Abandon Etomidate for RSI in Suspected Sepsis...

[mward04]

---

Members don't see this ad.

Anyone see the new RCT on Hydrocortisone in Septic Shock
in NEJM today?
http://content.nejm.org/cgi/content/short/358/2/111

I have been reviewing this data for a talk next month when I saw this article today...how timely 😀 Everything I had reviewed was retrospective data basically saying that a single dose of etomidate still caused adrenal insufficiency (the reason it is no longer used as an extended sedative) and increased mortality. I was already going to say that we should seriously reconsider Etomidate in RSI for patients with suspected sepsis. With this new 499 patient RCT, I think we should abandon etomidate in suspected sepsis.

Let the discussion flow...

 

[southerndoc]

Anyone see the new RCT on Hydrocortisone in Septic Shock
in NEJM today?
http://content.nejm.org/cgi/content/short/358/2/111

I have been reviewing this data for a talk next month when I saw this article today...how timely 😀 Everything I had reviewed was retrospective data basically saying that a single dose of etomidate still caused adrenal insufficiency (the reason it is no longer used as an extended sedative) and increased mortality. I was already going to say that we should seriously reconsider Etomidate in RSI for patients with suspected sepsis. With this new 499 patient RCT, I think we should abandon etomidate in suspected sepsis.

Let the discussion flow...

Does this article specifically look at mortality with etomidate use in sepsis? The article title suggests it's studying hydrocortisone.

(Disclaimer: I haven't read the article yet because I haven't switched over to the med school proxy yet to download the article.)

 

[bartleby]

I do not think that there is any sort of consensus or solid evidence for abandoning the use of etomidate across the board in sepsis for use in RSI. Not all people who are septic are adrenally insufficient...pressor resistent hypotension (which, from the academic sense, is itself an overly broad definition) is present in a relatively small proportion of people with sepsis. Might it be reasonable to pick another agent if you have someone in which you specifically worry about adrenal suppression (i.e. chronic steroids who has not yet been administered a stress dose of steroids)? Sure. But I wouldn't get the hammer and the nails out just yet.

 

[mward04]

Does this article specifically look at mortality with etomidate use in sepsis? The article title suggests it's studying hydrocortisone.

(Disclaimer: I haven't read the article yet because I haven't switched over to the med school proxy yet to download the article.)

The purpose of this study was to look at the effects of hydrocortisone on adrenal insufficiency but they did a subgroup analysis on the etomidate patients which showed that 60% of pt's given etomidate were adrenally insufficient vs. 43% of pt's not given etomidate. Additionally, the mortality of etomidate patients was 40-45% vs. 29-31% for non-etomidate patients.

This study also offers a couple of other key points:
1. Hydrocortisone does not treat adrenal insufficiency and vasopressor resistance
2. Those treated with Hydrocortisone actually had a higher mortality rate, although not statistically significant secondary to superinfection and recurrent sepsis

Additionally the following studies deal with etomidate & mortality:
1. Mohammad et al. 2006 showed that non-etomidate vs. Etomidate increased adrenal insufficiency from 51 to 76% (respectively) and mortality from 55 to 63% (respectively) in septic shock
2. Brinker et al. 2007 showed in septic peds that 95% of peds given etomidate had significantly elevated 11-deoxycortisol (because etomidate blocks 11-B Hydroxylase and conversion of 11-deoxycortisol into cortisol) vs. 65% in non-etomidate patients with mortality of 12% in non-etomidate and 30% with etomidate

 

[southerndoc]

mward, that's what I thought: it was a subgroup analysis, which doesn't really qualify as a great RCT study.

What we need is a specific study to examine etomidate in sepsis. I am aware of data from previous studies that suggests etomidate increases mortality in sepsis, but as of yet there is not conclusive data.

As bartleby has mentioned, sepsis patients that are adrenally insufficient are not a large percentage of septic patients. Most critical care physicians have moved away from giving steroids to every septic patient, and instead, only those that are pressor-dependent are given dexamethasone and those that are proven adrenally insufficient are given fludrocortisone and hydrocortisone.

 

[Haemr Head]

The concern about adrenal suppresion with one dose of etomidate is valid. However the data is still not compelling enough to stop using it without a better alternative. Barbituates and propofol are clearly contraindicated in hypotensive patients. Even with the adrenal risk, I would still choose etomidate over a benzodiazepine for RSI both for hemodynamic stability and success of the procedure. One rarely has the time to be able to use fentanyl effectively and narcotics supress sympathetic drive. So if not etomidate, then what induction agent would you use? Brutane?

 

[Hard24Get]

The concern about adrenal suppresion with one dose of etomidate is valid. However the data is still not compelling enough to stop using it without a better alternative. Barbituates and propofol are clearly contraindicated in hypotensive patients. Even with the adrenal risk, I would still choose etomidate over a benzodiazepine for RSI both for hemodynamic stability and success of the procedure. One rarely has the time to be able to use fentanyl effectively and narcotics supress sympathetic drive. So if not etomidate, then what induction agent would you use? Brutane?

Interestingly, this came up last night at BIDMC JC. I think the biggest problem with most of the studies so far is that etomidate was pretty much the induction agent for intubation as oppose to being compared with a different agent for just the reason Haemr Head cites. Those that need to be intubated are likely sicker, opening the door for a big confounder unless compare to those intubated without etomidate.

 

[docB]

You can successfully use versed for induction but you really need to dose it and wait several minutes before paralyzing or you're risking that the patient might be paralyzed and aware which is not cool. It's relatively rare that I have many extra minutes to play with in the patients I'm emergently intubating. That brings up my main concern with this issue. When we get patients in the ED who require emergent intubation my suspicion of sepsis has usually taken a back seat to the ABCs. I often don't have a temp, never have blood work back, no CXR, maybe an EKG. Granted I'll have a high index of suspecion on little old ladies with chronic foleys sent from the nursing home for hypotension but early on I'm probably not suspecting anything other than "sick."

 

[deleted65604]

Here's an abstract from a pilot study that tried to get at this question. It's not conclusive, but it leans toward etomidate being fine in sepsis

Available online http://ccforum.com/content/11/3/R56

Effect of induction agent on vasopressor and steroid use, and outcome in patients with septic shock

David Charles Ray and Dermot William McKeown
Department of Anaesthesia, Critical Care & Pain Medicine, Royal Infirmary of Edinburgh, Little France Crescent, Edinburgh EH16 4SA, Scotland, UK
Corresponding author: David Charles Ray, [email protected]

Critical Care 2007, 11: 3 (R56) (doi:10.1186/cc5916)
This article is online at: http://ccforum.com/content/11/3/R56
© 2007 Ray and McKeown; licensee BioMed Central Ltd.

Abstract
Introduction In seriously ill patients, etomidate gives
cardiovascular stability at induction of anaesthesia, but there is
concern over possible adrenal suppression. Etomidate could
reduce steroid synthesis and increase the need for vasopressor
and steroid therapy. The outcome could be worse than in
patients given other induction agents.
Methods We reviewed 159 septic shock patients admitted to
our intensive care unit (ICU) over a 40-month period to study the
association between induction agent and clinical outcome,
including vasopressor, inotrope, and steroid therapy. From our
records, we retrieved induction agent use; vasopressor
administration at induction; vasopressor, inotrope, and steroid
administration in the ICU; and hospital outcome.
Results Hospital mortality was 65%. The numbers of patients
given an induction agent were 74, etomidate; 25, propofol; 26,
thiopental; 18, other agent; and 16, no agent. Vasopressor,
inotrope, or steroid administration and outcome were not related
to the induction agent chosen. Corticosteroid therapy given to
patients who received etomidate did not affect outcome.
Vasopressor therapy was required less frequently and in smaller
doses when etomidate was used to induce anaesthesia. We
found no evidence that either clinical outcome or therapy was
affected when etomidate was used. Etomidate caused less
cardiovascular depression than other induction agents in
patients with septic shock.
Conclusion Etomidate use for critically ill patients should
consider all of these issues and not simply the possibility of
adrenal suppression, which may not be important when steroid
supplements are used.

 

[NinerNiner999]

The concern about adrenal suppresion with one dose of etomidate is valid. However the data is still not compelling enough to stop using it without a better alternative. Barbituates and propofol are clearly contraindicated in hypotensive patients. Even with the adrenal risk, I would still choose etomidate over a benzodiazepine for RSI both for hemodynamic stability and success of the procedure. One rarely has the time to be able to use fentanyl effectively and narcotics supress sympathetic drive. So if not etomidate, then what induction agent would you use? Brutane?

Ketamine?

 

[Haemr Head]

Ketamine?

That's actually a very interesting idea. There are a number of theoretical problems with ketamine (just like etomidate). One of the biggest is increased myocardial oxygen demand that might increase the risk of demand myocardial ischemia (a good number of the septic patients are elderly). Since we are starting to worry that some of the pressors actually worsen mortality in severe sepsis, ketamine which increases the oxygen debt substantially might be very detrimental. There is also the problems of increased secretion that can make the airway more difficult and I wouldn't be eager to toss atropine at a patient in septic shock. However, all of those points are as theoretical as the adrenal supression that worries us with etomidate. Ketamine may not induce patients quite as fast or reliably as etomidate or propofol, but it is real inductor unlike the benzos. We are firmly convinced in the US that it is absoultely contraindicated in patients with elevated ICP but the Europeens happily use it in this setting and say it is neuroprotective. I'm not sure how easy it would be to slide this one past an IRB but a randomized trial etomidate vs. ketamine or a cohort study on outcome in patients intubated with Ketamine might be very worthwhile. I think I'll go annoy some people who really understand this stuff later today. For all we know, it might end up giving a better result than all of goal directed therapy put together.

 

[VentdependenT]

Use what you are comfortable with.

If the patient is septic its highly unlikely that your choice of induction agent is going to have any impact on their outcome. They have bigger issues at hand.

I, personally, have avoided the use of etomidate for the past year or so. Reasons: makes people puke (not an issue here), myoclonic activity (again not an issue here), burns (big deal in this situation), expensive (what isn't these days), and most importantly the suppression of the adrenal glands (this could be a big problem).

All induction agents are basically equal when given in the proper amount (except perhaps ketamine).

The most important thing here is that you are comfortable with your induction and RSI. Get the tube in, get the CO2 down and the oxygen up, and get em on to the unit tanked up, abx runnen, and on a pressor if need be.

Thats my take, for what its worth.

 

[Dimoak]

I'm not an expert, but Ketamine also runs the risk of laryngospasm, which would suck during RSI.

 

[VentdependenT]

I'm not an expert, but Ketamine also runs the risk of laryngospasm, which would suck during RSI.

Laryngospasm has to do with the depth of anesthesia.

A light plane of anesthesia, even without any sort of airway manipulation, can lead to laryngospasm. Thats what sux is for though brotha.

 

[docB]

Ketamine also causes a rise in intracerebral and intraocular pressures so it's not appropriate in many cases such as ICB or multitrauma with an ocular component.

In my experience etomidate has settled into its role not because it's terrific butr because it is the best we have available. Lots of docs have been using it for procedural sedation although it's being eclipsed by propofol. We didn't adopt it because we loved it. We did it because we lost brevitol.

 

[Haemr Head]

I'm not an expert, but Ketamine also runs the risk of laryngospasm, which would suck during RSI.

No need to worry about laryngospasm with succinylcholine on board. That's only a problem with oral "awake" intubation and conscious sedation with Ketamine.

Go with what you know in airway management is a very good recommendation in terms of choice of approach (type of blade, nasal vs. oral, surgical cric vs. percutaneous cric) but in RSI you want an inductor that will take them down as fast as possible with as little consequence as possible. The only thing you need to know is the dosage and the contraindications.

In terms of all inductors being equal, I will respectively disagree in terms of vasodilatation, myocardial supression, ICP, speed of onset, and cortisol supression. Now one can argue that none of those really matter as much as just placing the tube but since we don't measure the outcomes 4 days later when we slam a tube in and ship them upstairs, it doesn't mean that there isn't a price to pay for a less adeqaute choice. I do agree that it is hard to measure or study, but plenty of good data of irreversible shock induced by thiopental in borderline shock patients. Go digging and you will find the same for propofol and midazolam. Think back to the time spent in the OR and ask yourself how many RSIs did you see with midazolam and ask yourself if it is equivalent, why aren't the anesthesiologists using this approach? And say what you will, they are involved with the consequences of airway management, have done the lion share of the research on inductors, and try to apply it as well as they can in their practice.

Not using etomidate because there is a slight drop in serum cortisol and because patients in the ICU on drips for days may do worse, hardly makes me feel that barbituates, propofol, or benzos are equivalent or preferable to a single shot of etomidate in a patient in shock. The difference in cost relative to the frequency of use, the overall cost of care of these patients, and the cost of dealing with more adverse events from airway management, really make this moot. If we aren't quite sure of the frequency and severity of the down stream effects, there is no way to measure what is cost effective and what is not. I would bet that most places not using etomidate, do so, because the Anesthesia Department managed to block the use of the drug in the ED, NOT because of cost or an evidenced based approach to the relative risk of each agent. I would also bet that patients in shock at these institutions are intubating using a fraction of the induction dose of midazolam, so basically Brutane.