طب اورژانس

This was a prospective, observational cohort study conducted without randomization. The study ran from February 2007 to October 2007 at a large, community tertiary care hospital in suburban Chicago, Illinois. The study included all patients older than 18 years who were intubated in the emergency department, had suspected or known infection, and who had two or more criteria for systemic inflammatory response syndrome. Criteria included tachycardia, tachypnea, hypercapnia, leukocytosis, and fever. Severity of illness was assessed by the Mortality in Emergency Department Sepsis (MEDS) score. This is a validated system that is based on data that can be easily obtained in the emergency department. The primary outcome measure was in-hospital mortality. The secondary outcome measure was LOS.

This study enrolled 106 patients: 74 received etomidate, 27 received other induction agents, and 5 received no agent. The most common alternative agent was a benzodiazepine. Baseline characteristics of the two groups were similar including age and gender. In-hospital mortality for etomidate-treated patients was 38% (95% CI 28%-49%) compared with 44% (95% CI 28%-61%) for patients receiving other agents. Mortality rate for patients receiving no induction was 20%. With regard to the secondary outcome of LOS, LOS for etomidate-treated patients was 8 days (interquartile range [IQR] 3-13] compared with 6.5 days (IQR 3-9.75) for patients not treated with etomidate (P = .18). In the etomidate-treated group, 62% (46 of 74) of patients survived to discharge compared with 56% (18 of 32) of patients receiving other agents or no induction. For surviving patients, LOS was 10 days (IQR 7-16.25) in etomidate-treated patients compared with 7.5 days (IQR 4.75-10.5) for patients not receiving treatment with etomidate (P = .08). Based on multivariate analysis of mortality, the only factor that remained a significant predictor of mortality was mean arterial blood pressure at time of intubation; for LOS, the only significant predictor was pressor use.

The authors conclude that this study did not show significant differences in mortality or LOS between patients receiving etomidate and patients receiving other induction agents. They note a statistically nonsignificant trend toward increased LOS among patients given etomidate. They further conclude that the data from this study do not support stopping use of etomidate for induction but that additional data are needed. Limitations of the study include that randomization was not performed, that care after admission was not standardized, and that the study was done at a single center. The authors also note that sensitivity analysis of the data showed that the results were not robust.