Etomidate back in the limelight

[epidural man]

We have had some great etomidate discussion on here previously. I still can't believe people use it.

Check out the latest (see attached)....

And I disagree with the authors that etomidate gives a more stable induction...that is pure BS.

---

Members don't see this ad.

 

[IkeBoy18]

very little experience talking here, but I read a few of these studies a few months ago. At a community hospital I rotated, they use etomidate for intubation during codes. I understand that etomidate has been implicated in cardiac issues later on and in the midst of a code its not a current concern obviously. Anyone use anything other than etomidate during codes for intubation?

 

[Guy Caballero]

very little experience talking here, but I read a few of these studies a few months ago. At a community hospital I rotated, they use etomidate for intubation during codes. I understand that etomidate has been implicated in cardiac issues later on and in the midst of a code its not a current concern obviously. Anyone use anything other than etomidate during codes for intubation?

I rarely use any medication while intubating during a code. It is rarely indicated.

 

[Scumlee13]

Article is retrospective. Propensity matching proves nothing. Sicker patients get etomidate, even propensity matched. Everyone knows that propofol induction in a critically ill patients requires substantially more vasopressor use, and the fact they found no difference indicates an unaccoutned difference in illness between their propofol analysis group and their etomidate group. Prospective trials are the only way to do this study properly.

 

[IkeBoy18]

I rarely use any medication while intubating during a code. It is rarely indicated.

Gotcha. Thanks. When I was on IM it seemed to be within whatever algorithm the IM residents used as each code they always pushed it.

 

[Guy Caballero]

Gotcha. Thanks. When I was on IM it seemed to be within whatever algorithm the IM residents used as each code they always pushed it.

If it is truly a "code," the patient is usually not breathing, unresponsive, +/- a pulse or BP. Chest compressions are frequently occurring. No reason to give drugs to a patient who will not fight you or recall the event (in most circumstances).

 

[IkeBoy18]

If it is truly a "code," the patient is usually not breathing, unresponsive, +/- a pulse or BP. Chest compressions are frequently occurring. No reason to give drugs to a patient who will not fight you or recall the event (in most circumstances).

I agree. Im not sure sure why they did. I had no clue what the drug was used for during my first few codes the first week of IM, I had assumed it was some sort of antiarrhythmic or pressor since it was used everytime; of course until I looked it up.

 

[deleted126335]

I used to use a lot of it. Now I only use it for known or strongly suspected increased intracranial pressure with precarious hemodynamic status.

 

[OB1🤙]

I use it exactly never.

I'm a cardiac anesthesiologist.

 

[Sublimazing]

I just used it today for a propofol "allergy" pt...even at the upper end of recommended dosing I never feel it is a smooth induction agent...there always seems to be unsettling myoclonus until the roc kicks in

 

[chmd]

Duplicate post

 

[chmd]

I use it exactly never.

I'm a cardiac anesthesiologist.

So what do you use? Just midazolam and fentanyl +\- ketamine. That's what we did for cardiac patients when etomidate was on shortage.

2 problems with that: 1) do large induction doses of fentanyl and midazolam prevent fast tracking? 2) do you feel comfortable giving AS patients enough ketamine to induce anesthesia?

We rarely even gave 2 mg/kg and I think it worked well because pts also got 5-10 mg of midazolam and 500-1000 of fentanyl.

 

[OB1🤙]

To answer that I first have to say that I have always found the idea of a "cardiac induction" involving high dose midaz/versed to be a vestige of an era that existed prior to short-acting cardioactive drugs.

Yes, you want to avoid tachycardia in CAD. There are other, better ways of accomplishing rate control than massive narcotic doses IMO. For example, esmolol.

I use benzos/opiates as I would in any other major case. Most of my hearts get 1-2.5 mg midaz for the a-line, and 500mcg fentanyl total for the case.

To induce a sick patient- I don't have a standard induction, I try to tailor each induction to the patient's particular ideal hemodynamic state. But some combination of 100-150mcg fent, .25-1 mg/kg ketamine, .5-2 mg/kg propofol usually does the trick. If I'm giving narcotic, I'll give that a few minutes up front. As we get going, ketamine goes in first, followed by NMBD (I don't test for maskability, but that's another thread), then judicious propofol to effect.

I don't use fentanyl if I want a higher heart rate (i.e. regurgitant valvular disease). I don't use ketamine if tachycardia poses a particularly bad threat, though doses around .5mg/kg don't cause much tachycardia. For most AS patients, ketamine + propofol does just fine.

For the truly, truly sick? They get a dash of midaz and then breathe sevo. And they go down nice and stable. Nothing more stable than an inhaled induction.

There are many ways to skin the cardiac cat without etomidate.

 

[chocomorsel]

To answer that I first have to say that I have always found the idea of a "cardiac induction" involving high dose midaz/versed to be a vestige of an era that existed prior to short-acting cardioactive drugs.

Yes, you want to avoid tachycardia in CAD. There are other, better ways of accomplishing rate control than massive narcotic doses IMO. For example, esmolol.

I use benzos/opiates as I would in any other major case. Most of my hearts get 1-2.5 mg midaz for the a-line, and 500mcg fentanyl total for the case.

To induce a sick patient- I don't have a standard induction, I try to tailor each induction to the patient's particular ideal hemodynamic state. But some combination of 100-150mcg fent, .25-1 mg/kg ketamine, .5-2 mg/kg propofol usually does the trick. If I'm giving narcotic, I'll give that a few minutes up front. As we get going, ketamine goes in first, followed by NMBD (I don't test for maskability, but that's another thread), then judicious propofol to effect.

I don't use fentanyl if I want a higher heart rate (i.e. regurgitant valvular disease). I don't use ketamine if tachycardia poses a particularly bad threat, though doses around .5mg/kg don't cause much tachycardia. For most AS patients, ketamine + propofol does just fine.

For the truly, truly sick? They get a dash of midaz and then breathe sevo. And they go down nice and stable. Nothing more stable than an inhaled induction.

There are many ways to skin the cardiac cat without etomidate.

nice post

 

[chocomorsel]

Nice post. Don't do hearts but in residency almost all hearts got etomidate. Hardly use it now tho.

 

[Dejavu]

If it is truly a "code," the patient is usually not breathing, unresponsive, +/- a pulse or BP. Chest compressions are frequently occurring. No reason to give drugs to a patient who will not fight you or recall the event (in most circumstances).

I agree with this. Dead people don't fight you when you intubate them.

 

[chmd]

Thanks for the thoughtful reply, Bruin.

I'd love to use less fentanyl and midazolam in cardiac cases. I worry about the titratability of volatile agent in sick cardiac patients, ie would rather keep agent low to off and rely on my peri-induction midazolam and fentanyl if needed during the preincision lull.

Also in the truly sick patients would you see too much of a sympathetic response with masking/dl/tee using agent only (likely minimal agent at that). I suppose that's where esmolol comes into play.

 

[sevoflurane]

Vomidate? Puuuhhhlease. Might mix a tiny bit with a little propofol once or twice a year... usually in the emergent heart with an occluded left main + balloon pump + BPs in the toilet. Not only does it cause you to puke your brains out, but it also is a very poor agent at inducing apnea. In fact, not only does it not produce apnea but it can give you some nice my0clonus.

Sorry, not my cup of tea.

That being said, I have heard of a cardiac anesthesiologist getting drilled by a lawyer for not using it during a case that went bad.

 

[Planktonmd]

I think the Etomidate debate will never end... similar to the crystalloids versus colloids, or low versus high tidal volume ventilation and many other endless masturbatory debates!

 

[IlDestriero]

I think the Etomidate debate will never end... similar to the crystalloids versus colloids, or low versus high tidal volume ventilation and many other endless masturbatory debates!

What else is there to talk about at meetings? How else am I going to get promoted if not lecturing on these fascinating clinically relevant topics?

 

[nimbus]

I think the Etomidate debate will never end... similar to the crystalloids versus colloids, or low versus high tidal volume ventilation and many other endless masturbatory debates!

These endless debates reveal the factors under consideration have little if any effect on outcome. Some s*** just doesn't matter.

 

[Idiopathic]

I think the Etomidate debate will never end... similar to the crystalloids versus colloids, or low versus high tidal volume ventilation and many other endless masturbatory debates!

who is arguing for high tidal volume ventilation? and the data is clearing up pretty quickly regarding colloids vs crystalloids, IMHO.

 

[Shimmy8]

So I'm an (anesthesia) intern on a cards elective and have a presentation to do, with complete freedom as to the topic. Was trying to keep it anesthesia related, and was considering just presenting this paper vs older etomidate papers.

With that said, anyone have any suggestions as to practice changing papers in the last 12-24 months? I have a few emails out to attendings I've worked with as well, but figured this may provide some interesting feedback. Thanks.

And by the way, though I rarely post, I sincerely appreciate the residents and attendings who do contribute regularly. It's something I plan on doing more of once this wonderful year is over.

 

[periopdoc]

Inducing and intubating a cardiac patient for cardiac surgery is no different than inducing and intubating the same cardiac patient for any other surgery so why would I use anything other than my routine induction meds for a patient with cardiac disease?

I tailor my induction regimen to the patient's diseases, not the fact that they are having cardiac surgery and I deal with significantly less hemodynamic disturbance between intubation and incision than I did as a resident/fellow when I used the traditional cardiac induction regimens.

Appropriately dosed propofol is the superior agent in every respect. Just be sure you know what your goal is (amnesia, hemodynamic blunting, relaxation...). The appropriate dose may range from as little as 10-25mg to more than 200mg. If I think it will take >200mg to achieve whatever goal I am targeting, then that goal is probably better achieved with the administration of an additional drug (esmolol, midaz, sux...)

If I overshoot with the propofol and the patient gets hypotensive I typically treat with an inotrope rather than a pressor. I never understood increasing the afterload on a sick heart that I just stunned, but I am a simple minded man.

I don't give opiates to blunt hemodynamic effects of induction/intubation. Why give something relatively long acting for something that is so momentary. If you truly need additional hemodynamic blunting esmolol or NTG are great choices depending on your goal.

- pod

 

[canjosh]

So I'm an (anesthesia) intern on a cards elective and have a presentation to do, with complete freedom as to the topic. Was trying to keep it anesthesia related, and was considering just presenting this paper vs older etomidate papers.

With that said, anyone have any suggestions as to practice changing papers in the last 12-24 months? I have a few emails out to attendings I've worked with as well, but figured this may provide some interesting feedback. Thanks.

And by the way, though I rarely post, I sincerely appreciate the residents and attendings who do contribute regularly. It's something I plan on doing more of once this wonderful year is over.

Not sure if this meets all your requirements, but it is recent and involves cards at least. Good luck.

Risk of major adverse cardiac events following noncardiac surgery in patients with coronary stents.
Hawn MT, Graham LA, Richman JS et al.
JAMA. 2013 Oct 9;310(14):1462-72. doi: 10.1001/jama.2013.278787

 

[Guy Caballero]

who is arguing for high tidal volume ventilation? and the data is clearing up pretty quickly regarding colloids vs crystalloids, IMHO.

Wait 20 years. It will all come around again. Everything old will be new again. Every issue that you believe is solved will be re-examined. What was once considered malpractice will be the new standard of care. What was once standard of care will be malpractice. I have been told the usual cycle is about 20 years.

 

[Planktonmd]

Inducing and intubating a cardiac patient for cardiac surgery is no different than inducing and intubating the same cardiac patient for any other surgery so why would I use anything other than my routine induction meds for a patient with cardiac disease?

I tailor my induction regimen to the patient's diseases, not the fact that they are having cardiac surgery and I deal with significantly less hemodynamic disturbance between intubation and incision than I did as a resident/fellow when I used the traditional cardiac induction regimens.

Appropriately dosed propofol is the superior agent in every respect. Just be sure you know what your goal is (amnesia, hemodynamic blunting, relaxation...). The appropriate dose may range from as little as 10-25mg to more than 200mg. If I think it will take >200mg to achieve whatever goal I am targeting, then that goal is probably better achieved with the administration of an additional drug (esmolol, midaz, sux...)

If I overshoot with the propofol and the patient gets hypotensive I typically treat with an inotrope rather than a pressor. I never understood increasing the afterload on a sick heart that I just stunned, but I am a simple minded man.

I don't give opiates to blunt hemodynamic effects of induction/intubation. Why give something relatively long acting for something that is so momentary. If you truly need additional hemodynamic blunting esmolol or NTG are great choices depending on your goal.

- pod

When you say that you give an inotrope to treat Propofol induced hypotension in a patient with CAD which inotrope are you referring to?
It seems to me that you need an inotrope that's not a chronotrope as well because increasing the heart rate is probably not something you want to do in a patient with CAD severe enough to require surgery.

 

[BLADEMDA]

Today I had a patient with an EF of 20 percent. I induced with propofol 50 mg IV and some rocuronium. Gentle mask ventilation with Sevo at 1.2 percent for two minutes. BP was rock stable the entire time including post intubation BP.

The CRNA was insistent on using etomidate but like others on this board I know that careful induction with Propofol goes a long way in maintaining hemodynamics stability even in patients with cardiomyopathy. I try and avoid etomidate as much as possible but readily admit that the peer reviewed evidence against it is weak at best. (Yes, I read the latest study).

 

[epidural man]

Article is retrospective. Propensity matching proves nothing. Sicker patients get etomidate, even propensity matched. Everyone knows that propofol induction in a critically ill patients requires substantially more vasopressor use, and the fact they found no difference indicates an unaccoutned difference in illness between their propofol analysis group and their etomidate group. Prospective trials are the only way to do this study properly.

Propensity score is all the rage these days. Know why we don't use aprotinin any more? Propensity matching. Know why we do a bunch of other things in anesthesia now? Propensity matching. I'm not sure your conclusion that retrospective with propensity matching proves nothing is correct.

Also, there is a great article (I can't find it) that shows that hypotension on induction with etomidate EQUALS propofol. In fact, in that study, people got hypotensive almost as often with just sux.

 

[Sonny Crocket]

Have not used Etomidate since residency. And I haven't really missed it at all. I do remember thinking it was useful when intubating ICU patients because they would usually maintain spontaneous respiration.

 

[Planktonmd]

It is unfortunate that people still think that it is cool to follow what the rest of the herd does!
If there is a general consensus that Etomidate is evil this should make you think... is it just the cool thing to do these days to pretend that you hate Etomidate since the big dogs say they hate it???

 

[GaseousClay]

who is arguing for high tidal volume ventilation? and the data is clearing up pretty quickly regarding colloids vs crystalloids, IMHO.

Can you clarify what you mean about the data clearing up on colloid vs crystalloid? Im still somewhat confused on what the consensus is becoming on this

 

[Idiopathic]

i think there are only a few cases where you can argue that colloid solutions are better than balanced salt solutions, and almost all of them involve severe hypoalbuminemia. i like to use them and i like to think that they are useful when i do, but there is a large body of evidence that suggests our use of them provides no benefit except for a few occasions. they are in most cases equivalent to salt solutions yet cost more and carry a theoretical risk of allergy and infection.

overall we infuse too much fluid, and i think we need to guide our therapy better.

 

[deleted171991]

I think the main problem is that the statistics behind many of these studies are so murky that nobody should draw final conclusions from them. A lot of the prerequisite conditions for these methods are poorly satisfied; plus the methods seem not to be the most accurate themselves. This is how one gets to the point where many of the (p<0.05) "scientific" studies are refuted years later. It's worse than faith (at least the Bible stays the same).

(Please read the chapter about propensity score matching in the etomidate paper from the OP, and see if it doesn't feel like they played with numbers until they got the desired conclusion. It's just murky "science".)

To paraphrase Warren Buffett: "If you don't understand the statistics, don't buy the results." The so-called "expert consensus" is like bulls versus bears in the stock market. And as in the stock market, since no "analyst" gets punished for false conclusions, they keep spitting out these studies for the sake of the art, or their academic promotion.

To quote the authors of the same etomidate paper:

The potential putative link between etomidate and worsened postoperative outcomes has yet to be studied in a large cohort of high-risk general surgical patients.

So why are we even talking about this?

P.S. Of course if we eliminated the noise from medical "research" we would get 100 times fewer papers, companies like LWW would make 100 times fewer billions, and there would be much fewer academic promotions based on resume thickness.

 

[Idiopathic]

I disagree about your assessment. It sounds to me like for the propensity matching, they considered reasons why people might get etomidate, and then matched them with people who got propofol while considering comorbid conditions such as age, gender, ASA status, emergency, etc. using a score that was within +/- 0.2 SD (or within about 10% of the mean on either side) w/r/t propensity score, and then matched that to the type of surgery.

This statement from discussion is telling: "Before the propensity score matching (Table 1, left panel), patients given etomidate for induction were generally older and sicker (higher Charlson comorbidity score and higher ASA status). They were also more likely to be male, have a lower body mass index, have cardiovascular and/or cere- brovascular disease, have emergent surgery, and to receive general anesthesia supplemented with regional anesthesia." This suggests that a study done without matching for covariates would have significant selection bias.

The first argument someone would make about a study such as this if it were purely retrospective (i.e. ASA III/IV patients who received etomidate were more likely to die than people who received propofol) is that those patients MUST have been sicker or appeared sicker, which is why they got etomidate. So the point of this study is to reduce selection bias, which I think they did as well as they could.

Their data appears to be very strong, suggesting that receiving etomidate instead of propofol increases the risk of death among sicker patients by more than 100% with 98% power and a p value of <0.001.

Studies like this are unlikely to be done in a prospective, controlled fashion. The data generated here is much cleaner than that from a purely restrospective study. Its results are strong, and we would be remiss to ignore them.

As far as this

The potential putative link between etomidate and worsened postoperative outcomes has yet to be studied in a large cohort of high-risk general surgical patients.

thats in their introduction and it explains why they did the study

 

[deleted171991]

To me, etomidate seems to be guilty mostly by association with very sick patients. I rarely resort to it, if I don't have to, but that's because of the lasting inhibition of corticosteroid synthesis, which is clearly measurable. (And for anybody who knows Physiology 101, one needs those steroids to be able to mount a sympathetic response, especially those ICU-level patients. There is a reason we give steroids to some of the sicker patients.)

However, this study does not measure much, just plays statistical hocus-pocus with many (non-quantitative) variables. Did you notice the tons of statistical tests (read "approximations") they used on their way to their conclusions? Interestingly, exactly those measurable steroid levels were not taken into account.

I wish I had as many dollars as the number of useless multivariate scores and studies in medicine... The more variables, the more useless the scoring system. In my limited thinking, when you approximate and then approximate and then approximate some more, and you do this with tons of variables, there is no way in hell (or Heaven) that the probability of a type I error is less than 0.05. Because even a 1% approximation error becomes huge when applied to enough variables. (1+0.01) times (1+0.01) times (1+0.01)... = 1.10, after only 10 factors. An error of 10%, from just 1% initially! That's how worthless the multivariate studies are.

Now if these biostatisticians were insurance guys (actuaries), working for a big insurance company, I would be much more inclined to believe their conclusions.

 

[Idiopathic]

why do you think "mounting a sympathetic response" is the answer to survival? there are several reasons to give steroids to a sicker patient, more often than not its to reduce the inflammatory response, certainly this is not what the body intended.

you are telling me that you would rather there be no evidence to support what you do as opposed to a relatively strong retrospective dataset that says "keep up the good work"?

 

[deleted171991]

why do you think "mounting a sympathetic response" is the answer to survival? there are several reasons to give steroids to a sicker patient, more often than not its to reduce the inflammatory response, certainly this is not what the body intended.

More often than not it's for suspected (relative) adrenal insufficiency.

you are telling me that you would rather there be no evidence to support what you do as opposed to a relatively strong retrospective dataset that says "keep up the good work"?

There is nothing strong about it. When one applies statistical methods in a cascade, where the output of one method becomes the input of the next, the amount of error increases exponentially. In my very logical (but totally non-statistical) mind, a strong multivariate (medical) study is almost an oxymoron. (There is at least one entire book written about the bad biostatistics that dominate our studies.)

When about truly important things, like medical therapies, I'd rather have no science than bad science. I am the kind of guy who would say "I don't know, I haven't seen a study that has convinced me yet", if that's the case. Of course, I will keep even a bad study in the back of my mind (because most of them are bad anyway), but in a risks versus benefits analysis, the proven benefits might matter more than the theoretical risks.

But that's just me. I am pretty straightforward with my patients about what I know or believe, and what I don't. Like I don't believe CRNAs and anesthesiologists have the same outcomes, despite all the contrary retrospective "studies".

 

[deleted171991]

And there we go again:

Crit Care Med. 2013 Mar;41(3):774-83. doi: 10.1097/CCM.0b013e318274190d.
Single-dose etomidate is not associated with increased mortality in ICU patients with sepsis: analysis of a large electronic ICU database.
McPhee LC1, Badawi O, Fraser GL, Lerwick PA, Riker RR, Zuckerman IH, Franey C, Seder DB.
Author information
Abstract
OBJECTIVE:
Retrospective analyses of several trials suggest etomidate may be unsafe for intubation in patients with sepsis. We evaluated the association of etomidate and mortality in a large cohort of septic patients to determine if single-dose etomidate was associated with increased in-hospital mortality.

DESIGN AND SETTING:
Retrospective cohort study at the Philips eICU Research Institute ICU clinical database.

INTERVENTIONS:
None.

PATIENTS:
Among 741,036 patients monitored from 2008 through 2010, we identified 2,014 adults intubated in the ICU 4-96 hrs after admission, having clinical criteria consistent with sepsis, severe sepsis, or septic shock. In all, 1,102 patients received etomidate and 912 received other induction agents for intubation.

MEASUREMENTS AND MAIN RESULTS:
The primary endpoint was in-hospital mortality, but we also evaluated demographic and clinical factors, severity of illness, ICU mortality, ICU length of stay, hospital length of stay, ventilator days, and vasopressor days. Competing risk Cox proportional hazard regression models were used for primary outcomes. Demographics and illness severity were similar between the groups. Hospital mortality was similar between the groups (37.2% vs. 37.8%, p = 0.77), as were ICU mortality (30.1% vs. 30.2%, p = 0.99), ICU length of stay (8.7 days vs. 8.9 days, p = 0.66), and hospital length of stay (15.2 vs. 14.6 days, p = 0.31). More patients in the etomidate group received steroids before and after intubation (52.9% vs. 44.5%, p < 0.001), but vasopressor use and duration of mechanical ventilation were similar. No regression model showed an independent association of etomidate with mortality, shock, duration of mechanical ventilation, ICU or hospital length of stay, or vasopressor use. A hospital mortality model limited to only patients with septic shock (n = 650) also showed no association of etomidate and hospital mortality.

CONCLUSION:
In a mixed-diagnosis group of critically ill patients with sepsis, severe sepsis, and septic shock, single-dose etomidate administration for intubation in the ICU was not associated with higher mortality or other adverse clinical outcomes.

 

[Triple AAA]

Honestly, I feel like people take and interpret these studies to the extreme. The study is NOT saying that etomidate is 'evil' or that it should NEVER be used. It simply states that there are potential risks to long term outcomes, and that we should understand those risks and use it judiciously. Granted, I trained at an institution where we almost always used propofol to induce anesthetics for elective OR cases, so I found ways around the hypotension that results from propofol.

Once the physiology of etomidate adrenal suppression is understood, one would realize that it is a) not permanent, and b) dose dependent. Yes, you could argue that it deprives septic patients of much needed corticosteroids, but those patients are often being replaced with exogenous steroids because they are presumably depleted in the first place. Once the steroid wean begins, the effects of the etomidate should be negated.

Ironically enough, those who are in the 'anti-etomidate' camp are usually the first to spout out that one should use it for a 'slow, controlled induction' when they are face to face with oral board examiners.

 

[Monty Python]

Today I had a patient with an EF of 20 percent. I induced with propofol 50 mg IV and some rocuronium. Gentle mask ventilation with Sevo at 1.2 percent for two minutes. BP was rock stable the entire time including post intubation BP.

Is that mask venting until exhaled sevo = 1.2%, mask venting with exhaled sevo @ 1.2% for those two minutes, or mask venting by the clock for two minutes, with the dial set at 1.2% (and not concerned about exhaled %)? Thanks.

 

[deleted171991]

Chest. 2014 Sep 25. doi: 10.1378/chest.14-1012. [Epub ahead of print]
Single-Dose Etomidate Does Not Increase Mortality in Patients with Sepsis: A Systematic Review and Meta-Analysis of Randomized Controlled Trials and Observational Studies.
Gu WJ, Wang F, Tang L, Liu JC.
Abstract
ABSTRACT:
Background:The effect of single-dose etomidate on mortality in patients with sepsis remains controversial. We systematically reviewed the literature to investigate whether a single-dose etomidate for rapid sequence intubation increased mortality in patients with sepsis. MethodsubMed, Embase, and CENTRAL were searched for randomized controlled trials (RCTs) and observational studies regarding the effect of single-dose etomidate on mortality in adults with sepsis. The primary outcome was all-cause mortality. The Mantel-Haenszel method with random effects model was used to calculate pooled relative risks (RRs) and 95% confidence intervals (CIs). Results:Eighteen studies (two RCTs and sixteen observational studies), involving 5552 patients, were included. Pooled analysis suggested that single-dose etomidate was not associated with increased mortality in patients with sepsis, both in RCTs (RR, 1.20; 95% CI, 0.84 to 1.72; P = 0.31; I2 = 0%) and observational studies (RR, 1.05; 95% CI, 0.97 to 1.13; P = 0.23; I2 = 25%). When only adjusted RRs were pooled in five observational studies, RR for mortality was 1.05 (95% CI, 0.79 to 1.39; P = 0.748; I2= 71.3%). These findings also were consistent across all subgroup analyses for observational studies. Single-dose etomidate increased the risk of adrenal insufficiency in patients with sepsis (eight studies; RR, 1.42; 95% CI, 1.22 to 1.64; P < 0.00001). Conclusions:Current evidence indicates that single-dose etomidate does not increase mortality in patients with sepsis. However, this finding largely relies on data from observational studies, potentially subject to selection bias, and hence high-quality and adequately powered RCTs are warranted.