The review included 18 RCTs (n=951 patients), one non-randomised prospective study (n=106 patients) and one retrospective study (n=159 patients). Jadad scores for RCTs ranged from 1 to 5; 14 RCTs scored 1 or 2 points, two scored 3 points and two scored 4 or 5 points. Two non-RCTs were at moderate to high risk of selection, performance and attrition bias, but at low risk for detection bias.
Cortisol levels (16 RCTs): Fourteen RCTs reported depressed cortisol levels post-induction with etomidate. Nine RCTs could be pooled.
For patients who were undergoing elective surgery and were graded ASA grade 1 or 2 (nine RCTs, n=201 patients), etomidate was associated with a significantly lower cortisol level than comparator groups at one hour (MD 6.1μg/dL, 95% CI 2.4 to 9.9), two hours (MD 13.3μg/dL, 95% CI 7.7 to 18.9), three hours (MD 12.6μg/dL, 95% CI 6.6 to 18.4) and four hours (MD 16.4μg/dL, 95% CI 9.7 to 23.1) after induction. There was no significant difference at five hours. Significant statistical heterogeneity was found (p<0.1) for all analyses. Grouping the trials by etomidate dose (lower and higher than standard doses) reduced the heterogeneity but did not significantly change results.
Three RCTs in elective patents that were not suitable for pooling showed consistent findings for cortisol in ASA I and III patients at or above baseline at seven and 12 hours post-induction.
It was not possible to pool data from four studies of critically-ill patients since studies assessed outcomes at different times. The largest RCT (n=655 patients with cortisol levels reported for 232) reported that etomidate was associated with significantly reduced cortisol levels compared with ketamine after a median of seven hours post-induction (16.0 versus 25.0μg/dL, p<0.001). Findings were similar in one of the three smaller RCTs at four to six hours; the other two RCTs reported no significant difference at 24 hours.
Mortality (six RCTs, one prospective observational study and one retrospective study): There was no significant difference in mortality between etomidate and comparator agents (five RCTs of 741 critically-ill patients).
Other outcomes: One of four RCTs reported that etomidate was associated with a longer period of ventilation, intensive care unit stay and hospital stay than comparator groups. One of three RCTs reported a longer hospital stay in the etomidate group.
Other results were also reported.