Hypercortisolism in septic shock is not suppressible by dexamethasone infusion.

Objective: To explore the feedback regulation of glucocorticoids on corticotropin secretion in patients with septic and nonseptic circulatory shock.

Design: Prospective study.

Setting: An intensive care unit of a general hospital.

Patients: Two groups of patients with septic shock (n = 11) or nonseptic shock (n = 7). A control group (n = 20) was also studied.

Interventions: Intravenous dexamethasone (1 mg/hr for 4 hrs) suppression test.

Measurements: Plasma concentrations of corticotropin-releasing factor, beta-lipotropin, and corticosteroid-binding globulin measured by radioimmunoassays, and plasma cortisol measured by radiocompetition assay; the ratio of cortisol to corticosteroid-binding globulin calculated as the free cortisol index.

Main Results: In both groups of patients, the concentrations of plasma cortisol and beta-lipotropin, and the ratio of cortisol to corticosteroid-binding globulin, were higher than normal subjects (p < .001) and were not different between septic and nonseptic shock patients, whereas the plasma corticosteroid-binding globulin concentration was significantly (p < .001) lower in septic shock patients than in normal subjects (444 +/- 154 vs. 696 +/- 56 nmol/L [22.0 +/- 7.6 vs. 34.5 +/- 2.8 mg/L]), but not significantly lower in nonseptic shock patients (607 +/- 157 nmol/L [30.0 +/- 7.8 mg/L]). In contrast to the complete suppressive effect of dexamethasone infusion on cortisol and beta-lipotropin concentrations in normal subjects, dexamethasone did not suppress cortisol or lipotropin in either septic or nonseptic shock patients.

Conclusions: During circulatory shock, hypercortisolism is associated with high concentrations of lipotropin, and is not suppressible by intravenous dexamethasone infusion.