. Welcome everybody. It is Friday, December 6th. We're almost at the end of the year. It's pretty crazy to see that. Today we have an amazing presentation. For those of you who are in ems, you surely have heard of this paper. It came out back in August and it really made a lot of waves. When I reached out to Dr. Dillon and he responded back as a yes, I was super excited. And his co author, Dr. Wang should be joining us here as well later on.

So this is going to be something where he'll give us a nice 15 minute presentation and would love to have a very open and robust conversation. He comes from UC Davis up in the northern part of California and he's an emergency medicine physician. He's got a PhD as well. And we cannot wait to hear about this paper. So Dr. Dhillon, welcome. Thanks for joining us. And take it away, my friend.

Fantastic. Thank you very much. So, yeah, thank you for that introduction. Thank you for the opportunity to present here. I'll try to keep this presentation reasonably short because a couple other times I've discussed this, there's been a pretty robust discussion afterwards. So I want to leave time for that. So like was just presented. My name's David Dillon. I'm an assistant professor of Emergency medicine at UC Davis. And today I'm going to be presenting and discussing a paper that came out in JAMA Network open in August, the association between Naloxone and patient outcomes and out of hospital cardiac arrest in California. Just real quick, some disclosures and conflicts. This work was funded by an NHLB IK38. Otherwise no conflicts to declare. And then some acknowledgments. This work has not been done in the vacuum. There's been many co authors, many collaborators. It's been a group that has representatives from UC Davis, UCSF, Dr. Wong when he comes on. We also have representatives University of Pittsburgh and Unity Health up in Toronto, Canada as well.

So just a little bit of background, we're really going to be talking about two related problems today. The first problem is that of out of hospital cardiac arrest. I know with this audience, I don't really need to go into this too much, but just to remind you of some numbers, there's approximately 356,000 out of hospital cardiac arrests every single year in the United States, which is around 1,000 cardiac arrests every single day. And even though we've been trying to develop new interventions for these cardiac arrests, nearly 90% of out of hospital cardiac arrests are fatal. And that number hasn't really moved in several decades. Problem number two that we've been looking at is that of opiate overdose. So opiate overdose causes over 80,000 deaths in the United States in 2023 and for several years now has been the leading cause of deaths for middle aged Americans between 25 and 64. That figure that you see is the 12 month kind of rolling drug numbers for overdose deaths. And the two lines I want you to pay attention to are the black one and the brown one. That black one is opiate overdose deaths and the brown was in synthetic opioids. There is a major spike around the COVID 19 pandemic and now it's kind of flattened off and may be decreasing over the past several months. However, we're still sitting at a level that's approximately twice as many opiate overdose deaths per capita compared to the months and years before COVID 19. So just four years ago.

When you take these two problems and you combine them together, you get opiate associated cardiac arrest. This is event where somebody either takes opiates or overdoses on opiates and then ends up in cardiac arrest. And it's estimated that between 6 and 17% of all out of hospital cardiac arrests are thought to be opiate associated. The American Heart association actually has recognized this for a while, that this is a growing problem. And they released a very thorough scientific statement March 2021 calling for additional investigation. So against this background is kind of where we, where we started to take this research project and the aims of this is to really evaluate the association of naloxone administration on patient clinical outcomes in out of hospital cardiac arrest. The idea here is that naloxone, commonly called Narcan, thought of as the antidotes to opiate overdose in patients with a pulse. We hypothesized that naloxone administration would be associated with improved clinical outcomes in patients who had opiate related or drug related cardiac arrests. How did we do this? Our inclusion and exclusion criteria. We looked at patients treated by EMS who were in cardiac arrest in three Northern California counties over approximately nine years. All the patients were aged at least 18 years or older. And we exclude the traumatic cardiac arrests or those missing any of the medication administration data in the EMS.

Electronic health record. The exposures that we evaluated was the EMS administration of naloxone as recorded in the medication administration record. And we had two real outcomes that we were looking at. We were looking at a slightly more proximal outcome of sustained ROSC. So ROSC for 20 minutes or upon transfer of care to an emergency department. So it's very similar to the CARES definition. Then the outcome, the more patient centered outcome we were trying to look at was survival to hospital discharge.

We knew that this was going to be a retrospective analysis that had many limitations and was going to be at a risk of bias. So we kind of prepared our statistical analysis plan with this in mind. You'll see there a list of covariates that we included, including age, the sex of the patients presenting cardiac rhythm, whether they had any comorbid conditions or not. That was a yes no for us. We also had whether the cardiac arrest was witnessed by bystanders or not, and then whether it was presumed drug related or not. I'm going to take a second just to kind of dive into that. Because presumed drug related in our data set was a decision made by the treating EMS provider. There was no strict criteria. It was a gestalt evaluation made by the EMS provider based on the information they had on scene when they were taking care of this patient. So we really kind of used two separate approaches to evaluate these data. First being a logistic regression where we had an interaction term. And this interaction term kind of looks at. We anticipated that naloxone administration and presumed drug related cardiac arrest would be related independently of any outcomes. You'd be more likely to get naloxone if you had a presumed drug related cardiac arrest. We also looked at two different propensity score models. Propensity scores. Just the one liner here is that propensity score looks at the probability based on the patient's covariates that they would receive the treatment. So independent of whether they received it or not, just whether they look more like patients who received the treatment or more like a control.

So just diving into the propensity score models just a little bit, just to give a little bit of information on how we tried to adjust for our anticipated biases. In this retrospective cohort, we looked at nearest neighbor propensity matching. What this does very, very simply is it takes that propensity score and then it tries to compare a patient in the treated group with a similar patient in the control group. So what you see over in that figure is in top treated group and in that bottom control group, it looks for patients with similar propensity scores or in that figure, similar colors, and it tries to compare those against each other. So even though those groups are, they look different from each other, we try to compare them as equally as possible. The second model we used was an inverse propensity weighted regression adjustment. I'm not going to go as in depth into this. But essentially what this does is it uses propensity scores to adjust the weighting of the patients. So for rare patients, that is somebody who was treated with naloxone but who otherwise looked like a control, it would increase that weighting of that patient in our analysis to try to adjust for these differences in those groups. The reason why I spent so much time on this and why this is really important is when you look at our baseline characteristics, the very first thing you should see is that every single one is statistically significantly different. Right? So these populations between the patients that got naloxone, which is that middle column, and the patients that didn't get naloxone, which is the column on the right, they look different. So patients who got naloxone are younger, they tend to be more male, they tend to be much more likely to have a presumed drug related cause of cardiac arrest. And this is all important because those can independently affect the outcome of the cardiac arrest. This is why we tried to spend so much time preparing a robust analysis plan to adjust for this as much as possible.

Jumping into our results a little bit, here's the propensity score models. It's a little bit of a busy slide, but what I really want you to focus on are these two boxes. So over on the left we're looking at the nearest neighbor propensity matching model on the top and the inverse propensity or probability weighted regression adjustment on the bottom. And these are for return of spontaneous circulation. What you can see here is that the absolute risk difference for both models was positive, meaning that in patients who got naloxone you had increased chance whether somewhere between 12 and 15% increased chance of having return of spontaneous circulation. And then that box on the right, you see that for patients who receive naloxone, you had somewhere between a 4 and 6% absolute risk increased chance of surviving the hospital discharge if you received naloxone. This is a little bit abstract. So to put it another way, if this was translated into a number needed to treat, the number needed treat for naloxone for ROSC was nine and the number needed to treat for survival to hospital discharge was 26. That means you would need to give naloxone to nine patients if this was causal. And the caveat here that we'll get into is we can't say whether this was causal or not. But if this was causal, naloxone would need to be given to nine patients to see one improved outcome for ROSC and to 26 patients for one improved outcome for survival to hospital discharge. And to put this a little bit in context, the number needed to treat for epinephrine in out of hospital Cardiac arrest is 112. So these are pretty wild numbers just to kind of put that all there, even after all the adjustment that we did.

The next results table that I'll focus on is the one with the regression with the interaction term. What this allowed us to do was to break it into four separate groups. You'll see them over there under the characteristic column. It's a pairing between drug related and non drug related cardiac arrests and whether they received naloxone or didn't receive naloxone. What I really want people to focus on are the two pairs at the bottom. Both of these receive naloxone and you either way, non drug related cardiac arrest. You received naloxone or, or you were a presumed drug related cardiac arrest. You're seeing naloxone. You'll see the odds ratios for both return to spontaneous circulation and survival to hospital discharge were statistically significant in both of these groups. So what this means is that naloxone in our population was actually associated with improved outcomes. So both return spontaneous circulation and survival in both the presumed drug related and the non drug related cardiac arrest groups. This, I have to say, was a little bit of a surprising finding for us, but it remained robust even after we tried to adjust away everything that could. So we reported it as we found it.

So there are many limitations to this study. We've touched already briefly on the retrospective nature, which there's going to be selection bias, there's going to be other things that differentiate these populations aside from the naloxone. We tried to adjust for this and create our analysis plan in such a manner that we could be as robust as possible, but you can't adjust away everything. We also don't have data on naloxone dose or timing, so we can't say when the naloxone was given during the resuscitation or how much. We can't drill down into individual cases either. So for any given cardiac arrest, we can't say why naloxone was or wasn't given. And this is important because you saw in some of those data slides that the naloxone and the presumed drug related cardiac arrest don't match up perfectly. There are some presumed drug related cardiac arrest cases where naloxone was not given and vice versa. We had some cases where it was not presumed to be drug Related, but naloxone was still attempted. Additionally, we have no data on non EMS naloxone. So we don't know whether bystanders or police gave intranasal naloxone or some other version. And finally, I'll touch on. We have no data on neuro intact survival. As much as survival to hospital discharge is a great patient centered outcome, it doesn't really matter as much if patients survive to discharge. If they're neurologically devastated. This would have been an ideal better outcome. However, we unfortunately don't have the data for it. The final caveat that I'll give here is association does not equal causation. Even though these are very, very interesting results and very encouraging, I would say we cannot say that naloxone is actually causal here. So additional research is needed. So really the conclusion that we came from, from this study was that EMS administered naloxone in cardiac arrest is associated with improved rates of both ROSC and survival to hospital discharge. With that major caveat that interventional trials, prospective interventional trials are needed to assess for true causation here.

Thank you very much.

All right, and I know that we have the addition and we want to welcome Dr. Ralph Wang. Ralph, welcome to the program. I know you're the senior author on this paper. David mentioned that you'd be joining us. So if you want to say hello and make a quick introduction and a few comments before we get into the other comments.

Oh, hi. My name is Ralph Wong. I'm an investigator at ucsf and you know, I've been working on this project with David for several years now. We're excited to really.

Dig into the role of naloxone in cardiac arrests. And yeah, thanks for the invite and I'm really happy to see the great turnout and happy to answer questions.

Awesome. So I'm gonna, I'm gonna kick it off. And I know we have other people who already put in questions in the chat. So, you know, we run pretty large systems here and for years we turn people away from jumping right to naloxone. Obviously the cops get there, they'll do it when we get there. We were pushing people to just to do, you know, standard ABCs and actually by bagging them down from an end title of 100 down to something normal, oftentimes that would actually, you know, kind of just improve their outcome that way. So we were pushing people away from just giving, you know, 16 milligrams or more of naloxone. Then your paper came out and now all of a sudden we're kind of like, it's like a shot over the bow. And now we're thinking, okay, how does this impact us as medical directors trying to steer a large system? We do about 60 cardiac arrests a month just at one single agency. And so this could impact quite a number of people. What do you say to us on the EMS side with respect to training, education and the use of naloxone, given the caveat that you just mentioned, and we get that part, does that change the way that you think about naloxone for the pre hospital environment in any way?

David, why don't you take this. But I do have some thoughts as well.

Okay.

So I think that's, that's a fantastic question and it kind of, it really drills down into the, the major takeaway from this. So we did this great paper, we did this amazing research, Yay us. But now it's like, so what? Right? So what, what is this actually going to do? How are you going to operationalize this? As a very classic researcher, I want to pump the brakes just a little bit. So I think that this is absolutely fantastic and I think that we did our best with the looking at the associations and trying to adjust the way the confounding we anticipated. But I think it is a leap too far to start changing your current clinical practice based on this paper alone. Now I want to also kind of put out there there's been several other papers, two major ones in the last couple of months as well, that kind of came out the same time. One of them showed relatively similar findings to arsh show the positive association between naloxone and patient outcomes and one was a negative finding. So I mean like this isn't like a overwhelming amount of evidence here that's come out in the past six months. And I think with that in mind, I would really recommend to the medical directors, recommend not necessarily jumping off the deep end to start giving 16 milligrams of naloxone to every single cardiac arrest and to kind of, to speak to maybe between the lines. Part of the question is for many, many years people were saying don't give naloxone in cardiac arrest because opiate overdose is a respiratory problem. Once somebody is in cardiac arrest and we start assisting their breathing and assisting their respiration, naloxone really has no role because we're already taking over that portion. And the thing that's going to be more helpful is going to be doing rigorous ACLs. So I still, I agree with the. Do rigorous ACLs get those high quality compressions in that's been shown to be very, very helpful for these patients get those shocks. However, our hypothesis and some of the data behind our work has shown that naloxone may have some effects more than just in breathing. It may have effects directly on the cardiovascular system, directly on the heart. It may be predisposing kind of the brain neurons and the opiate receptors in the brain to when ROSC happens to like really latch on and kind of run with it as opposed to just go right back into cardiac arrest. So there is some pathophysiology here saying that naloxone may have an effect more than just respiratory depression. But to bring this full circle, even with all that, I would say follow the AHA guidelines. Consider naloxone. If you truly think that somebody is in opiate related cardiac arrest and it's part of your current EMS guidelines, great, consider it, give it. Maybe it will help. But don't let it detract from the actual data driven things like aed. Use high quality compressions and transporting patients once they're stable.

Ralph, do you want to make a comment? And I'll go to Tanner and then to Dr. Pepe.

Yeah, I would echo those comments and just say that this should not change clinical care. This, we take this to mean that an investigation using a randomized clinical trial design is the next step and that could influence care. But I think this article in itself should not influence care. There was ILCOR put out a review in November that.

Said that there's no good evidence for the effectiveness of naloxone. And I personally agree. So I would hold back. I would, yeah, I, I would try to dissuade providers from focusing on naloxone and instead focus on standard resuscitation.

And actually let me, let me go to Dr. Pepe because I know you have a limited time that I go to Tanner. Paul, you have a comment? Love to hear it.

Oh, you're on mute.

Yeah, I know I had a tap to speak. It was just an odd thing that I have here. So three things is, I think you already raised the issue is what is the mechanism? You know, and I'm just really still not aware of it. I've looked into it. If you know, naloxone has an effect on other organ systems and all that. So that's the first thing that comes to mind. You know, unless, you know, you get it on board and.

Unless it comes on board and it wakes somebody up and they, we don't have to innovate them and knock them off with positive pressure breathing, you know, that kind of thing. So that, that, that's one positive possibility. So I have to think about that a lot. Secondly, and everybody's leaning on, well, we have to have an RCT. We have to have an RCT. I've really come to really distrust the RCTs that we have in emergency medicine. And I really look at other, you know, mechanisms to evaluate this. That's like, then the last thing is what is going to be the cost of this? You know, for example, If I have 100 cardiac arrests, I have to add that to my budget now. So what, what would be the, you know, I haven't looked into it. I don't know what the cost is. But you know, just, that's another thing. If it's cheap, you know, like we give these little nasal things out to people, then that maybe that's not a big issue at all, but just something else to be thinking about as we go longer. Right.

Well, the great comment. And Paul, I will say that after this paper came out, I kind of spent part of my weekend a few months back looking at this, and I found several papers regarding naloxone in sepsis that it basically.

It had that impact on the body's endorphins. And I saw one paper where in steroid resistant shock, they gave Narcan and boom, the blood pressure came right up. So even in non arrest situations, we're seeing that. So, David, any, any comment on that mechanism, per se?

Yeah, well, it could have a effect on the vascular system in that if it isn't an overdose. Right. You, you, you theoretically that whatever drug you're given, you know, can actually have all those other depressions. So that may be part of the Ross issue there.

So that, that this is a non opioid. This is a non opioid patient. The body's own endorphins are reversed. Yeah, right. That's right. That's the mechanism.

But it's kind of along the same lines, I think. I'm thinking all that. And by the way, Charlie Ingram wanted to know, when do you have free time?

He put in the chat, so. All right, well, thank you. Hey, what is the cost, Peter? Do you know what it would be for us?

I mean, I know the state of Florida is giving away Narcan for free. Okay. I don't know how much a standard dose of Narcan cost the system today, but I think it's relatively negligible. I think, per se.

Yeah, that's us thinking too. But I just want to at least raise those issues and all that kind of thing.

Right.

So. All right, well, listen, thanks. I hope that was into, you know, that was fantastic presentation. Just the fact that you're doing this is great. And it, it really is hypothesis generating in terms of what's the mechanism. Right. As we go through this.

So. So, David, any. Any follow up before we go to Tanner and then we'll go. Oh, Ralph, go ahead.

Can I ask, Paul, what, what study would you do, aside from a trial?

Well.

Well, yeah, let me get back to you on that because I'm getting a pager that I got to answer for Dr. Schepke here. But bottom line is that I'll write to you later if you don't mind, if that's okay. I apologize. Yeah, yeah. Because I do. I do have. If you look at some things that have come out recently, particularly the critical care literature, the use of.

Other techniques have been probably more worthwhile. And especially when we have biasian statistics coming out more and more with the data that we have. So I'll talk more. It's a little bit complicated to just mention it, but I'm really distrustful of RCTs nowadays and cardiac arrest. It's just one of those things. But I think there, there are some solutions that improve our outcomes. I got to go. I'm sorry about that, but really nice to meet. Nice to meet you too. And give my best to Chris Colwell. Okay, so. That's right. See you later. Bye. Bye.

Okay, David, you want to comment on the mechanism and then I'll go to Tanner, then to Charlie.

Yeah, sure.

And I'll make it super quick because I think you already kind of touched on one of the. So Dr. Pepe, I think hit the nail on the head here, is that the simple answer is we don't know. There have been some papers out there that you mentioned that this idea of in non opiate related shock situations, there's a Cochrane review that showed that there's an increase in blood pressure. There's also been some papers out there showing that they're in cardiac arrest. When you give Narcan, you have changes in EKG rhythm. So there are opiate receptors on the heart itself that could be positive. You also have this whole idea about opiate receptors in the brain and predisposing towards latching onto that ROSC state. I think finally there's been just straight up animal literature saying that in opiate related cardiac arrests in rats, when you're given naloxone, it also just is associated with ROSC in these studies. Although again, the exact physiology is a little bit unclear and I'm not quite sure we're going to know in the next several months.

Okay, excellent. Tanner, we like to bring you in and then Charlie, you're next.

Yeah, I appreciate it. And again, first, I want to thank you both for coming on and presenting this to us. And I just want to say first that at least from the perspective of a student of epidemiology and biostatistics, from my perspective, I think this was a very rigorous study. You did everything possible with the data you had. It was a well written study, well balanced discussion. So just wanted to say that up front. So I have three comments and I'll give them all to you at once as not to monopolize time and you can respond to whichever one you want, if any.

So first of all, I think that the biggest limitation, kind of speaking to the group here of these sorts of studies where you compare delivering an intervention to not delivering an intervention is what economists would call omitted variable bias. Or in medicine, we call it confounding by indication, right? As healthcare clinicians, whenever we administer a treatment to a patient, there's typically a reason for it, right? It might be physiologic state, it might be things like, you know, the area of practice, it could be the training of the provider. And because of that, you know, you end up, as in this case, with two groups of patients that differ pretty dramatically in baseline characteristics. Now, you touched on this in the presentation. Patients who received naloxone were about two decades younger. About 40% of those patients rested in public places versus 20% of the patients who didn't get naloxone. I thought it was really amazing that you were able to include comorbidities in this paper. And about 80% of the patients who got naloxone had no comorbidities, compared to only about 50% of the patients who didn't get it. And so I think the first thing to mention, when you have kind of these drastic baseline imbalances, there's the possibility, regardless of how good your statistical models are, of residual confounding. So just because you put, for example, age into a statistical model doesn't necessarily mean that that totally accounts for all of the bias introduced by age. In addition, right. There's always the possibility of unknown and unmeasured confounders. So there are some utine factors that weren't able to be adjusted for in these analyses. Things like response time, for instance. Some other kind of possible confounders that I was thinking about in relation to these analyses was, and you mentioned these again, kind of the route of drug delivery, the dosing of naloxone, whether the naloxone was administered pre arrest, post arrest, during active resuscitation. There are also probably things that, that predict administration of naloxone other than things like age and race and sex, etc. One thing that I thought of is the presence of agonal respirations when providers arrive on scene. And we know that, you know, perhaps especially in an area where there's frequent opioid overdoses, those patients might be more likely to get a dose of naloxone even before cardiac arrest is recognized. And then also, of course, you know, agonal respirations are kind of a positive prognostic indicator for any cardiac arrest patient. So that' another confounder. I also noted some of the differences kind of persisted as you looked at the treatment of these patients. So there was differences in the proportion of patients who received epi. So more patients that got naloxone also got epi. Also, interestingly, among the patients who got naloxone, less of those patients there was about a 15% difference in advanced airway management, about a 10% difference in advanced airway management. And so kind of, that's my first comment. There's always the potential speaking to the group here about residual confounding, unknown and unmeasured confounders. As a related comment, another thing I was thinking about as I was reading through this is the possibility for the inclusion of patients who were never truly in cardiac arrest. So it's kind of a pseudo PEA population. I know that the evidence really shows that regardless of level of experience, whether you're a physician or a paramedic, we're not fantastic at palpating pulses. And on the floor with the potentially bradycardic opioid overdose patient, I imagine that kind of compounds the difficulty. And so I kind of imagine a scenario where there's a person who, you know, maybe they don't palpate a pulse, they get a cycle of cpr, you know, naloxone gets administered, respiratory drive returns, and they say, okay, well, now we have Ross, right? But that patient was never, never truly had a cessation of circulation. I've noticed that looking at, I've worked quite a bit with the ESO annual data sets and looking at their outcome data, there's a very small proportion of cardiac arrest patients who leave from the emergency department against Medical advice. Just, just again, as I looked through the data set myself, I found that was interesting and kind of working backwards. A lot of those patients are patients who had initial rhythm of asystole or PEA, were drug overdoses, returned to a GCS of 15 in the field. And so I think there is a population out there of patients who are not truly in cardiac arrest or being categorized as cardiac arrest by ems. Again, nothing you could do about that with your data, just something to think about. And then so finally, and we've already talked a little bit about mechanism, but I wanted to talk a little bit about some of, kind of the laboratory and clinical rationale for using naloxone in cardiac arrest. And actually, I do have one slide to share. I pulled together some figures from some of the animal papers out there. And so some of the animal studies that exist previously are basically rat models of asphyxial cardiac arrest. So they take rats, they clamp their endotracheal tubes, wait till they get to a mean arterial pressure less than 10 millimeters of mercury, leave them there for five minutes, start CPR, administer naloxone. And in those studies, as you noted, they did find better outcomes, better hemodynamics in animals who received naloxone compared to saline placebo. However, I noticed that the doses of naloxone that they were giving were heroic compared to kind of the doses that we administer in clinical practice. So the figure on the left hand side of the screen here, or actually we'll start, we'll start on the right hand side, the bottom group. So the patients with the low or the rats with the lowest coronary perfusion pressure were, that was the saline placebo group. And then they had a group that received 0.5 milligrams per kilogram of naloxone and then a group that had 1 mig per kilogram of naloxcone. Again, for 100 kilogram person, that'd be 100 milligrams of naloxone, which I don't really think reflects clinical practice in the United States. And that's the dose that they had to give to basically get equivalent.

Ross.

To administering the equivalent of 4mg of bolus dose epinephrine to these rats. I also found there was a single animal study in pigs which are probably a better corollary to human cardiac arrest. I've personally worked with rat asphyxial cardiac arrest models and I will say that it's very hard to keep a rat in cardiac arrest. And in this pig Studies, they actually use 10 milligrams per kilogram of naloxone. So they were giving a gram of naloxone to each of these animals and they found basically no hemodynamic changes at all in these animals. I think at 30 minutes there was a slight increase in mean aortic pressure, something like that. But for coronary perfusion and a lot of these other important metrics, they found no difference at all. And so I think that the, the animal data is still a bit mixed. And as far as the. We talked already about the patients who experience increases in mean arterial pressure and hemodynamics for patients that are in sepsis or have sepsis are in the icu. Most of those studies enrolled patients who are already on very high doses of pressors, mostly dopamine. And there is, there is that. That Cochrane review you mentioned showed about a 10 millimeter mercury increase in the mean arterial pressure for patients who receive naloxone. Kind of within that patient population, there are animal studies that suggest that while naloxone benefits animals that are in kind of endotoxin precipitated shock in other pathophysiologies, so like hemorrhage in shock that's precipitated by ischemic guts, et cetera, naloxone does not have an effect. So I think it's probably still an open question as well whether naloxone has beneficial hemodynamic effects for kind of, you know, the population that we're interested in, these human cardiac arrest patients following opioid overdose. And so I'll leave it there. Thank you again very much for presenting this paper. I really enjoyed reading it, very thought provoking and I look forward to more work from your group and others.

Yeah, I mean, I'm going to be completely honest, absolutely no notes. I think that you hit like all, like this is a very complex topic and I think you hit on the things that we've discussed internally and externally previously and had a very good summary of both the lack of definitive data on both a animal model and in a human population. Which to circle all the way back around to the original question, should you be changing your practice?

No.

At least not based on these data alone. And then also this idea of pseudo pea. I'm right there with you is very real possibility that like one of the things, one of the reasons we saw.

A positive association here is because even though they were classified as cardiac arrest and they were treated as cardiac arrest, were they really in cardiac arrest? We might not be so sure. So honestly I have nothing really to add. I think that was a fantastic summary of both some of the difficulties with this work, some of the limitations, and then also some of the other hypotheses as to why the naloxone association may be seen.

And now you know why I have Tanner as a co host. Because he, he, he, he brings the brains to this operation. Now let's go over to Charlie Ingram who trained at Cooper. So Charlie, I don't know if you have any bias to the, to the Cooper paper, which was a negative paper, but that Charlie Ingram, medical director with me here down in South Florida, love to hear your comment.

Yeah, so I did, I did train at Cooper. Some of their data set was probably me giving that naloxone in the field when I was a resident back in the day. So, you know, mild. But no, I mean I, I think it's refreshing to hear that you guys like put off this like massive paper and I think it, it definitely sent some shockwaves across kind of the bow, which I think is good. And it didn't change my practice in the ED nor in the EMS systems that we run. But it did make me consider saying we could pull off this study. We could, we have, we have the data sets to do something prospective like this.

Right.

And so I think that that's where it really kind of provoked thought for me. My biggest, my biggest thing was what I was hoping to see in your paper was if like I think the witnessed unwitnessed is great, but I wonder if they bystander cpr. And I didn't see that anywhere. And the data is not available. The data is not available. And unfortunately even something where the data is available, is the data reliable? We just don't know anymore. And so it's tough. And then like you mentioned, like those neuro outcomes for me are huge. But also because like there's been some kind of thought of like the naloxone in arrest after they're already in arrest worsens their neurologic outcome. And so like it's questionable, like am I saving organ donors or am I saving people with outcomes? It's so hard to say. And so I think that like whatever we as a physician community do in the next five years, making sure that we have those outcomes and just to make sure that we can try and link them and then the kind of nitty gritty data of like when was it given in the arrest, was it given in the arrest or was it given just before the arrest or just after the arrest? Like those are kind of the points that I think is interesting. Back to Paul's comment on like cost. Interestingly the IV and like the injectable solutions are cheaper than the nasals. And I like personally, Narcan is given with a needle. For me, whether it's IV or IM or something. I think that like when we do these studies, like we should be able to control the dose. We don't know how much is absorbed through the nose, especially in these shock states or arrest states. And so like I think when we give the needle, the iv, the IO, et cetera, I think it's a better study.

So.

But no, it's an exciting thing to, to definitely look at all of these, these data sets and, and take a look and see like what's next for sure. So awesome work guys.

Thank you. I think one of the things.

Go ahead.

Yeah, sorry.

Go ahead, David.

One of the things I want to highlight, I agree 100% with everything you just said there and one of the things I wanted to highlight as well is this whole idea that even if you drink the Kool Aid and you think naloxone might be working, there is this risk that we haven't touched on yet that opiates have very firmly been shown to be neuroprotective. And like post Ross, post isch. So if we're going and reversing all of these opiates, whether it's exogenous, endogenous, doesn't matter. Are we actually causing neurologic harm to these patients by giving naloxone? Sure, you get Ross, but then you're neurologically devastated. And that's one of the risks that's kind of baked into this of even if you buy that naloxone may be associated with ROSC and increased survival, if we don't have those data on neuro intact survival, we're not going to be able to directly answer that question. So that was one of the reasons why it was highlighted as a limitation. So I do think that that is a very, very important question to answer if we're moving forward with this research.

Totally.

And it begs the question of like, why do we have that data that says that it's neuroprotective? Like it might just be because all of the patients that are awake enough after cardiac arrest get fentanyl drips. Like, who knows? Yeah, totally.

And then, and I just, I just wanted to comment, just, just about. Just go to Ralph's comment earlier, the discussion we're having with Paul and if I had to guess And I'm just, you know, I don't speak Paul Pepe, but I, but I can maybe assume is what he was getting at is that, you know, he's been involved in some rtts in the past, that because of the variation in the quality of cpr. And I'll take you back to the ITD study where as everyone knows, they only found a significant difference once they relooked at the data and only looked at the agencies who are actually performing quote unquote back then, high quality CPR with compression rates within 100 to 120, then they found the difference. And so then this whole concept of it's the bundle. Right. And so I think that the question of cardiac arrest care that's going to come up even if in his mind, I think if you do the RCT is, you know, is the CPR being done correctly? Was it that one item like people are trying to figure out intubation or supraglottic or BVM work. This doesn't figure that one out either. So I think his reservation is you're going to go and do this entire RCT and then people are going to come back to you and say, well, we don't really know whether it was a positive or negative study. They're still going to come back and say, well, it's because. And they'll throw out one of five things that would suggest why that didn't work. You see what I'm saying? Yeah, I think that's that, that's, that's his problem with it 100%.

And the one thing I see, we have a hand up, so I'll make this quick. But the one other thing I'll bring us back to is this whole question of like, what are the outcomes we truly care about as well? Because on one hand we were just discussing about we want neurologically intact outcomes for these patients. That's what we want to see. That would be fantastic. But then when you actually take a step back and say, okay, me giving naloxone and out of hospital cardiac arrest days to possibly months prior to this outcome of out of hospital or like neuro intact survival, is that truly like a real outcome or is it the fentanyl drip that they got? Is it an improved ICU care? Is that the level of center that they. It's so founded that it's very, very difficult to tease this out and also tease out outcomes that truly matter for these patients. I know we're running a little bit over time and I see a hand so I'll.

Yeah, yeah, yeah, Lou, go ahead.

Yeah.

You noticed in your non shockable rhythms you had about twice as much as systole as pea. But I didn't see differentiating as far as the outcomes. Presuming as systole is certainly cardiac arrest and PEA might just be profoundly impaired circulation. Are you able to differentiate if that mattered?

So that was not something that we looked at. We looked at it in a subset and I don't think we moved forward with it. We are doing another study with a much larger data set, a national data set that would be powered to break those apart because at least for this, it would be difficult to break out the pea, break out the assistantly in the data set that we have. So I agree 100% because I think that there have been other papers that look specifically at a PEA subgroup as opposed to just a non shockable subgroup because I think there is a hypothesis there of the pseudo pea. So that is not something that we were able to do with this paper, but that is something we're doing with an upcoming larger study that we're hoping to publish shortly.

Thank you.

Excellent and great question. Go ahead, Ralph.

We think if it's going to work, it's going to work in pea.

Okay. And so just to kind of start to end here, I wanted to say that I gave a talk in Vegas and I mentioned your paper. This is talked to about, you know, about 150 paramedics. And I had the paper up there and obviously people were like looking at me like I'm crazy and the whole thing. And then one paramedic comes up to me afterwards and he's shaking a little bit. I said, what's going on? He says, I just saw what you just said. And he says, I have three cases of young women, they're all in their 20s, who I thought were all drug addicts, they were on cardiac arrest. So I gave them Narcan. He's like, I followed up on all three of those girls afterwards and none of them were drug addicts. And so he, he goes home, he emails me back and this is like the non scientific part of this whole thing, but he emails me, redacted the three charts because he wanted me to believe and that look, this could be something here, but kind of bringing it back full circle and I'm going to give you both the last word here, which is yes, this is great, it looks like it could be something, but let's pump the brakes, let's do what we currently do and let's Let the professionals like you guys and Tanner and other smart researchers take the wheel and help us guide practice in the future. So, David, we'll start with you and then end with Ralph. Give us some parting words. And again, thank you so much for coming on the webinar.

Of course. And thank you again for having us. I think this has been a fantastic discussion. I think my parting words are to just kind of reiterate what you just said. I wouldn't change practice. I'd pump the brakes and start and give naloxone to everybody. I think we've had a very robust and actually like very thorough discussion about how while this paper was really interesting and while our work and the work of several other groups has been really thought provoking over the past six months, it's not enough to change what we're doing for patients right now. If you're giving naloxone in some cases because you're following the AHA's consider guidelines, great, continue to do it. But I wouldn't start rolling out in naloxone for everybody program just based on these data.

Excellent, Ralph.

Yeah, I would say we're at the beginning of discovery of a journey and.

We haven't really scratched the surface.

It's very refreshing, I must say, to hear you guys put a lot of work into this. A lot of us do research and we want to put a lot of work into our project and then we want to see it change care. It's very refreshing to hear both of you come on and say, listen, we put years of our lives into this, but there's still a lot more work to be done. And so as a fellow scientist, not as smart as you guys are, very nice to hear that from the both of you and very appreciative that you joined us today. So from all of us here in Florida, thank you very much and we hope to meet you guys in real in real life soon. For those of you on the call, NAMSP is in January. The Florida meeting. The dinner is sold out. Unfortunately, we've reached capacity of like 70 people. If you really want to make it, and maybe I'll just kick Charlie out or something, but just give me a call. We'll see if we can figure it out. Thanks everybody. David, Ralph, thank you so much. Tanner, thank you so much. I'm going to put in the Dr. Pepe webinar link here. Unless, Tanner, you have it handy. I got to just search for it real quick.

I do. I got it in there.

Oh, perfect. You already did. You're the best. Okay, we'll see you on the other side. Happy holidays, and we'll see you next week.

Bye, everybody.