ABSTRACT
CONTEXT: Dehydroepiandrosterone (DHEA) replacement in sepsis has been advocated because of the sepsis-associated decrease in serum DHEA sulfate (DHEAS). However, experimental sepsis in rodents leads to down-regulation of DHEA sulfotransferase, which inactivates DHEA to DHEAS, theoretically resulting in higher DHEA levels. OBJECTIVE: The objective of the study was to test whether serum DHEA and DHEAS are dissociated in septic shock and to determine their association with circulating cortisol in the context of severity of disease and mortality. DESIGN, SETTING, AND PARTICIPANTS: This was a cross-sectional study consisting of 181 patients with septic shock, 31 patients with acute trauma, and 60 healthy controls. MAIN OUTCOME MEASURES: Serum cortisol, DHEA, and DHEAS were measured before and 60 min after ACTH stimulation. RESULTS: Serum cortisol was increased and DHEAS was decreased in both septic shock and trauma patients (all P < 0.001). However, compared with healthy controls, DHEA was significantly increased in sepsis but decreased after trauma (all P < 0.001). In sepsis, neither cortisol nor DHEA increased significantly after ACTH. Most severely ill patients had higher cortisol (P = 0.069) and lower DHEA (P = 0.076) and a significantly higher cortisol to DHEA ratio (P = 0.004). Similarly, the cortisol to DHEA ratio was significantly increased in nonsurvivors of septic shock (P = 0.026), whereas survivors did not differ from controls (P = 0.322). CONCLUSIONS: The observed dissociation of DHEA and DHEAS in septic shock contradicts the previous concept of sepsis-associated DHEA deficiency. Increased DHEA levels may maintain the balance between glucocorticoid- and DHEA-mediated immune and vascular effects. However, most severe disease and mortality is associated with an increased cortisol to DHEA ratio, which may represent a novel prognostic marker in septic shock.