Effect of Etomidate vs Propofol for Total Intravenous Anesthesia on Major Postoperative Complications in Older Patients: A Randomized Clinical Trial

IMPORTANCE: Older patients may benefit from the hemodynamic stability of etomidate for general anesthesia. However, it remains uncertain whether the potential for adrenocortical suppression with etomidate may increase morbidity. OBJECTIVE: To test the primary hypothesis that etomidate vs propofol for anesthesia does not increase in-hospital morbidity after abdominal surgery in older patients. DESIGN, SETTING, AND PARTICIPANTS: This multicenter, parallel-group, noninferiority randomized clinical trial (Etomidate vs Propofol for In-hospital Complications [EPIC]) was conducted between August 15, 2017, and November 20, 2020, at 22 tertiary hospitals in China. Participants were aged 65 to 80 years and were scheduled for elective abdominal surgery. Patients and outcome assessors were blinded to group allocation. Data analysis followed a modified intention-to-treat principle. INTERVENTIONS: Patients were randomized 1:1 to receive either etomidate or propofol for general anesthesia by target-controlled infusion. MAIN OUTCOMES AND MEASURES: Primary outcome was a composite of major in-hospital postoperative complications (with a noninferiority margin of 3%). Secondary outcomes included intraoperative hemodynamic measurements; postoperative adrenocortical hormone levels; self-reported postoperative pain, nausea, and vomiting; and mortality at postoperative months 6 and 12. RESULTS: A total of 1944 participants were randomized, of whom 1917 (98.6%) completed the trial. Patients were randomized to the etomidate group (n = 967; mean [SD] age, 70.3 [4.0] years; 578 men [59.8%]) or propofol group (n = 950; mean [SD] age, 70.6 [4.2] years; 533 men [56.1%]). The primary end point occurred in 90 of 967 patients (9.3%) in the etomidate group and 83 of 950 patients (8.7%) in the propofol group, which met the noninferiority criterion (risk difference [RD], 0.6%; 95% CI, –1.6% to 2.7%; P = .66). In the etomidate group, mean (SD) cortisol levels were lower at the end of surgery (4.8 [2.7] μg/dL vs 6.1 [3.4] μg/dL; P < .001), and mean (SD) aldosterone levels were lower at the end of surgery (0.13 [0.05] ng/dL vs 0.15 [0.07] ng/dL; P = .02) and on postoperative day 1 (0.14 [0.04] ng/dL vs 0.16 [0.06] ng/dL; P = .001) compared with the propofol group. No difference in mortality was observed between the etomidate and propofol groups at postoperative month 6 (2.2% vs 3.0%; RD, –0.8%; 95% CI, –2.2% to 0.7%) and 12 (3.3% vs 3.9%; RD, –0.6%; 95% CI, –2.3% to 1.0%). More patients had pneumonia in the etomidate group than in the propofol group (2.0% vs 0.3%; RD, 1.7%; 95% CI, 0.7% to 2.8%; P = .001). Results were consistent in the per-protocol population. CONCLUSIONS AND RELEVANCE: Results of this trial showed that, compared with propofol, etomidate anesthesia did not increase overall major in-hospital morbidity after abdominal surgery in older patients, although it induced transient adrenocortical suppression. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT02910206