What is the approach to post-extubation stridor in COVID-19?

Welcome back to Depth of Anesthesia. My name is David Hao and I'm an anesthesia resident at the Massachusetts General Hospital. This episode will be featured on Anesthesiology News. You can find them at anesthesiologynews.com If you find the podcast educational, please be sure to subscribe, rate, and share this podcast with your friends and colleagues. My guest today is Dr. Tiara Calhoun.

Dr. Calhoun is a resident in the Department of Medicine here at MGH, and she's an active contributor to the MGH Flair newsletter. Listeners of our show might remember that our last episode on ARDS and COVID-19 was based on a flared newsletter by Dr. Corey Hardin. Flare is a newsletter that critically appraises some of the rapidly evolving literature on COVID-19. Dr. Calhoun, welcome to the show. Thanks so much for having me here, David.

I'm really excited to be here So we'll start with our case today, which is that of a 55 year old man with a history of hypertension and diabetes who's been intubated in the ICU for 17 days with COVID-19. As the team is preparing to extipate, The resident mentions concerns about post extubation complications like strider in this population. The team agrees and proceeds to perform a leak test and then administer a dose of steroids Dr. Calhoun, what are some of the claims in this case? So the resident here seems to be invoking COVID-19 as or patients with COVID-19 as a population who may have a higher incidence of post-extubation strider than the general population. Secondly, they are claiming that the cuff leak test may predict development of post-extubation strider.

And third, they think that steroids may reduce the um incidence of post-extirpation strider. To our listeners, what is the level of your agreement with the claims laid out by Dr. Calhoun? And what is the level of evidence for what you believe? Dr. Calhoun, let's start by talking a bit about what post-extubation strider is. So uh strider is a high pitched wheezing sound that occurs during inspiration.

It's thought to be caused by turbulent airflow in the upper airway and is often invoked to indicate upper airway narrowing or obstruction. In ICU patients with strider after extirbation, laryngeal edema due to trauma from the endotracheal tube is usually invoked as the cause. Should we be using larynguledema and strider synonymously? So to clarify uh laryngula edema versus strider Some degree of laryngeal edema is thought to occur in just about all intubated patients. You know, ET tubes are traumatic to the upper airway and some swelling should be expected.

Strider, on the other hand, is used as a sign of edema that's significant enough to cause some respiratory distress And I think the reason this matters is that ringuledema and strider may lead to respiratory failure and reintubation. That relationship is certainly a claim in and of itself. But luckily I think the studies often evaluate those entities separately. Speaking of the studies, do we have any literature on the incidence of post-extubation strider pre-COVID-19? The data in this space is incredibly poor.

The reported incidence of edema, strider, and re-intubation due to complications of these varies dramatically from paper to paper. And that's probably because there really isn't a commonly accepted definition or mode for assessing peri -intubation laryngeal edema or strider. Some papers try to define very specific sounds or focus on subjective work of breathing and respiratory rate And others have used things like fiber optic video to directly visualize airway diameter. The best data we do have comes from a review by Witakamp and colleagues, first published in 2009, I believe, and updated by some other members of that research group in 2015. They found that anywhere from 5 to 55% of patients developed laryngeal edema post-extubation, and then 1.5 to 26% developed strider.

So really, really wide ranges and even wider ranges of of patients who developed those complications required reintubation. They found, depending on the study, that anywhere from 10 to 100% had to be re-intubated. That said in the the study that quoted 100%, there was only eight patients, so it was a very small sample size. And as you mentioned before, it seems like the variability is due to lack of standardization about not only methods for assessing edema and strider, but also inconsistency with definitions. That is exactly right.

So the none of these these papers used one common definition to assess either laryngeoledema or strider. What are some of the risk factors reported for laryngeal edema and strider? Risk factors for this is are basically anything that can contribute to airway trauma. So longer durations of intubation, high cuff pressures, larger tube size, history of a difficult or traumatic intubation, or histories of self-extubation Lots of uh authors invoke female gender, but some think that that's primarily because female airways are smaller and often uh a correspondingly smaller ET tube isn't used And high BMI is invoked, but high BMI means more soft tissue pressure on the airway. Is there anything different about COVID-19 patients?

For COVID, particularly, there are a couple of things that may contribute to increased risk of laryngeal injury. And I think that these are prolonged duration of intubation and use of prone ventilation. So starting with prolonged durations, most centers haven't published their average durations yet, and I think it'll be a really long time before we know exactly what the averages in this cohort are But in the couple of published cohorts to date, the average duration were nine and ten days. In our center, I'm frequently seeing two or more weeks of of intubation, which is certainly more than than we see in in other critically ill patients for other reasons. Regarding proning We're using a lot of it in COVID patients , and prone ventilation sounds intuitively more traumatic to the airway than supine.

Especially when done on a traditional ICU bed meant for the supine position. You know, we distort the neck at odd angles, put traction on the airway. And this impairs the venous drainage of the neck and contributes to more compressive trauma. And unfortunately, the major trials of proning and ARDS, like Presiva, just don't address this There's a little bit of data in the spine surgery literature, including one prospective study that observed increased laryngeal edema in patients who were operated in the prone position compared to supine They used cufflake test, which we'll get into in a little bit, as well as videolaryngoscopy to assess this. But despite finding more edema, they found no evidence of airway obstruction in the prone group when they measured uh that using flow volume loops.

And even if there were better evidence in surgical populations , There are really big differences between the surgical population and the critically ill COVID-19 population that may make them difficult to compare to one another. Uh the first that comes to mind is is the duration is measured in hours in surgery and days to weeks in critical illness. For another example, in spine surgery, people get large volumes of IV fluids that may contribute to acute edema around the face and airway Whereas uh in COVID-19 patients and all patients with ARDS, we use conservative fluid management strategies. So it's really unclear how much, if any, that literature bears on COVID-19 patients. I think you make a few really good points.

We still don't have great data on accurate median duration of ICU stays. The Griselli paper, for example, which you mentioned reporting a median of nine days, at the time of their publication, more than fifty percent of their patients were still in the ICU and presumably still intubated. That's exactly right. And you know, with these few reports out to date, we we just we barely know what the incidence of post-extuation strider is in the large body of patients who have experienced um prolonged mechanical ventilation, it's really impossible for us to say at this point what the incidence is in the COVID-19 group you know, lots of folks are circulating anecdotal reports via word of mouth and social media that they're seeing this more often than they have in the past.

But I I certainly don't feel like I'm in a place to make an assessment and I don't think they should feel that way either. And one interesting tidbit about that prospective study you mentioned When you start anesthesia residency at MGH, you're assigned two tutors who teach you the basics of anesthesia over the course of a month. That first author, Dr. Michelle Zabot, was actually my second tutor. Oh, that's so funny. And that that is the only trial that I was able to find in the literature that that compared prone to supine ventilation and laryngeal edema.

So there's there's certainly a a a space to be filled for for future research Moving on to our second claim, which is that a cuff leak test is an effective predictor of post-extubation strider. What exactly is a cuff leak test? Sure, so a cuff leak test essentially checks for air movement around the endotracheal tube while the cuff is deflated. So this can be measured either qualitatively or quantitatively. In the qualitative couff leak test, the practitioner deflates the tube and just listens for audible leak.

If they hear it, it's there, and they feel reassured that there is a small enough amount of edema that the airway will be patent after the tube is removed. In the quantitative approach, patients are placed on volume control ventilation, and the difference in their expired title volume when the cuff is inflated versus deflated is calculated. So an adequate cuff leak volume is not standardized, but lots of providers use 110 mls as a general rough cutoff of being adequate cuff leak. And this comes from a a study from 1996 that first started to evaluate this test. Do we have any evidence that a cough leak test is effective as a predictor for either strider, edema, or reintubation?

That's a great question. Um, because in my very limited experience as a resident in the ICU, we always check a cuff leak prior to extivating an ICU patient. When what I've been able to find, there was a 2008 meta-analysis by Ochoa and colleagues that analyzed nine studies that all assess the diagnostic accuracy of CoughLeak They used pretty different methods to assess laryngeoledema or strider post-extubation. But in the end, they found a pooled sensitivity of only about 56 % and a somewhat better specificity at 92% And I also noticed in that systematic review and meta-analysis, only three studies looked at the accuracy of the cuff leak tests for reintubation. The pooled positive likelihood ratio was about four, and the negative likelihood ratio was 0.46 When you frame this in the context of pretest probability, if you assume your pretest probability of reintubation starts at 1%. then even with an absent cuff leak test, which would suggest edema, a positive likelihood ratio of four would only bring the post-test probability to four percent And that's well within the acceptable extubation failure rate that's commonly cited in the ICUs of 15% Yeah, that's exactly right.

The cuff leak may not move the mean the needle much if you already have a a reasonable pretest probability of not needing reintubation. Are there any other prediction strategies and is there any evidence for their efficacy? So in my literature few uh review I found a couple None with nearly as much data as CuffLeak, just to to preface that. Uh the first was Loryngia Ultrasound. So this has been studied a handful of times.

It was in like four or five different papers. And this method, they use the width of an air column in the trachea. that's measured both before and after the cuff is deflated. It's hypothesized that a smaller width difference may portend a higher risk of edema in strider. But this method is pretty limited evidence and is probably best used uh in the hands of a very experienced operator, because I imagine there's a lot of um interoperator dependence and in how how well that width difference is is measured. The second method I was able to find was video laryngoscopy So this is exactly what it sounds like.

You directly visualize the trachea to see if edema is present and to look at the diameter of the airway. This method has not been systemically evaluated anywhere I can find. And so I it's I actually don't feel comfortable commenting on whether this might be a good method or not. Yeah, I I can't even begin to imagine how tough it would be to create a study that relied on something so subjective and completely visual. We do do it sometimes in the SIQ and the vast majority of the time when I take a look it's either eh looks fine or oh that looks pretty bad and that's really the level of fidelity we get with that And onto our third claim, do we have any evidence that steroids work for minimizing post-extubation strider, edema, or reintubation?

So steroids before extubation is by far the best studied intervention to treat laryngeal edema and prevent strider after extubation. And by way of in a historical context, there were some earlier studies, um, like in in the very early nineties, that gave Then later into the two thousand In 2006, Chang'an colleagues compared a single dose of 40 milligrams of methylpread methylpredicolone 24 hours prior to extubation. against control and also compared four doses of methylloprenosolone spaced out in the 24 hours prior to extubation So at 24, at 18, at 12, and at 6 hours against a control group. They found that the repeated dosing mechanism reduced strider in the treated group compared to the control. And the the single dose at 24 hours prior trended in the right direction but didn't reach significance.

That same group did another study a couple of years later in 2011 where they compared a single dose of 40 milligrams of methylpread four hours prior to extubation to control, and this time also achieved a significant result. So I think that set of studies by Chengen colleagues suggests that Steroids may be effective at preventing laryngeal edema. A longer course probably isn't necessary, and 24 hours ahead of time is probably too long. One thing to note , in all of these studies, Cheng and colleagues identified and included only at-risk patients who had been intubated for 24 hours or more and who had low cuff leak. I think they used 24 % or less in their studies.

Something a little bit more interesting. So another trial around the same time as the first um set of Cheng trials, randomized patients to receive three doses of methylpread over the 12 hours prior to excavation or placebo without assessing cuff leak at all. So all you had to do was be intubated for 36 hours or more And there was no other pre-enrollment assessment of risk for laryngeal edema. And this group still found a benefit in the treated group compared to the control group. I think this is a pretty cool finding because it suggests we may be able to give prophylactic steroids to all comers who are intubated for a couple of days or more and still see a benefit.

That is, in the absence of other risk factors, which we'll get to in a minute. One other paper of interest that I'll mention is one by Abbasi and colleagues in 2014 in which they randomized patients who had been intubated for 48 hours or more to receive uh inhaled butesinide or placebo in four doses over the 48 hours after extubation. So this requires no pre-extubation treatment but it was all done after the fact. They also found a decreased incidence of respiratory distress and of reintubation in the treated group compared to placebo And this trial again didn't require any assessment of cuff leak or assessment of risk for post-extubation strider other than the duration of intubation So it was maybe some more evidence that all comers could benefit.

What is your overall assessment of the evidence that you presented? On balance, it seems that the body of evidence suggests that anticipatory steroids could be effective in reducing post-extubation, laryngeal edema, and strider. among adult patients who have been intubated for prolonged periods of time. In the papers I've mentioned here, it was anywhere from 24 to 48 hours at least. I do wonder how this potential benefit will measure up specifically in viral pneumonias such as COVID-19, since steroids have been associated with harm in other viral pneumonias. I think that's a really important point.

All this data we're reviewing here today is truly extrapolated to this population. The last thing I do want to touch on is the Cochrane review, corticosteroids for the prevention and treatment of post-extubation strider, and I know you purposely left that out of your flare newsletter. Yeah, um so I I did come across the the Cochrane review from 2008 on uh post-extubation, Laringula edema and Strider. The reason I haven't discussed it, I didn't discuss it in my flare and I haven't discussed it here is that it lumps together. neonatal, pediatric, and adult data, and only included five studies in adult patients. Several of the adult studies occurred after that review was published and Uh as an adult doctor, I I worry about uh including children, babies, and adults together because the mechanism and management of laryngeal edema in children may be very different from adults.

So I've I've chosen to to not evaluate that piece, although I recognize it's there and it it's useful in context That makes a lot of sense, and I'll be sure to include that review in our show notes. There is a subgroup analysis for adults in that paper that might be worthwhile for our listeners to take a look at. Dr. Calhoun, what are some of your key takeaways from our discussion today? So I think in summary, uh critically ill COVID-19 patients. have a lot of risk factors for larynguladema, and those risk factors are fully extrapolated from non-COVID-19 populations. I think the most important risk factor is probably prolonged mechanical intubation.

Proning could be a risk factor, but we don't have enough information to say so yet. Secondly, at this point, we really just need more data to understand whether post-extubation strider and larynguledema are of increased risk in the COVID-19 population. Lots of anecdotal reports abound , but it's just not yet clear. So and and and lastly, the available data does suggest that glucocorticoids given prior to extubation may be helpful in preventing reintubation and post-extubation laryngula daemon strider. But the the the timing for this and whether it is likely to be helpful in the COVID population is still up for debate Excellent.

And so for our listeners, the three claims we started with were as follows. Patients with COVID-19 have a higher incidence of post-extubation strider The cuff leak test is effective at predicting post-extubation strider, and steroids are beneficial in reducing post-extubation strider. So to our listeners again, what is your level of agreement with the claims? And what is the level of evidence for what you believe? Dr. Calhoun, thanks for joining us on the show today.

I think this is really timely information. At least in Boston we seem to have hit the plateau, and so Nowadays we're doing more extubating than intubating, and I think a lot of this this information and evidence will go a long way in Allowing us to better take care of our patients. I'm so glad to be of help and thank you very much for having me. Thanks for joining me on the show today. I continue to be sincerely appreciative for all the support I've gotten from you, the listeners.

As a way of supporting you, I want to share with you some information about the Emotional PPE project This is a project that connects healthcare workers in need with licensed mental health professionals who are available to help. The service is completely free, no insurance involved, just a trained professional to talk to. You can reach him at emotionalpe. org or emotional ppe at gmail. com. I highly recommend it. Thanks again for tuning into the show and as always stay hungry and keep asking questions.

But the waste of three smallers , it's a waste of three split, it's a massive, it's a waste of three , it's a massive thing.