How do you reverse neuromuscular blockade? Part 1

Welcome to Depth of Anesthesia. I'm David Hao and I'm an anesthesia resident at the Massachusetts General Hospital. If you haven't heard, our first episode was recently featured on ACRAC. A big shout out to Dr. Wolpaw for his support in helping to share our content. Another new and exciting development for the podcast is that Anesthesiology News will be featuring some of our episodes on their website.

They're a go-to resource for me for up-to-date information and interesting content, and I highly recommend you check them out. My guest today is Dr. Daniel Saddawi-Konefka. He's an attending anesthesiologist in the Critical Care Division, and he's the program director here as well. Dan, welcome back to the show. Pleasure to be back.

So for our case today, we have a 42-year-old woman with no past medical history who's presenting for a laparoscopic gynecology case. She was paralyzed on induction with Rocuronium, which was redosed three times. At the end of the case, the resident elicits four twitches with no fade using a facial nerve stimulator It has been 60 minutes since the last dose of rocuronium, and they're planning to proceed to extivation. Their attending makes them recheck twitches on the ulnar nerve, and when the attending feels no fade, they then proceed to extivation. Dan, what do you think are some of the claims in this case?

Wow, it actually it seems like there's a bunch of claims here. Uh and again, this is in the context of a healthy patient who's been paralyzed and redosed with rocuronium. So it seems like the residents' practice habits suggest that they believe the facial nerve stimulation is sufficient for excavation. Or maybe they're claiming that if it's been longing long enough since you're redosing, you don't actually have to re-check any differently Uh the attending also has some claims, I'm inferring. Uh they suggested that the facial nerve maybe was inadequate because they wanted to recheck on the ulnar nerve, so there's a claim there that the sites are different They then felt no fade and proceeded to extubation, so there's a claim there that feeling fade is sufficient for extubation.

So just to succinctly state these claims First, the site where you checked witches matters. Second, you can go with how long it's been since the last dose, as far as whether or not you need to reverse Third, that you can discriminate fade just by feeling. And then another one I suppose is what fade ratio is acceptable for extivating So to our listeners out there, what's the level of your agreement with the claims that Dan just talked about? And what's the level of evidence for what you believe? And Dan, we're gonna be doing something new this time, right?

With uh two-part series here. Yeah, I think it'd be quite a lot to try to bring all these claims into one episode and analyze even just one or two articles per claim, and also keep it you know, something manageable for someone to listen to. So we'll we'll do the first few claims, the the site for checking the uh sort of the pharmacokinetic argument of it's been a while and then how much uh anesthesiologists can discriminate by looking or feeling. And then we'll leave the Well, what does it matter and what ratio do you use to determine whether you're not to uh reverse? We'll leave that for the next episode.

Perfect, sounds like a plan. And to keep this manageable, we've tried to pick just one or two articles per claim. When we looked through the literature, we tried to pick whatever the highest level of evidence. existed was so if that we only had observational studies we picked observational studies. If we could find something randomized, we'd do that. And if there was nothing, if it was just a couple of reviews, we'd we'd pick that.

Alright, so let's get started with our first paper, and this is a paper in anesthesiology that's titled Intraoperative Neuromuscular Monitoring Site and Residual Paralysis Right, so this is to look at the claim that the uh facial nerve is just as good as the ulnar nerve. Alright, so this is a prospective cohort study, 180 patients Um and uh this is the one paper that we'll look at to assess this claim. There are plenty more and and as always we encourage you to To take a look at some of the other evidence if there's something you think we should have talked about that we didn't uh make some comments. So what they did was they looked at 180 patients that were scheduled for elective surgery and uh where there was anticipated use of neuromuscular blockade.

And they did this at two institutions. They excluded extremes of age, so kids and people over 80 Uh and the providers, this is kind of an interesting side note, they did not really tell them much about the aim of the study. They didn't want to color their practice. And they the people knew that there was a study ongoing, but they weren't told whether or not this specific patient was going to be in this study. The providers had access to qualitative TRAN-4 monitoring.

So in other words, they could not quantitatively assess in the OR what the TRAN-4 ratio was. Toph watches did have the availability of uh tetany at 50 or 100 Hz, but again no quantitative measure. And what they did was they they looked at how much residual paralysis was there in the PACU and determine whether or not there was residual paralysis, they measured TRANO4 ratios within five minutes of PACU arrival. So what was their definition for residual paralysis? So they measured TRAN-4 ratios as soon as the patients got to the recovery room.

Any train-of-4 ratio less than 90% they defined as residual paralysis. Now something else about the paper, right? This is comparing different sites. They did not prescribe which site the providers would monitor at In other words, it was up to the providers to pick the site, and then they just did an association, a correlation of their residual paralysis outcome with the site that the provider had selected So it was basically up to whoever was doing the anesthesia. That's exactly right.

So just to summarize again, you got a got a bunch of patients going in for surgery. The providers are doing qualitative monitoring either at the facial nerve or the ulnar nerve. Just standard of care type stuff. And they're looking for the incidence of residual paralysis and they're gonna see Did the site make a difference? Alright, so what did they find?

So they included 150 patients. 99 of them had had facial nerve monitoring, so ostensibly they're looking at the orbicularis oculi or corrugator supracelii muscles. 51 patients had had the ulnar monitoring, so they're looking at the ad doctor polysis. Of the facial nerve monitoring group 52% had residual paralysis. That was 51 out of 99 patients. So those patients in the PACU within five minutes of testing. had a train of four ratio less than 90%. That's right.

And of the Olmar group, 22%, or 11 out of the 51 patients? had residual prowls and that that's uh that difference was statistically significant p-value of less than 0.01 Now what if they hadn't used a cutoff of 0.9? What if they used 0.7? So I don't know about the statistical difference between the two groups. But the incidences would have been 17% in the facial nerve group and 11.8% in the alnar nerve group So still ostensibly higher uh residual paralysis in the group that just teased the facial nerve. Fair enough.

Now, did these groups differ by any other metrics like hospital or surgery type? Yeah, those are exactly what they differed by. Because remember this is this was not a randomized uh trial, this was an association trial. So they did have different surgical types. And in particular the eye group, the facial nerve group were more likely to have had abdominal or thoracic surgeries.

So when I read that initially I thought Well, if it's abdominal or thoracic, maybe they're getting more paralytic, and that's why there is a difference. The problem with that Assumption though is they actually report the amount of paralytic that the patients got. Between groups, there was no difference. No difference in how much they got, when the last dose was. Those were all matched And even if they weren't mentioned, they actually did a multivariable logistic regression that adjusted for a bunch of things.

Age, gender, hospital, ASA class, time from the last dose. Type of surgery, duration of surgery, BMI, all of those things. And there was still a stick statistically significant difference between groups. The group that was monitored with facial nerve had an odds ratio of over five more likely to have residual paralysis. Gotcha.

So uh what's the punchline for this study? So in this observational study, patients having qualitative train of four monitoring of eye muscles had a greater than fivefold higher risk of post-operative residual paralysis compared to those who had monitoring at the adductor palsis Now, one I think still important residual confounder is the fact that you've different providers And they were, you wouldn't be able to do adjustment or regression at the individual provider level. But if there were differences in the providers, that might muddy the results Just the same though, this is the level of evidence we have about differences. And and I will say there's biological plausibility, right? The the thought that different muscle groups recover at different rates is something we see in the operating room all the time.

Yeah. Yeah. You've got no twitches but the diaphragm's moving. So this at least passes the WIF test. And again, this is an observational study, right?

So it needs to be interpreted in the context of how it's designed. We cannot establish causality-only associations. That's well put. Not related to this site claim, but just a couple of other interesting points One is that residual paralysis was more common in the regression in patients with higher BMI. The other thing, and this will tie into one of our later claims, They saw that the time since the last dose of paralytic was related to the Likelihood of residual paralysis.

Kind of on a on a gross scale, every 10 minutes that went by since the last dose of paralytic was associated with a 10% reduction and the risk of residual paralysis. So that might tie into the next claim. If if you want to introduce that, we can look at some of those studies Yeah, definitely. So the next claim that some people will make is that if it's brit if it's been a certain amount of time since the last dose of paralytic, it may not be worthwhile to even check twitches. What kind of studies do we have investigating that claim?

So for this claim we picked out two studies. The first was published in 2005 in Acta Anesthesiologica Scandinavica. The second was published in Anesthesia and Analgesia in nineteen ninety-five. So this first one is called Variability of Duration of Action of Neuromuscular Blocking Drugs in Elderly Patients. So this was only in older patients.

So it may not apply to our younger patient in the example that we gave just up front. This was 66 older patients all over the age of 60. In the paper, they state that the patients all had normal renal and liver function. I do want to point out that they defined normal renal function as a creatinine of less than 1.5 Which in some older patients doesn't feel totally right to me. Sure.

Anyways, these were all patients, these 66, who are having elective procedures that require general anesthesia with endotracheal intubation. Any exclusion criteria? Yep. They excluded anyone with neuromuscular disease. And we'll see that as a common exclusion for all of these neuromuscular blocking drug trials.

And they also excluded obesity, so anyone with BMI over 30. And that fits with the previous paper we looked at because the incidence of residual paralysis was higher in that population. So in this study, they randomized these 66 patients to receiving one of three different drugs. either Rocuronium, Vecuronium, or Cisatracurium for neuromuscular blockade. And they used doses that were about twice the ED95. So that's 0.6 per kilo for rocuronium 0.1 per kilo for vecuronium and 0.1 per kilo for cis.

Now they were comparing the variability between these three. I think what's more interesting for us is actually to look at the variability within each group, because if we're making the argument that if it's been so long since our last dose were okay, then we would want to see that there's a fairly tight range So how do they define the times in the study? Right. So it was how long it took after the in induction dose of paralytic, one one-time dose. To get back to 25% of a control twitch height.

So what that means is they calibrated the accelerometer to get the initial the T1 height And after it went down, they intubated at twenty-five percent of the T1 height Then ostensibly the height would get lower as the paralysis becomes more dense, and as soon as it came back up to 25%, that was the time. So Dan, I think I'll I'm wondering, and I think a lot of listeners would also wonder, what exactly does that mean to have your T1 height returned to 25%? Right. I think most of us use train of four, my myself included. Yeah.

So turns out that as the as the neuromuscular blocker starts to work your T1 height starts to diminish When it gets to about 25%, which is what they use here, that's about when your T4 twitch disappears. Okay. So basically a 25% hype for your T1, that's about a train of 4 ratio of zero. So what do they find with respect to the duration of action of these three medications? So the median duration of action of each of them was pretty similar.

Vecuronium was 62 minutes, rocuronium 63 minutes. Cisatracurium 57 minutes. And again, that's to go from basically zero train of four back to just at the cusp of getting your T4 back But what was really interesting about this was the variability. So cisatracurium had the narrowest range of variability, which maybe makes sense based on its metabolism. But even that, and remember this is a fairly small study, 66 patients total, the range on minutes was 37 to 81 minutes.

So almost half an hour to almost an hour and a half For rocuronium, it was 33 minutes to 119 minutes, a half an hour to two hours. And for vec eronium, it was 35 minutes to 137 minutes to over two hours. This was with a single intubating dosage, right? That's right. That's right.

That's pretty remarkable. Yeah. And it's it's hard to know what that variability is due to. It's hard to blame it on any single um say organ dysfunction because you even see these differences with cisatricureum. Right.

I think what it does suggest though is that it's hard to um sort of put all your stock in a specific amount of time. So the punchline is There's a pretty big range of durations given the same doses of neuromuscular blocking drugs in this elderly population. Things are probably better with time, as was suggested with the first paper that we looked at, but it might be hard to predict when any individual patient would be, quote, okay to reverse or okay to extubate And we have one more study on this claim, right? Right, this is uh called Reversal of Residual Neuromuscular Blockade with neostigmine at one to four hours after a single intubating dose of Vecuronium.

So this is a papal paper by James Caldwell from 1991 out of UCSF He was studying something that's a little bit tangential to what we'll talk about in the paper. He was studying different doses of neostigmine and their effects on neuromuscular blockade after a single intubating dose of Ecuronium. So similar to the last paper, they used 0.1 milligrams per kilo for vacuronium for their intubating dose. And again, that's a pretty typical dose we would use to intubate someone. And they looked at 60 patients who were having general endotracheal anesthesia.

These are all ASA-1 and II patients, no kidney or liver issues, no neuromuscular issues. And these 60 patients were allocated into six different groups. So they were reversed with neostigmine either 40 mics per kilo or 20 mics per kilo at either one, two, three, or four hours after that intubating dose of acuronin. Okay. The really relevant part for us was what the train of four ratios were before and they measured them after, but I think for you know, assessing our claim here that time is enough.

You don't actually have to check, you could just excavate. Their train of four ratio is measured before. So again there's ten patients in each of these six groups that they had And at one hour after the intubating dose, the train of four ratio, the uh median train-of-4 ratio, was 0.29 with a range across ten patients of zero to point eight six. The range was zero to point eight six? Yeah.

For ten patients. Yeah. And that's after one hour. What about at two hours or three hours or four hours? Okay, so at two hours, uh the median was 0.87, so that's pretty good.

Yeah Problem's the range again. It's 0.43 to 0.98. At three hours, i the median was 0.9. At four hours the median was 0.91 But again, the the ranges for three hours it was 0.62 to 1 and for 4 hours it was 0.68 to one. So even at four hours after a single intubating dose, you have someone o of these ten patients who's hanging out at. 68 for their train of four.

I don't want to jump to conclusions, but I'm sensing a theme here. Yeah, and well, and I don't know I don't know if you've ever had this where you have a patient in the OR. and gosh it's been a while. You decide to check twitches just for the heck of it. You can't explain why it's low when you go around to different sites. Right?

So your or your twit they call for a different Twitch monitor. I've had that exact case. So at least this paper, small paper, would suggest that it's not a twitch monitor issue. And that makes sense because in the case that I'm recalling, we actually gave Sugammadex to reverse the Rocuronium. and demonstrated pretty quickly that it wasn't the twitch modded at all. 'Cause the train of floor increased? Yeah.

Now this is uh Vecuronium uh 1995. This is uh that paper. So um the punchline here is It seems like there's a good amount of variability. Even in again, this is ASA 1 and 2 pretty healthy patients. So if you were to introduce a little bit of kidney or liver disease.

I don't have the data here in front of me, but you might imagine that the variability would increase. That's fair. So I think that wraps it up for that claim. And we have one more claim, right? Two more claims Right, one more for this show and then we'll explore the other one in a lot more depth in the next episode.

Perfect. And so our final claim that we'll review in this episode is that if you don't have quantitative train of form monitoring feeling or even visually looking at looking for fade would be sufficient. And we have a very nice paper here called Tactile and visual evaluation of the response to train of four nerve stimulation. What do they do in this study, Dan? So this was a study done in three parts.

We'll talk about each of the parts. So in the first one, what their question was, was if you take inexperienced anesthetists, people who know about train of four ratio, can they reliably detect fade? So they took twenty-nine such individuals. Who assessed FAID on 30 patients, and they would do it more than once, for a total of 95 evaluations by these 29 individuals. Each of the patient was their own control.

So while they would check Trainophore quantitatively on one hand, the anesthetist would assess For fade, it was just binary. Is there fade or not? On the other hand. And they had 95 of those evaluations. So what do they find?

Could inexperienced observers detect fade? Unfortunately not very well. So they would do sequential tests and they'd either estimate visually just by holding the hand and looking. or manually by holding the thumb back and seeing if they could feel fade with sort of a constant preload sort of thing. And when the train of four ratio in these patients was point three or less, so pretty pretty good amount of fade. They were all pretty good at telling.

But when the Trinidad ratio was not so obvious, say 0.4 to 0.5 , Only 37% could detect it visually and only 57% manually. There's no statistical difference between those. And when the train of war ratio was 0.6 to 0.7, to levels we still wouldn't consider acceptable, at that point only 8% and 16% could detect fade visually or manually. So inexperienced observers not so good at detecting fade. Wait, wait, wait, but this is for inexperienced observers.

What if you were experienced? Right, so introduce group two So these were they called them moderately experienced observers. This was 24 anesthetists. And these folks they didn't have any special interest in neuromuscular monitoring, but they all had at least two years of experience in the use of nerve stimulators. So these 24 anesthetists did 80 responses in 29 patients, same setup as the last one.

And there was no statistical difference between their performance and the inexperienced synesthetist performance. The specific numbers, so 77% could detect fade when it was less than 0.3. About half of 'em could detect fade if the terrain of four was 0.4 to 0.5. And if the terrain of four was 0.6 to 0.7, it's only about a third. Alright Dan, I know this third group is coming, so what about people who are really into neuromuscular monitoring as much as you could be into it?

So they're hard to find. So this is group three, and they only had two observers. But these are folks with long experience in neuromuscular monitoring and a special interest in as well. So the idea here is, could you take real, real experts and then maybe they can detect fade? Um There was no statistical difference between how they did and how group 2 did.

There was a statistical difference between how they did and how the totally inexperienced people did. But they still missed a lot, so they were actually able to detect fade of less than 0.4 reliably But when you get in the. 4 to 0.5 range, only two out of three times could they detect fade. 0.6 to 0.7, they only detected that fade a quarter of the time. And we'll have to put the graph from this paper on depth of anesthesia. com because it's absolutely striking just seeing a visual representation of how well these different groups performed. Yeah, it's it's pretty incredible. To me, I'm not very good at darts, but if there was like a dartboard and here's the actual tough and I'm supposed to throw it at that Like it'll be kinda closer to it than maybe something else, but man, there's a huge what did what do they even say here?

But note the wide scatter in estimated ratios. Alright, so what's the punchline for this study? So the way I take the punchline here is that uh it's very difficult, if not impossible, to reliably estimate train to floor ratio either visually or manually. Even if you're an expert. Alright, so what were the three claims that we looked at?

Right, so the first one was does the site of monitoring matter or not? The second one was that uh if it's been a while, I'm okay to just uh reverse or extubate. And the third one was if I don't have a quantitative trainer for monitoring Tactually assessing or visually assessing is sufficient. We only looked at one or two papers for each of those claims. And to our listeners out there, as always, take a second and think about your level of agreement with those claims.

And what do you think is the level of evidence supporting those Alright, Dan. Alright, I'm excited to record the next episode. Would it be cool to give them a little uh sneak peek what we'll be talking about? Yeah, sounds good. Alright, so most of the next episode will be looking at does it really matter?

You know, is there a train of four ratio that we really have to push for? Um or the claim I guess that It does matter or it doesn't matter. So if you kinda want to have some fun with it, see if you can answer that claim right now, or at least your belief in that claim. And if you like, before listening to our next episode, do a quick uh lit search, see what you can find, see if you can substantiate your beliefs, um, see if you find the same papers that we found, or different ones. Yeah, I think that's an awesome idea.

I think that that'd be a lot of fun. I think it sounds like fun. As always, you can reach us at depthofanesthesia at gmail. com or on Twitter at depthanesthesia. And I think what we're going to do for this episode, because it's in two parts, is we're going to conduct a Twitter poll. I haven't yet decided what exactly we're going to ask, but it's going to be related to this two part series.

And what I'll do is during the next episode or the next part We'll talk about the results of the Twitter poll and compare and contrast that with a survey of current management of neuromuscular block in the United States and Europe. which is a study published in Anesthesia and Analgesia uh that actually reviews clinical practice and opinions amongst Anesthetists in the United States and Europe. To our listeners, thanks for joining us on the show this week and thanks for continuing to think critically about your clinical practices. Stay hungry and keep asking questions.