Background
For as long as I can remember emergency and critical care clinicians have debated the optimal induction agent for tracheal intubation in the critically ill. When I started my anaesthetic placements I uswed only two drugs for an RSI, Thiopentone and Suxamethonium. Anything else was considered ‘fancy’ amongst the older anaesthetists. The cool kids were using Etomidate/Sux as an alternative, and as I wanted to move with the times, that became my go-to regime for many years. That changed when Ketamine became the standard here in Virchester and these days it’s rare that I use anything else to be honest. I even still occasionally use Etomidate in selected cases (mostly cardioversions), but the logic behind that is not clear (even to me).
In other parts of the world Etomidate holds on. It is a cardiovascularly stable drug, it’s pretty easy to dose and use, (Ed – and it makes the anaesthetists happy because it’s white and reminds them of Propofol, which we almost never use in sick patients). Remember that my practice in an MTC and in PHEM means that the patients I anaesthetisse are typicaly critically unwell/injured.
Etomidate, has a rapid onset and reliable haemodynamic profile, but fell out of favour largely due to concerns about adrenal suppression, which can last for as long as 72 hours after a single bolus. This biochemical effect prompted concerns that etomidate might worsen outcomes by impairing physiological resilience at precisely the moment critically ill patients need it most. Observational studies have linked etomidate to higher mortality, particularly in septic patients, and some guidelines have recommended avoiding it entirely.
Ketamine is pretty ubiquitous here in Virchester. It’s considered to be cardiovasculaly stable and without the adrenal suppression concerns ofm Etomidate. It’s easy to give and dose and can be given through a variety of different routes. It tends to preserve airway reflexes and sympathetic tone, which are also perceived benefits in unwell/shocked patients.
So we have seen fashions change, but the evidence base has not really been that robust. This week we see the Randomized Trial of Sedative Choice for Intubation (RSI Trial) published and discussed at the Critical Care Reviews Down Under conference. The paper itself is published in NEJM and social media has declared that Ketamine is dead as a result. But is it? Maybe we need to dive a little deeper. First habve a read of the abstract and of course read the full paper. Make up your own mind and don’t just believe what you read on a blog or social media.
Abstract
BACKGROUND For critically ill adults undergoing tracheal intubation, observational studies suggest that the use of etomidate to induce anesthesia may increase the risk of death. Whether the use of ketamine rather than etomidate decreases the risk of death is uncertain.
METHODS In a randomized trial conducted in 14 emergency departments and intensive care units in the United States, we randomly assigned critically ill adults who were undergoing tracheal intubation to receive ketamine or etomidate for the induction of anesthesia. The primary outcome was in-hospital death from any cause by day 28. The secondary outcome was cardiovascular collapse during intubation, defined by the occurrence of a systolic blood pressure below 65 mm Hg, receipt of a new or increased dose of vasopressors, or cardiac arrest.
RESULTS A total of 2365 patients underwent randomization and were included in the trial population; 1176 were assigned to the ketamine group and 1189 to the etomidate group. In-hospital death by day 28 occurred in 330 of 1173 patients (28.1%) in the ketamine group and in 345 of 1186 patients (29.1%) in the etomidate group (risk difference adjusted for trial site, −0.8 percentage points; 95% confidence interval [CI], −4.5 to 2.9; P = 0.65). Cardiovascular collapse during intubation occurred in 260 of 1176 patients (22.1%) in the ketamine group and in 202 of 1189 patients (17.0%) in the etomidate group (risk difference, 5.1 percentage points; 95% CI, 1.9 to 8.3). Prespecified safety outcomes were similar in the two groups.
CONCLUSIONS Among critically ill adults undergoing tracheal intubation, the use of ketamine to induce anesthesia did not result in a significantly lower incidence of in-hospital death by day 28 than etomidate. (Funded by the Patient-Centered Outcomes Research Institute and others; RSI ClinicalTrials.gov number, NCT05277896.)
What kind of study is this?
The RSI Trial is a pragmatic, multicentre, randomised controlled trial conducted across 14 emergency departments and intensive care units in the United States. Randomisation was 1:1 between ketamine and etomidate, with allocation concealed until the point of enrolment. The pragmatic design is important: clinicians selected doses from a weight-based nomogram but otherwise practised as they normally would, allowing the trial to reflect real-world emergency airway management.
The study was unblinded, an unavoidable limitation given the visible and pharmacological differences between the medications, but carefully structured to ensure independent data collection: trained observers documented haemodynamic data and procedural characteristics, while follow-up data were collected separately from the immediate clinical team.
The primary outcome was all-cause in-hospital mortality by day 28. A secondary outcome—cardiovascular collapse during intubation—was defined prospectively, encompassing profound hypotension (systolic blood pressure <65 mmHg), initiation or escalation of vasopressor support, or cardiac arrest between induction and two minutes after the procedure. A range of exploratory haemodynamic and clinical outcomes were also collected.
With more than 2300 enrolled patients, this is the largest randomised study ever conducted to compare induction agents for emergency intubation. The sample size was specifically powered to detect a clinically meaningful difference in mortality, something previous studies were never large enough to do reliably.
Tell me about the patients
The study population was typical of patients requiring emergency intubation in 6 EDs and 8 ICUs in the US. The median age was 60 years, and approximately half of all intubations occurred in each of the two settings. Almost half of the patients had sepsis or septic shock at the time of intubation. Around one-fifth were already receiving vasopressors before induction, and physiological severity was substantial, with a median APACHE II score of 18. Nearly 30% ultimately died by 28 days.
Comorbidities such as cirrhosis, heart failure, and chronic steroid use were common, and acute respiratory illness was the leading indication for intubation.
Most importantly for me is the list of exclusions. Patients were excluded if they were known to be pregnant, were known to be prisoners, were presenting with a primary diagnosis of trauma, or had an immediate need for tracheal intubation that precluded randomization. If I look a the patients I see I suspect that the vast majority would be excluded. So the relevance of this paper to my practice is potentially limited. Similarly patients were also excluded if the treating clinicians determined that the use of ketamine or etomidate was either necessary or contraindicated.
Table 1 tells me that these patients are very different to the ones I get involved with. Roughly 50% are patients with sepsis, 57% with respirato
Tell me about the intervention
I’ve reviewed many trials on medications and it’s always tricky to work out whether you just let clinicians decide a dose or whether you specify in advance. In reality we often adjust drug induction doses for a whole variety of reasons, but that does not always translate well into a trial protocol. In this study the drug doses were as follows, allowing clinicians to do some adjustment.
Ketamine at 2, 1.5 or 1mg/Kg
Etomidate at 0.3, 0.25 or 0.2mg/Kg
Interestingly, 30% of patients received Sux rather than Roc.
My only comment here is that in truly critically unwell patients 1mg/Kg ketamine is still a lot of ketamine (I’ve used a lot less in some cases). In the study the median dose of ketamine was 140mg, which feels a lot higher than in my practice. Similarly the median dose of Etomidate was 20mg which based on my past experience would again be a pretty big dose for practice here.
Across both groups, adherence to allocated medication was excellent, with more than 99% receiving their assigned induction agent. Use of neuromuscular blockade was almost universal. The study population therefore reflects real-world emergency intubations in a medically unwell cohort, and the internal validity of the comparison is high.
What were the measured outcomes in this study?
The primary outcome was in-hospital death by day 28. Given historical concerns about adrenal suppression and haemodynamic effects, mortality is arguably more important than short term physiological measurement.
That said, the main debate has been around CVS stability and the authors sensibly looked at a range of these.
- Cardiovascular collapse during intubation was defined as a specific secondary outcome. They defined this as during the 2 minutes after tracheal intubation: a systolic blood pressure below 65 mm Hg, receipt of a new or increased dose of vasopressors, or cardiac arrest. This is quite a short period of time and I think could only be reliably determined if invasive arterial line monitoring would be in place (but only about 20% had this). Is a very short period of hypotension really that important (we don’t have data on this) or would a better approach be to look at time and depth of hypotension (the dose of badness is severity + time in this case). Is the first two minutes enough to really know?
- Exploratory outcomes included detailed haemodynamic measures (such as the lowest blood pressure, changes in blood pressure, oxygen saturation parameters), procedural metrics (including first-pass success and time from induction to tube placement), and later outcomes such as ventilator-free, vasopressor-free, and ICU-free days. These should be considered hypothesis generating
What are the main results?
The headline figure, and the most important one is that there is no difference in mortality at 28 days. It did not differ between the two groups. By day 28, 28.1% of patients in the ketamine group and 29.1% in the etomidate group had died. The adjusted risk difference was −0.8 percentage points, with a 95% confidence interval ranging from a possible 4.5-point benefit to a 2.9-point harm. In other words, the confidence interval excludes any clinically important difference. Sensitivity analyses, including those accounting for missing data and alternative definitions of mortality, yielded the same conclusion. Subgroup analyses, including septic patients, those receiving vasopressors, and those with high severity of illness, showed no meaningful difference either.
The secondary outcome (CVS collapse) is different. Cardiovascular collapse occurred more frequently in the ketamine group (22.1%) than the etomidate group (17.0%). The absolute difference of just over five percentage points is clinically notable. Importantly, the groups in which ketamine is often assumed to offer haemodynamic advantages, sepsis, vasopressor-dependent patients, and those with higher APACHE II scores showed the most striking differences. Among patients with sepsis, cardiovascular collapse occurred in 30.6% of those induced with ketamine compared with 20.9% with etomidate. In those with APACHE II scores ≥20, collapse rates were 31.4% and 20.7% respectively.
Exploratory haemodynamic outcomes consistently favoured etomidate. Patients receiving ketamine experienced lower nadir systolic blood pressures, more episodes of systolic BP <80 mmHg, and more substantial drops in blood pressure (>30 mmHg). There was also a higher incidence of ventricular tachycardia in the ketamine group, although absolute numbers were small.
Procedural outcomes such as first-pass success and oxygenation did not differ significantly between groups, nor did downstream clinical outcomes such as ventilator-free or ICU-free days. Haemodynamic effects at 24 hours were broadly similar.
In summary, ketamine did not reduce mortality and was associated with noticeably more haemodynamic compromise during intubation.
So is this definitive?
This study is a great study, but I’m not sure it really helps me as an EM/PHEM clinician.
On the positive side of things, it is large, randomised, and pragmatic, capturing practice across multiple centres with diverse clinical teams, we love that at St Emlyn’s so lots of positives from us. They also managed really good adherence to treatment arms, they were really good at following up their patients, and the statistical methods were clear and well designed.
The pragmatic design reflects real-world airway management and enhances external validity. In addition, the haemodynamic data, captured by independent observers is great.
There are, however, limitations. It is unblinded, and the use of observers rather than electronically collected data introduces the possibility of bias in haemodynamic assessments and vasopressor titration. While many of the physiological parameters are objective, decisions about vasopressor escalation are clinician-dependent and could be influenced by expectations about each drug. The definition of cardiovascular collapse is reasonable, but does not really reflect practice. A single low BP recording equates to cardiovascular collapse. Is that really ‘collapse’ I don’t think so. Similarly the two minute window is really short. Really, really short. It’s clearly important but I’m interested in CVS stability over a longer period of time than a couple of minutes. There is also the concern about Etomidate and adrenal suppression, that does not happen within a 2-min window, so one might argue that this trial observation period favours investigating just one potential complication.
Julian Morgenstern has a great blog on this trial that highlights the fact that it is arguably underpowered. It was set up to detect a difference of 5.2% which is HUGE, and much greater than almost all critical care interventions. The realistic chances of finding such a difference were really small and therefore this trial was unlikley to find a difference from the very beginning. In fact ketamine performed better than etomidate, but only by 1% which this study is too small to detect. That said, a 1% difference in this group of patients is (I think) important. So it’s a shame that this trial is too small to detect clinically important differences (Ed: Note – that 1% difference only applies to in-hospital mortality, so may not exist in real patient outcome terms).
The exclusion of trauma, and patients requiring immediate management really limits generalisability to my practice. In addition, although the study is powered to detect meaningful differences in mortality, the confidence intervals still allow for very small differences in effect, though none large enough to justify preferring one drug on mortality grounds alone. And just as a reminder, the main outcome measure shows – no difference.
Should we change practice based on this study?
Not for me. I like Etomidate and would be happy to use it, but I would not change on the basis of this trial. It’s not my patients, the observation period is too short and the main outcome measure (mortality ) shows no difference. Many of the patients I intubate, (notably trauma patients who were excluded here) also need a good dose of analgesia which Etomidate does not deliver, so another drug would be needed to deliver this.
On social media people are saying that this is the end of Ketamine in RSI: I don’t think this trial, good though it is, justifies the removal of ketamine as an RSI drug in my practice. If you spend your time intubating lots of septic patients, then this might persuade you to use Etomidate more, but do remember that it won’t make any difference to mortality.
Zaf’s thoughts: We have largely moved to ketamine as the sedative of choice for our intubations in the US, but there are still enough of us who have been comfortable with and still use etomidate instead. Unlike some other countries, there are no restrictions to use in the US at this time. I think this study will at the least be reassuring for those worried about the role of etomidate in outcomes in critically ill patients. I think the bigger problem we face is that for a variety of reasons we’ve moved to a “one size/drug fits all” when intubating these super sick patients when in reality, as clinicians, we should adopt nuanced, individualized approach to drug choices and overall intubation approach. This paper adds to our knowledge base about drugs we commonly go to, and that can only make us better in the choices we make when we take over a patient’s airway.
On the bottom line site they have also pointed out that we don’t really know whether other doses might produce less instability, or whether a DSI approach may change matters. All areas that we could explore in future trials.
Summary
The RSI Trial is a good trial of ketamine vs. etomidate for induction of anaesthesia during emergency tracheal intubation in (some) critically ill adults. It shows no difference in mortality between the two agents. Etomidate may be more cardiovascularly stable in the two minutes after intubation. At the doses given in this trial both drugs can result in early cardiovascular instability.
Those claiming on social media that this trial is the death of ketamine …… are wrong 😉
References and further reading.
- Article: Casey JD, Seitz KP, Driver BE, et al. Ketamine or Etomidate for Tracheal Intubation of Critically Ill Adults. N Engl J Med. Published online December 9, 2025. doi:10.1056/NEJMoa2511420 (includes links to protocol and 94 page supplement)
- Further Reading: EMCrit Wee – The RSI Trial (2025)
- Interview with one of the leading investigators: ICU Ed and Todd Cast – CCR Down Under: RSI with Jon Casey (2025)
- Results Presentation:Critical Care Reviews: RSI trial presentation by Matthew Semler and John Casey
- RSI: Ketamine or Etomidate for Tracheal Intubation of Critically Ill Adults. https://www.thebottomline.org.uk/summaries/icm/rsi-ketamine-or-etomidate-for-tracheal-intubation-of-critically-ill-adults/
- Morgenstern, J. Ketamine versus Etomidate | The RSI trial, First10EM, December 22, 2025. Available at:
https://doi.org/10.51684/FIRS.144668 - Josh Farkas. PulmCrit: Hot take on RSI trial of ketamine vs etomidate. https://emcrit.org/pulmcrit/rsitrial/
