Young, healthy male patients undergoing body surface surgery under fentanyl-nitrous oxide-oxygen anesthesia were given an etomidate bolus for induction (0.3 mg/kg body weight) followed by a continuous infusion (0.36 mg.kg-1.h-1). The total etomidate dose was 63 +/- 6.4 mg (mean +/- standard error). Basal cortisol levels as well as the adrenocortical response to exogenous synthetic ACTH1-24 (0.25 mg i.m.) were determined preoperatively, at the end of anesthesia, and 6 and 20 h postoperatively. The reduction of the response to ACTH stimulation measured postoperatively as compared to preoperatively was calculated as percentage of the preoperative response. Plasma concentrations of etomidate were measured at the same sample times by gas chromatography. At the end of anesthesia the plasma level was 570 +/- 131 nM (mean +/- standard error). It decreased to 99 +/- 23 nM 6 h later and to 43 +/- 3 nM 20 h postoperatively. The plasma concentrations were plotted against the inhibition data. We found a sigmoid relationship between the relative inhibition and the log of the etomidate concentrations. We determined an EC50 for inhibition of 110 nM that agrees well with published data. We also found that the cortisol response was not completely suppressed at the concentrations measured here. This supports the results of a previous study in which we found that patients undergoing cardiovascular surgery still exhibited intra- and postoperative cortisol increases although they had been given large doses of etomidate. We conclude that the suppression of corticosteroid synthesis caused by etomidate is concentration-dependent. The suppression is negligible at concentrations lower than 40 nM.(ABSTRACT TRUNCATED AT 250 WORDS)