UMEM Educational Pearls

As emergency physicians, we use etomidate to intubate patients most of the time, although there was controversy whether etomidate would suppress critically ill patients’ cortisol production. Whether etomidate was associated with mortality was controversial. A recent meta-analysis investigated the issue again.

Methods: meta-analysis of randomized trials using etomidate for intubation versus other agents. Outcome = mortality as defined by the authors. Mortality was defined from 24 hours to 30 days by study’s authors.

Results: 11 RCTs, including one new RCT in 2022

319 (1359, 23%) patients received etomidate died vs. 267 (1345, 20%) receiving other agents died; Risk Ratio 1.16, 95% CI 1.01-1.33, P = 0.03.

Etomidate was also associated with higher risk ratio for adrenal insufficiency, when compared with other control agents (147/695, 21% vs. 69/686, 10%, RR 2.01, 95% CI (1.59-2.56), P < 0.01.

Etomidate was also associated with higher risk ratio of mortality, when compared with ketamine, for mortality, as defined by each study’s author (273/1201, 23% vs. 226/1198. 19%. RR 1.18, 95% CI 1.02-1.37, P = 0.03).

Discussion:

The authors used fixed effects model, as they claimed that their meta-analysis had low heterogeneity (I2 =0%). However, fixed effects model should only be used when there is no difference among patient population. In this study, the outcome definitions were different, the patient populations were different (trauma, pre-hospital, ED, ICU). Therefore, random effects model should be used. Random effects models tend to yield larger 95% CI, thus, more likely yield non-statistically significant results.

The authors claimed a Number Needed To Treat (NNT) for etomidate of 31, so basically many ED patients would die, while most of patients being intubated by Anesthesiology, regarding settings, would not die, as anesthesiologists mostly use propofol.