Using Etomidate in Patients With Suspected Sepsis Does Not Increase LOS

Megan Brooks

September 24, 2010

September 24, 2010 — In a prospective, randomized study, patients with suspected sepsis who received a bolus of etomidate before rapid sequence intubation spent no more time in the hospital than similar patients who received a bolus of midazolam.

Etomidate, the researchers note, is ideal for rapid intubation in septic patients because of its "predictability in dosing, reliable hypnotic effect, rapid onset, and short duration of effects and because it has no effect on histamine release, which contributes to its hemodynamic stability." However, previous studies have found adverse outcomes from a single bolus dose of etomidate in septic patients.

But in this study, along with similar hospital stays, the two groups also spent similar amounts of time in intensive care and on a ventilator, according to a September 10th online paper in Annals of Emergency Medicine.

But the trial was underpowered to detect any differences in mortality, "which is the more meaningful clinical outcome," first author Dr. Karis L. Tekwani, from Advocate Christ Medical Center in Oak Lawn, Illinois, noted in an e-mail to Reuters Health.

"Our study found a non-significant 7% increase in mortality in patients given single bolus doses of etomidate," she said. "Another randomized controlled study comparing etomidate to ketamine for rapid sequence intubation by Jabre et al also found a similar non-significant 8% increase in mortality in patients with sepsis given single bolus doses of etomidate."

Therefore, Dr. Tekwani said, it is still possible that a larger study would detect a significant difference, "and hence we would urge clinicians to at least consider using alternative induction agents in the critically ill population at risk of adrenal suppression."

Dr. Tekwani and her colleagues had randomly assigned 122 critically ill patients with presumed sepsis to etomidate (0.3 mg/kg IV) or midazolam (0.1 mg/kg IV) before emergency department intubation at their hospital. The investigators were blinded to treatment assignment. Demographics and baseline characteristics were fairly similar between the two groups. The median patient age was 70 in the etomidate group, 73 in the midazolam group.

Two patients in the etomidate arm were lost to follow-up. In intention-to-treat analysis, the median number of days in the hospital (the primary outcome) was 7.3 in the etomidate group and 9.5 in the midazolam group.

There was also no significant difference in intent-to-treat analysis in the secondary outcome of ICU stay: 3.1 days in the etomidate arm vs 4.2 days in the midazolam arm. The same was true for number of ventilator days (median 2.1 with etomidate and 2.8 days with midazolam).

The in-hospital mortality rate was 43% (26 of 61) with etomidate and 36% (21 of 59) with midazolam. "Kaplan-Meier survival analysis likewise showed no statistically significant differences between groups (p = 0.22)," the authors note.

The decision to use steroids was left to the treating physician; 20 patients in the etomidate arm (33%) got steroids in the ED versus 28 (47%) in the midazolam arm. Overall, there were no significant effects from the use of steroids, "although trends toward increased mortality were evident with supplemental steroid use, irrespective of induction agent," they report.

Etomidate has been shown to cause "measurable adrenal suppression after a single bolus dose," the authors point out, although the clinical significance of this in patients with sepsis "remains controversial." The current study, however, was not designed to obtain quantitative measures of adrenal function.

Pulse oximetry results and systolic blood pressures were similar in the two groups, both before and after intubation; 25% of patients in each group had a systolic blood pressure < 90 mm Hg after intubation.

A per-protocol analysis of 45 etomidate- and 51 midazolam-treated subjects with confirmed sepsis yielded largely findings similar to the intention-to-treat analysis.

In their report, Dr. Tekwani and colleagues try to reconcile their findings with prior studies that hinted at increased harm with etomidate in patients with sepsis. They say, for instance, that to their knowledge theirs is the only randomized, double-blind comparison of etomidate to an alternative agent for intubation in patients with presumed sepsis.

And they cite several methodological problems with past studies. The current study, they say, "overcomes many of the limitations of previous observational studies and raises the possibility that previous findings were influenced by confounding variables."

Nevertheless, they admit, this study was relatively small. This, combined with the non-significant increase in mortality in the etomidate group, raises the possibility that there may be a clinically important difference in mortality after all.

Clearly, more study is needed, they conclude.

Ann Emerg Med. Published online September 10, 2010. Abstract

Reuters Health Information 2010. © 2010 Reuters Ltd.