Etomidate is a short-acting intravenous hypnotic with a superior haemodynamic profile compared with alternative drugs such as propofol in patients at risk of acute cardiovascular instability.1,2 In a randomised controlled trial (RCT) in patients with severe aortic valve stenosis, etomidate was half as likely as propofol to evoke hypotension on induction; hypotension was also less severe and rescue therapy requirements were substantially reduced.3 In coronary artery bypass graft (CABG) surgery, haemodynamic stability was better maintained during anaesthetic induction with etomidate than with sevoflurane.4 In another RCT in cardiac surgery, etomidate use was not associated with increased vasopressor requirements or worse outcomes.5 In two cohort studies, including more than 6000 cardiac patients, single-dose etomidate was not associated with hypotension, worse outcomes or mortality.6,7
However, etomidate reversibly inhibits 11-β-hydroxylase, a mitochondrial enzyme required in the conversion of cholesterol to cortisol, thus producing transient adrenocortical suppression.1,8,9 Although this has been linked in retrospective studies to increased mortality, RCTs failed to confirm this association.5,10,11 Many clinicians nevertheless replaced etomidate with propofol. The discussion about the safety of etomidate continues.12 More evidence from RCTs is needed for an unbiased assessment of the effects of etomidate in those at cardiovascular risk.
The aim of the present study was to compare the effects of a single induction dose of etomidate with that of propofol on haemodynamics, adrenocortical responsiveness and early outcomes in patients undergoing elective CABG or mitral valve surgery (MVS). We tested the hypotheses that single-dose etomidate increases cumulative vasopressor requirement, is associated with prolonged time to extubation and length of stay in the ICU and has more adverse clinical effects.
Patients and methods
The study was approved by the Local Research Ethics Committee (Kantonale Ethikkommission Bern, N° 074/06) on 21 July 2006 and registered (NCT00415701, ClinicalTrials.gov). Signed written informed consent was obtained from all patients. Reporting complied with CONSORT guidelines.13
The single centre, double-blind, parallel group, prospective RCT13 was performed in elective cardiac surgery patients at the University Hospital, Bern, Switzerland, and started on 9 November 2006. Two cohorts of patients scheduled to undergo either on-pump CABG or MVS (repair or replacement via median sternotomy; underwent random allocation to treatments separately within each cohort.
CABG patients underwent randomisation to one of three treatment arms: anaesthetic induction using a single dose of etomidate 0.15 mg kg−1 combined with placebo (0.9% saline;); propofol 1.5 mg kg−1 with placebo; or etomidate 0.15 mg kg−1 combined with hydrocortisone 100 mg. Within each of the three arms, risk stratification was achieved by block randomisation according to low risk [primary CABG, left ventricular ejection fraction (LVEF) more than 40%; per treatment arm] or high risk [reoperation and/or LVEF ± 40% and/or expected duration of cardiopulmonary bypass (CPB) greater than 97 min].14
Elective patients suffering from mitral regurgitation and scheduled for MVS combined, if indicated, with CABG were allocated randomly to one of two treatment arms: anaesthetic induction using a single dose of etomidate 0.15 mg kg−1 or propofol 1.5 mg kg−1.
Eligible patients were adults between 18 and 80 years of age who had signed written informed consent and were scheduled for elective CABG or MVS. Exclusion criteria included participation in another clinical trial, known adrenocortical insufficiency, use of etomidate or propofol within 1 week preoperatively, use of glucocorticoids within 6 months preoperatively, known sensitivity to etomidate, propofol, or its emulsifier, severe hepatic dysfunction (serum bilirubin concentration >51 μmol l−1), severe renal dysfunction (plasma creatinine >180 μmol l−1), sepsis, endocarditis or other chronic inflammatory disease, insulin-dependent diabetes mellitus, positive HIV serology, haemodynamically significant carotid artery stenosis, other serious illness, pregnancy or breast-feeding, requirement for rapid sequence induction, emergency surgery and history of allergic asthma.
All patients received lorazepam 2 mg orally 1 h preoperatively and underwent a standardised anaesthetic induction sequence, starting with intravenous midazolam (50 to 100 μg kg−1), fentanyl (3 to 7 μg kg−1) and 2 ml of lignocaine 1% to obtund injection pain. The study drug (Etomidate Lipuro 0.2%, B. Braun Medical AG, Sempach, Switzerland or Disoprivane 2%, AstraZeneca GmbH, Zug, Switzerland) was administered through the same intravenous cannula over 120 s. Orotracheal intubation was facilitated with pancuronium (0.1 mg kg−1). To maintain mean arterial pressure within the range of 60 to 80 mmHg, noradrenaline (5 to 10 μg bolus) was administered. Maintenance of anaesthesia, haemodynamic management and CPB weaning were standardised and protocol-driven. In our department, postoperative monitoring with pulmonary artery catheter is omitted if preoperative myocardial function is good, intra-operative haemodynamics are stable and echocardiography reveals no abnormalities. All patients were transferred sedated (propofol 2 mg kg−1 h−1) and ventilated to the ICU. Postoperative care was standardised as per institutional routines.
In a sub-group of CABG patients who received etomidate, stress-dose hydrocortisone replacement (Solu-Cortef, Pfiser AG, Zürich, Switzerland) was administered intravenously instead of placebo to 30 patients in a randomised blinded fashion as a loading dose (100 mg intravenously) at anaesthetic induction, followed by 100 mg after 8 and 16 h (i.e. 300 mg on the day of surgery). The hydrocortisone dose was reduced to 100 mg twice daily on postoperative day 1 [morning dose given after adrenocorticotropin (ACTH) test] and to 100 mg in the morning of postoperative day 2.14 To ensure blinding, the other two treatment arms received placebo injections of identical appearance (0.9% saline) instead.
In all patients, a 250 μg adrenocorticotropin (ACTH) test (Synacthen, Ciba-Geigy, Basel, Switzerland) was performed on the day before surgery, at 7 and 24 h after anaesthetic induction, and in the morning of postoperative day 5 or 6. Plasma cortisol concentration, including corticosteroid-binding globulin without stimulation testing was measured 30 min following initiation of CPB, and 30 min after successful weaning from CPB. Absolute adrenal insufficiency was defined as a maximum serum cortisol concentration less than 500 nmol l−1 (18 μg dl−1) after ACTH stimulation. Relative adrenal insufficiency was defined as an increase in serum cortisol concentration less than 248 nmol l−1 after ACTH stimulation irrespective of basal cortisol concentration.15
Allocation to treatment arms occurred by central randomisation using a computer-generated list of random numbers (www.randomiser.org). The allocation ratio was 1 : 1 : 1 in CABG, and 1 : 1 in MVS. Allocation sequence was concealed in sequentially numbered, opaque, sealed envelopes. An independent study nurse not involved in the study opened the randomisation envelope and delivered appropriate syringes. Patients, investigators, caregivers, laboratory personnel and assessors were blinded to group assignment. Blinding was achieved by identical study drug appearance (white lipid emulsion of etomidate 0.2% or propofol 2% in identical syringes) and interventions (lignocaine pre-treatment, injection volume and speed). Also, hydrocortisone and placebo syringes were prepared to be indistinguishable. After completion of data collection, the database was closed. Investigators were deblinded after statistical analysis. No interim analyses were planned or performed.
The primary end point was cumulative noradrenaline and adrenaline dosage up to 24 h after induction on an intention-to-treat basis. Secondary end points included the incidence of failure to wean off CPB on first intention; serum lactate concentration at the end of surgery, and at 8 and 24 h; time to extubation; length of stay in ICU; and total duration of hospitalisation. The incidence of adrenal insufficiency was recorded.
Sample size calculation was based upon institutional registry data (Intellect 1.6.5, Dendrite Clinical Systems, Henley-on-Thames, UK), and assumed a mean of 180 ± 80 μg cumulative vasopressor load per patient and a maximal difference of 33% between groups. This required n = 20 per group for two groups, and n = 28 per group for three groups (α 0.05, β 0.80). Data are presented as mean ± SD, median Inter-quartile range (IQR) or as number (%). All analyses were performed using the intention-to-treat groups. Proportions were compared using χ2 testing. For group comparisons, Kruskal-Wallis equality-of-populations rank test was used. Vasopressor data were normalised to body weight. Effect size was calculated as absolute mean difference of cumulative vasopressor requirement, and as standardised mean difference (mean difference per standard deviation unit). Negative values for mean difference indicate lower vasopressor requirements for etomidate. Effect of standardised mean difference is rated as small (0.2), medium (0.5) or large (0.8). Risk ratios for adrenal insufficiency were calculated. Low and high-risk CABG subgroups were analysed accordingly.
A two-sided P value of less than 0.05 was considered significant. All statistical analyses were performed using Stata 13 for Mac OSX (StataCorp LP, Texas, USA).
The study flow chart is shown in Fig. 1. None of the 130 patients recruited were excluded from analysis, except for mortality, where two patients were lost to follow-up. Two protocol deviations occurred; however, analysis included both patients in their assigned group. Follow-up ended 30 days after induction of anaesthesia. Demographics are summarised in Table 1 and haemodynamic and surgical characteristics in Table 2.
In the CABG patients (n=90), the three treatment arms (n=30) did not differ in cumulative noradrenaline dose (P = 0.438, mean absolute difference −0.47 to 0.02 μg kg−1 per 24 h, standardised effect size 0.00 to 0.08) or adrenaline dose (P = 0.226, mean absolute difference −12.2 to −3.44 μg kg−1 per 24 h, standardised effect size 0.22 to 0.33). In the MVS patients (n=40), propofol (n=20) was associated with a small but significant increase in cumulative noradrenaline dose (P = 0.047, absolute mean difference −5.86 μg kg−1 per 24 h, standardised effect size 0.47) but not of cumulative adrenaline dose (P = 0.496, absolute mean difference −3.26 μg kg−1 per 24 h, standardised effect size 0.29) (Fig. 2).
In patients undergoing induction with etomidate with hydrocortisone supplementation (n=30), the serum lactate concentration was significantly increased both intra-operatively (P = 0.006) and at 8 h (P = 0.038). Postoperative haemodynamics, fluid and transfusion requirements as well as all other secondary end points showed no significant between-treatment effects (Table 3).
In both surgical groups, absolute adrenocortical insufficiency was evident only at 7 h following etomidate, regardless of hydrocortisone supplementation (CABG, P < 0.001; MVS, P = 0.004; Table 4). Relative adrenocortical insufficiency (RAI) occurred in all treatment arms at 7 and 24 h following anaesthetic induction. The incidence of RAI was significantly increased at 7 h after any etomidate administration (CABG, etomidate 83%, etomidate with hydrocortisone 80%, propofol, 37%; P < 0.001: MVS, etomidate 95%, propofol 35%; P < 0.001). At 24 h, an increased incidence of RAI persisted after etomidate only in CABG patients (CABG, etomidate 37%, etomidate with hydrocortisone 3%, propofol 3%; P < 0.001: MVS, etomidate 20%, propofol 10%; P = 0.376) (Table 4). In CABG patients, risk ratios for absolute and RAI ranged from 3.67 to 22 and from 1.08 to 2.0, respectively. In MVS patients, corresponding risk ratios were 14 and 2.11.
Within high (n=10) and low-risk (n=20) subgroups in CABG patients, administration of the induction hypnotic was not associated with differences in cumulative noradrenaline dose (low risk, P = 0.555, mean absolute difference −0.03 to 0.36 μg kg−1 per 24 h, standardised effect size 0.01 to 0.11; high risk, P = 0.368, mean absolute difference −2.02 to −0.66 μg kg−1 per 24 h, standardised effect size 0.09 to 0.18) or cumulative adrenaline dose (low risk, P = 0.097, mean absolute difference −13.51 to −0.03 μg kg−1 per 24 h, standardised effect size 0.32 to 0.45; high risk, P = 0.488, mean absolute difference −10.28 to 0.75 μg kg−1 per 24 h, standardised effect size 0.02 to 0.52).
In both risk subgroups, only patients exposed to etomidate fulfilled the criteria of absolute adrenocortical insufficiency 7 h after induction, irrespective of hydrocortisone supplementation (P < 0.005). Etomidate significantly increased the incidence of RAI in both risk groups 7 h after induction (low-risk, etomidate 80%, propofol 35%; etomidate with hydrocortisone 75%; P = 0.005: high-risk, etomidate 90%, propofol 40%, etomidate with hydrocortisone 90%; P = 0.014). At 24 h, this effect persisted in less than half of the low-risk CABG subgroup (etomidate 45%, propofol 5%, etomidate with hydrocortisone 5%; P = 0.001).
Secondary endpoints and 30-day mortality did not differ between risk subgroups. There were two perioperative deaths in the 130 patients, both after propofol induction (right heart failure, CABG low-risk, n = 1; unplanned double-valve replacement and thoracic aortic repair, MVS, n = 1). A causative relationship of these serious adverse events with study drug exposure was judged unlikely by the investigators.
In the present study, patients undergoing elective cardiac surgery were randomised to receive propofol or etomidate as the anaesthetic induction agent. We demonstrated that cumulative vasopressor requirements and haemodynamics during the first 24 h after induction were not significantly different between groups, except for a minor increase in cumulative noradrenaline dosage following propofol induction for MVS. This was found despite transient adrenocortical suppression following reproducible etomidate exposure, though not exclusive so. The use of etomidate also had no detrimental effect on procedural end points such as failure to wean from CPB, time to extubation, length of stay in ICU and in hospital, or all-cause mortality.
In patients undergoing MVS, we found a small but significant increase in noradrenaline dosage associated with propofol. Low-dose milrinone was part of the routine CPB weaning protocol in MVS patients, and a probable explanation for the increased noradrenaline requirements is the interaction of propofol with milrinone-induced vasodilatation. However, this difference appears of low clinical importance according to its effect size.
Transient inhibition of adrenal steroid synthesis associated with the use of etomidate has been linked to increased mortality.1,8,9 Although the evidence is conflicting,9,12 this has led many clinicians to replace etomidate with the haemodynamically less benign propofol, even for single-bolus induction in the operating theatre. However, severe hypotension after induction of anaesthesia occurs in 10% of patients and propofol, in contrast with etomidate, has been identified as an independent predictor of post-induction hypotension.2 In RCTs in cardiac surgery, etomidate has clearly been shown to provide better haemodynamic stability during anaesthetic induction.3,4
In cardiac surgery, two retrospective outcomes analyses of 6181 patients did not find any evidence that etomidate exposure was associated with more severe hypotension, longer mechanical ventilation or hospital stay, or increased in-hospital mortality.6,7 Our results confirm the findings of this large series and of another RCT in low-risk cardiac surgery.5 The findings of that RCT and our study are consistent in terms of vasopressor use, outcomes and incidence of adrenocortical insufficiency.
Beyond that, our study adds confirmative results for patient subsets at higher risk, that is, with LVEF less than 40%. In this high-risk subgroup, we did not find any differences in cumulative vasopressor requirements. A non-significant trend toward lower adrenaline dosage in both low and high-risk etomidate arms appears of low clinical importance according to its effect size. The results for adrenocortical insufficiency were similar to those in the entire CABG cohort.
Moreover, in our study, we observed no beneficial effects from hydrocortisone supplementation, but found increased serum lactate concentrations intra-operatively and in the early postoperative period. However, the absolute lactate concentrations were still at the upper limit of normal, and the practical significance of this finding remains unclear. This resembles findings from the ‘Dexamethasone for Cardiac Surgery’ trial in cardiac surgery, where a single high intra-operative dose of dexamethasone led to increased serum lactate concentrations during the first 15 h.16 The risk of developing RAI after etomidate exposure did not change with hydrocortisone supplementation, whereas the incidence of absolute adrenal insufficiency was reduced. However, stress-dose cortisol supplementation is not supported by good evidence even in perioperative or critically ill patients with adrenal insufficiency.17–19 As our study found no clinical benefits beyond the correction of laboratory values, steroid replacement to counteract the effect of etomidate on inhibition of 11β-hydroxylase is not supported by our results.
Our study found no haemodynamic or other clinical correlate of drug-induced hypoadrenalism, the suspected mechanism by which etomidate should increase morbidity and mortality. Suppression of adrenal function by hypnotics other than etomidate (e.g. thiopental, propofol, ketamine and midazolam) has been a consistent finding, ranging from 12 to 57%.5,11,20 In our study, the incidence for propofol was 36%. This indicates that mechanisms other than etomidate-mediated inhibition of 11β-hydroxylase also play significant roles in suppressing intra-operative plasma cortisol concentration. Moreover, the mechanism of recovery of plasma cortisol early postoperatively also appears multifactorial. There is evidence that corticotropin stimulation by inflammatory cytokines may contribute.21 In our study on post-induction day 1, 63% of CABG and 80% of MVS patients showed a restored response to cortisol stimulation, with no patient remaining in absolute adrenal insufficiency.
A limitation of this study is its small size, and in particular it has insufficient power for analysis of mortality, other important clinical outcomes and low rate adverse events. Furthermore, our results might have differed with a comparator other than propofol.
In conclusion, the results of the current study do not support the hypothesis of increased perioperative vasopressor requirement after exposure to a single dose of etomidate. Etomidate does induce more frequent adrenocortical insufficiency. Clinically, this does not translate into increased vasopressor requirements, less favourable haemodynamics, prolonged length of mechanical ventilation and ICU or hospital stay, or worse outcomes. Without solid evidence to the contrary, etomidate should remain one of several useful induction hypnotics in the armamentarium of cardiovascular anaesthetists.
Acknowledgements relating to this article
Assistance with this study: the authors would like to thank the research nurses Judith Kaufmann Erni, Natalie Araya, Torsten Konrad, Zita Bischofberger-Schmidli, Gerald Kleemanns, Peter Zurbuchen, Michael Lensch and Monika P. Stucki for their excellent technical support and Catherine Reid MD for proofreading. Co-authors Antje Rindlisbacher and Esther Begert made an equal contribution to the study.
Financial support and sponsorship: the study was funded by the Foundation for Research in Anesthesiology and Intensive Care Medicine at Inselspital Bern, and by the Department of Anesthesiology and Pain Medicine and the Department of Intensive Care Medicine, University Hospital, University of Bern, Bern, Switzerland.
Conflicts of interest: none.
Presentation: data from this study were presented at the Annual Meeting of The Swiss Society of Anesthesiology and Resuscitation, Interlaken, Switzerland, 6 to 8 November 2014.
1. Forman SA. Clinical and molecular pharmacology of etomidate. Anesthesiology
2. Reich DL, Hossain S, Krol M, et al. Predictors of hypotension after induction of general anesthesia. Anesth Analg
3. Bendel S, Ruokonen E, Pölönen P, Uusaro A. Propofol causes more hypotension than etomidate in patients with severe aortic stenosis: a double-blind, randomised study comparing propofol and etomidate. Acta Anaesthesiol Scand
4. Cheong KF, Choy JM. Sevoflurane-fentanyl versus etomidate-fentanyl for anesthetic induction in coronary artery bypass graft surgery patients. J Cardiothorac Vasc Anesth
5. Morel J, Salard M, Castelain C, et al. Haemodynamic consequences of etomidate administration in elective cardiac surgery: a randomised double-blinded study. Br J Anaesth
6. Wagner CE, Bick JS, Johnson D, et al. Etomidate use and postoperative outcomes among cardiac surgery patients. Anesthesiology
7. Heinrich S, Schmidt J, Ackermann A, et al. Comparison of clinical outcome variables in patients with and without etomidate facilitated anesthesia induction ahead of major cardiac surgery: a retrospective analysis. Crit Care
8. Varga I, Rácz K, Kiss R, et al. Direct inhibitory effect of etomidate on corticosteroid secretion in human pathologic adrenocortical cells. Steroids
9. Hohl CM, Kelly-Smith CH, Yeung TC, et al. The effect of a bolus dose of etomidate on cortisol levels, mortality, and health services utilization: a systematic review. Ann Emerg Med
10. Jabre P, Combes X, Lapostolle F, Dhaouadi M, et al. Etomidate versus ketamine for rapid sequence intubation in acutely ill patients: a multicentre randomised controlled trial. Lancet
11. Tekwani KL, Watts HF, Sweis RT, et al. A comparison of the effects of etomidate and midazolam on hospital length of stay in patients with suspected sepsis: a prospective, randomised study. Ann Emerg Med
12. Erdoes G, Basciani RM, Eberle B. Etomidate: a review of robust evidence for its use in various clinical scenarios. Acta Anaesthesiol Scand
13. Moher D, Hopewell S, Schulz KF, et al. CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trials. Br Med J
14. Kilger E, Weis F, Briegel J, et al. Stress doses of hydrocortisone reduce severe systemic inflammatory response syndrome and improve early outcome in a risk group of patients after cardiac surgery. Crit Care Med
15. Loisa P, Uusaro A, Ruokonen E. A single adrenocorticotropic hormone stimulation test does not reveal adrenal insufficiency in septic shock. Anesth Analg
16. Ottens TH, Nijsten MW, Hofland J, et al. Effect of high-dose dexamethasone on perioperative lactate levels and glucose control: a randomized controlled trial. Crit Care
17. Payen JF, Dupuis C, Trouve-Buisson T, et al. Corticosteroid after etomidate in critically ill patients: a randomized controlled trial. Crit Care Med
18. Yong SL, Coulthard P, Wrzosek A. Supplemental perioperative steroids for surgical patients with adrenal insufficiency. Cochrane Database Syst Rev
19. Boonen E, Vervenne H, Meersseman P, et al. Reduced cortisol metabolism during critical illness. N Engl J Med
20. Kenyon CJ, McNeil LM, Fraser R. Comparison of the effects of etomidate, thiopentone and propofol on cortisol synthesis. Br J Anaesth
21. Naito Y, Tamai S, Shingu K, et al. Responses of plasma adrenocorticotropic hormone, cortisol, and cytokines during and after upper abdominal surgery. Anesthesiology