OBJECTIVES: To compare the safety and efficacy of ketamine and etomidate as induction agents to facilitate emergent endotracheal intubation.

DATA SOURCES: We searched MEDLINE, Embase, Cochrane Clinical Trials Register, and ClinicalTrials.gov from inception to April 3, 2024.

STUDY SELECTION: We included randomized controlled trials (RCTs) that compared ketamine to etomidate to facilitate emergent endotracheal intubation in adults.

DATA EXTRACTION: Reviewers screened abstracts, full texts, and extracted data independently and in duplicate. We pooled data using a random-effects model, assessed risk of bias using the modified Cochrane tool and certainty of evidence using the Grading Recommendations Assessment, Development, and Evaluation approach. We pre-registered the protocol on PROSPERO (CRD42023472450).

DATA SYNTHESIS: We included seven RCTs (n = 2384 patients). Based on pooled analysis, compared with etomidate, ketamine probably increases hemodynamic instability in the peri-intubation period (relative risk [RR], 1.29; 95% CI, 1.07-1.57; moderate certainty) but probably decreases the need for initiation of continuous infusion vasopressors (RR, 0.75; 95% CI, 0.57-1.00; moderate certainty) and results in less adrenal suppression (RR, 0.54; 95% CI, 0.45-0.66; moderate certainty). Ketamine probably has no effect on successful intubation on the first attempt (RR, 1.01; 95% CI, 0.97-1.05; moderate certainty) or organ dysfunction measured as the maximum Sequential Organ Failure Assessment (SOFA) score during the first 3 days in ICU (mean difference, 0.55 SOFA points lower; 95% CI, 1.12 lower to 0.03 higher; moderate certainty) and may have no effect on mortality (RR, 1.00; 95% CI, 0.83-1.21; low certainty) when compared with etomidate.

CONCLUSIONS: Compared with etomidate, ketamine probably results in more hemodynamic instability during the peri-intubation period and appears to have no effect on successful intubation on the first attempt or mortality. However, ketamine results in decreased need for the initiation of vasopressor use and decreases adrenal suppression compared with etomidate.

Emergency Medicine rater

In this SR/MA, there was no difference between etomidate and ketamine as an induction agent on first-pass success rate, organ dysfunction (as per max SOFA score), and mortality. Interestingly, there is some increase in hemodynamic instability with ketamine (RR 1.29), but paradoxically less need for vasopressors (RR 0.75). The hemodynamic instability is consistent with other recent studies, although would go against traditional teaching. The increase in adrenal insufficiency with etomidate is not surprising and consistent with other publications, although it's effect from a patient-centric standpoint is unclear and probably insignificant. Regardless, either etomidate or ketamine can be used for induction prior to intubation without a difference in clinical outcomes. Further studies may be able to delineate cohorts where one agent may be better than another, but that is not clear based on this study alone.

Respirology/Pulmonology rater

The choice of an induction drug in emergent endotracheal intubation has long been controversial. This review/meta-analysis helps clarify this, but only modestly. My read of the results is that ketamine, in general, is no worse than etomdiate and may be desirable in certain subsets. Overall, the results do not materially change my clinical practice.

Respirology/Pulmonology rater

This is a good attempt but a poorly conducted meta-analysis. Indirectness due to the various definitions of hemodynamic instability is very severe and the GRADE Rating should be downgraded by two levels. Cointerventions used during rapid sequence intubation, such as fentanyl and propofol, were not considered, which negatively impacts generalizability. Using different neuromuscular blockers should also be considered as a reason for downgrading due to indirectness. Although the attempt to compare etomidate and ketamine should be commended, the current evidence does not provide solid conclusions due to clinical heterogeneity across the included studies.