Résumé : Introduction: Intravenous etomidate infusion is effective to rapidly lower cortisol levels in severe Cushing's syndrome (CS) in the intensive care unit (ICU). Recently, etomidate treatment has also been proposed at lower doses in non-ICU wards, but it is not yet clear how this approach compares to ICU treatment. Methods: We compared data from patients with severe CS treated with high starting doses of etomidate (median: 0.30 mg/kg BW/day) in ICU or with lower starting doses (median: 0.025 mg/kg BW/day) in non-ICU medical wards. Results: Fourteen patients were included, among which ten were treated with low starting doses (LD) and four with high starting doses etomidate (HD). All patients had severe and complicated CS related to adrenal carcinoma (n = 8) or ectopic ACTH secretion (n = 6). Etomidate was effective in reducing cortisol levels below 500 nmol/L in a median of 1 day in the HD group compared to 3 days in the LD group (P = 0.013). However, all patients of the HD group had etomidate-induced cortisol insufficiency and needed frequent monitoring, while no patient from the LD group required hydrocortisone supplementation. No patient in either group died from complications of CS or etomidate treatment, but final outcome was poor as six patients in the LD group and all four patients in the HD group died from their cancer during follow-up. Conclusion: Our study suggests that, for patients with severe CS who do not require intensive organ-supporting therapy, the use of very low doses of etomidate in medical wards should be considered.