Sadly, the jury is still out. Just months after Critical Care Medicine published the systematic review demonstrating an association between use of etomidate and mortality in sepsis, now they're back with a retrospective data-mining expedition that draws the opposite conclusion.
This is a multi-center prospective registry of critically ill patients entered into a research database who were retrospectively data-mined for septic, intubated patients. Of the 42,000 patients in the database, approximately 2,000 met this definition, and about half were identified as receiving etomidate as their induction agent. In their cohort, there was no in-hospital mortality difference between the patients who received etomidate and the patients who received a different induction agent for intubation.
Unfortunately, as an observational, retrospective study of imperfectly matched cohorts, there are far too many uncontrolled confounders to base clinical practice on these findings. Studies such as these, even robust, prospective cohorts, are capable of doing little more than suggesting a hypothesis contrary to the findings of prior work.
If you believe etomidate has a chance to harm patients in sepsis, this doesn't change your practice.
"Single-Dose Etomidate Is Not Associated With
Increased Mortality in ICU Patients With Sepsis:
Analysis of a Large Electronic ICU Database"