Newsletter home

Mark Twite, MA, MB, BChir, FRCP

Luis Zabala, MD


President's Message

Letter from the Editor

Contrast Agents and Pediatric Cardiac Catheterization


Etomidate: The controversy continues

CCAS Meeting Reviews

Morning Sessions

Afternoon Sessions




Etomidate: The controversy continues

Reviewed by: Sana Ullah, MD
Associate Professor of Anesthesiology
UT Southwestern
Dallas, TX

Komatsu R et al. Anesthetic Induction with Etomidate, Rather than Propofol, Is Associated with Increased 30-Day Mortality and Cardiovascular Morbidity After Noncardiac Surgery. Anesth Analg 2013; 117:1329-37

Wagner CE et al. Etomidate Use and Postoperative Outcomes among Cardiac Surgery Patients. Anesthesiology 2014; 120:579-89

Etomidate has been used as an induction agent since the early 80s. The minimal hemodynamic effects derived from a full induction dose make it a very useful agent in patients at high risk of cardiovascular instability such as in the cardiac operating room, ICU, trauma and ER setting. Based on its favorable cardiovascular profile, it became popular as a sedative in the ICUs, but was very quickly linked to an increased mortality due to sepsis in multiple trauma patients, and is no longer used in this capacity [1].  Subsequent research revealed that etomidate causes adrenal insufficiency by inhibiting the enzyme 11B-hydroxylase, and it is purported, but not proven, that this adrenal suppression may be the key mechanism for increased mortality. Even sub-anesthetic doses can produce adrenal suppression which can last over 24 hours in critically ill patients.

Despite many publications, including several meta-analyses, there is on-going controversy about the safety of etomidate in the critical care setting, particularly in sepsis. However, most of these studies are hampered by small numbers and retrospective design. Although there is no direct link between etomidate and increased mortality, there have been calls for abandonment or severe curtailment of its use as far back as ten years ago [2,3]. Even the recent American Heart Association (AHA) PALS guidelines of 2010 recommend that it should not be routinely used in pediatric patients with septic shock [4]. Two recent publications in the anesthetic literature add more fuel to the controversy about the use of etomidate as an induction agent in the non-cardiac and cardiac surgery setting.

The first paper by Komatsu et al. looked at outcomes between two groups of adult patients having non-cardiac surgery who received either etomidate or propofol at induction. This was a retrospective, observational study based on data extracted from their perioperative electronic databases. The study population consisted of adults, ASA status 3 & 4, having non-cardiac surgery under general anesthesia, with or without regional anesthesia, requiring at least one night of postoperative hospitalization. Anesthesia was maintained with volatile agents supplemented with opioids and muscle relaxants. The types of surgeries were fairly typical of what might be seen at a large tertiary care center in the United States – major abdominal, including aortic resections, urological and orthopedic.

The two groups of patients were matched using propensity scores based on all prespecified potential confounding variables. Propensity scoring is a statistical technique used to reduce "selection bias” in observational studies where there is a non-random allocation of treatment [5]. In simplistic terms, it simulates randomization in non-randomized study populations and its use is increasing in analyzing observational data. 2144 patients given etomidate were matched with 5233 patients given propofol, based on the prespecified variables and the type of surgery.  The primary outcomes were 30-day mortality, any major in-hospital cardiovascular morbidity, and any major in-hospital infectious morbidity. Secondary outcomes were the relationship between the amount of etomidate received scaling by weight and the primary outcomes, intraoperative vasopressor use, and minimum systolic and diastolic blood pressures intraoperatively.

The results were interesting. The etomidate group had 2.5 times the odds of dying versus the propofol group (odds ratio [OR] 2.5, 98% confidence interval [CI], 1.9-3.4).  The mortality was 6.5% with etomidate versus 2.5% with propofol. The cardiovascular morbidity was also higher – 7.6% versus 4.9% - (OR 1.5 [1.2-2.0]). Etomidate was also significantly associated with a prolonged hospital stay (P < 0.001), such that these patients were 18% less likely to be discharged from hospital at any given point postoperatively compared with the propofol patients. Rather surprisingly, there was no difference in infectious morbidity or intraoperative vasopressor use between the two groups. There was also no difference in intraoperative blood pressures either. However, we do not know if there was any difference in vasopressor use post-operatively.

The second publication, another single institution, retrospective, observational study, looked at the effects of etomidate on outcomes in a large group of adult cardiac surgical patients, and came to a different conclusion. Wagner et al. did a retrospective analysis on all adult cardiac surgeries carried out at their institution over a two-year period between 2007 -2009.The patient information was obtained from the Society of Thoracic Surgeons Database and included age, sex, body mass index, race, type of surgery, history of diabetes mellitus, history of congestive heart failure, preoperative angiotensin-converting enzyme inhibitor use, ejection fraction, last preoperative creatinine level, presence of cardiogenic shock, status of the surgery (elective, urgent, and emergent), mechanical ventilation hours, hospital length of stay, and in-hospital mortality. The induction agents included etomidate, propofol, midazolam, and fentanyl.

Sixty two percent of 3127 patients in their study were given etomidate at induction. Anesthesia was maintained with isoflurane and all patients were placed on a propofol infusion during transfer to the ICU. The clinical outcomes they looked at were severe hypotension, total time on mechanical ventilation, hospital length of stay, and in-hospital mortality. Severe hypotension was defined as a norepinephrine requirement of greater than or equal to 15mcg/min for greater than 1h to maintain a mean arterial BP of 60-70 mmHg.

The baseline demographics of the etomidate and non-etomidate groups were similar with the exception of the etomidate group having a more patients with a history of CHF (23 vs 18%; P = 0.002), and were less likely to present with cardiogenic shock (1 vs 4%; P < 0.001). This latter finding is explained by the fact that patients in cardiogenic shock arrived in the operating room already intubated and were not given an “induction” agent, although it is not known what sedation was used to facilitate intubation before arrival to the operating room.

Overall, there was no difference in unadjusted and adjusted (regression model & propensity score approach) outcomes between the etomidate and non-etomidate groups. Based on this study, the authors concluded that, after adjusting for potential confounders, there is no evidence to suggest that etomidate exposure at induction is associated with severe hypotension, longer mechanical ventilation, longer hospital stay, or increased mortality.
So, as cardiac anesthesiologists who more frequently use etomidate in our practice when compared to our general counterparts, what can we learn from these two studies with conflicting results, albeit, in different patient populations?

The study by Komatsu et al. is very concerning if we are to accept their conclusion that etomidate use increases the odds of dying by 2.5 times, and also is associated with increased cardiovascular complications. Despite the large number of patients (2616) and the statistical methods used to reduce treatment selection bias (propensity score matching), it is still a retrospective study which shows an “association” and not a “correlation”. Retrospective studies do not allow controlling of all the potential confounders, which means that there may be many unknowns which might have affected the outcomes. Despite propensity score matching, it was a very heterogeneous surgical population. We do not know the level of perioperative monitoring that was used (arterial, central venous etc.) to guide fluid and hemodynamic management. We do not know what fluids were administered – crystalloid, colloid or blood products – which can affect outcomes. We have little information about other medications used perioperatively, such as antibiotics, steroids, and anticoagulants. There is no information about the skill levels of the anesthesia providers (residents, fellows or attendings) and the surgeons. Where the patient was managed postoperatively may have affected outcome – intensive care, high dependency unit or a ward. There can be many individual physician or institutional practices which may bias a study. All these considerations do not imply that we should simply dismiss the findings, but we should be cautious in interpreting them and applying them to our own practices.

The study by Wagner et al. comes to a different and somewhat reassuring conclusion that in adult patients having cardiac surgery, etomidate did not increase mortality or morbidity. Their study population is much closer (but still different in many respects) to the patients that pediatric cardiac anesthesiologists are used to treating. Perioperative management in cardiac surgery, whether adult or pediatric, is fairly standardized. There is likely to be very little variation in anesthesia or surgical practices in a particular institution and all patients are intensively managed in a critical care setting after surgery. All this is clearly going to have an effect on outcomes compared with a diverse general non-cardiac surgery population with multiple differing co-morbidities and variation in perioperative management.

So, based on what is known about etomidate so far, and in light of these two recent publications, should pediatric anesthesiologists taking care of patients having cardiac surgery reconsider the use of etomidate in their practice? In many circumstances, etomidate is the ideal choice for induction – severe valvar stenosis, hypertrophic cardiac myopathy, or severe ventricular dysfunction. The usual alternative, ketamine, although having a good cardiovascular safety profile, has the undesirable side effects of tachycardia, postoperative delirium, and possible adverse effects on the pulmonary vascular resistance.

Until better evidence about the link between etomidate and clinically relevant outcomes is available, etomidate should remain in the cardiac anesthesia armamentarium. However, large prospective randomized controlled trials of etomidate versus other agents may be difficult to perform due to the informed consent process. Will patients (or their parents) consent to the etomidate group given what is already published about the possible link between this drug and adverse outcomes?

The issue of altered adrenocortical function in the post-operative period, and its relevance to clinically important outcomes is also controversial, as is the use of steroid supplementation, knowing the well-known risks of steroids [6]. Etomidate may also be a confounding factor in testing for adrenocortical function post-operatively.

In summary, the debate about etomidate will continue until there is more compelling evidence about its influence on postoperative outcomes [7]. Whether large prospective randomized trials will be feasible to definitively answer this question remains to be seen. Until then, etomidate will remain a good choice in certain groups of patients who are at high risk of cardiovascular instability during the critical induction period.


  1. Watt I, Ledingham IM. Mortality amongst multiple trauma patients admitted to an intensive therapy unit. Anaesthesia 1984; 39:973-81
  2. Morris C, McAllister C. Etomidate for emergency anaesthesia; mad, bad and dangerous to know? Anaesthesia 2005; 60:737-740
  3. Annane D. ICU physicians should abandon the use of etomidate! Intensive Care Med 2005; 31:325-326
  4. Kleinman ME, Chameides L, Schexnayder SM et al. Part 14: Pediatric Advanced Life Support: 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 2010; 122: S876-S908
  5. Okoli GN, Sanders RD, and Myles P. Demystifying propensity scores. Br J Anaesth 2014;112: 13-15
  6. Green ML, Koch J. Adrenocortical function in the postoperative pediatric cardiac surgical patient. Curr Opin Pediatr 2012;24: 285-290
  7. Van den Heuvel I, Wurmb TE, Bottiger B et al. Pros and cons of etomidate – more discussion than evidence? Curr Opin Anesthesiol 2013;26: 404-408
Back to top