Postoperative adrenocortical function was compared in 23 out-patients receiving either thiopental, 4 mg/kg, for induction and a thiopental infusion, 0.26 mg . kg-1 . min-1, in combination with nitrous oxide 70% for maintenance of anesthesia (control); etomidate, 0.4 mg/kg, for induction followed by an etomidate infusion, 0.02 mg . kg-1 . min-1, and nitrous oxide 70% for maintenance (etomidate I); or etomidate, 0.4 mg/kg, for induction and a thiopental infusion, 0.22 mg . kg-1 . min-1, in combination with nitrous oxide 70% for maintenance (etomidate II). The norepinephrine response to anesthesia and surgery did not differ significantly between the three groups. The postoperative cortisol response to ACTH stimulation was normal in the control group (maximum rise in plasma cortisol was 20.1 +/- 2.9 micrograms/dl [mean +/- SEM] ), however, it was decreased in all patients receiving etomidate, whether by a short infusion (mean change in plasma cortisol was -3.8 +/- 1.9 micrograms/dl) or as a single induction dose (mean change in plasma cortisol was -4.0 +/- 2.0 micrograms/dl). Similarly, the postoperative aldosterone levels in the control group increased normally in response to ACTH (+ 10.2 +/- 3.0 ng/dl) but decreased in both the etomidate I and etomidate II groups (-3.0 +/- 0.7 ng/dl and -3.3 +/- 1.0 ng/dl, respectively). Because ACTH was administered exogenously, etomidate-induced suppression of adrenocortical response appeared to be a direct effect on the adrenal gland, which was present at a time when the serum etomidate levels were in the subhypnotic range.(ABSTRACT TRUNCATED AT 250 WORDS)