Adrenal response in children with septic shock.

OBJECTIVE: To describe the serum cortisol profile and evaluate the adrenal response in children with septic shock, and determine the influence of these factors on the outcome and mortality in this group. METHODS: Between May and November 2003, 22 children with septic shock admitted to two pediatric intensive care units in southern Brazil were followed. Adrenal function was evaluated based on the levels of cortisol measured on the occasion of the diagnosis of septic shock and on the response of serum cortisol 30 min after the administration of intravenous corticotrophin (0.5 microg/1.73m(2)). Adrenal insufficiency was defined as a baseline serum cortisol below 690 nmol/l and/or a cortisol response to corticotrophin less than 250 nmol/l. RESULTS: Adrenal insufficiency was detected in 17 patients (77.3%). All patients who died had baseline cortisol higher than 690 nmol/l. A cortisol response to corticotrophin less than 250 nmol/l was associated with a 60% mortality (RR = 7.2, 1.03-50.28). Regression analysis showed that the combination of baseline cortisol higher than 690 nmol/l and a cortisol response to corticotrophin less than 250 nmol/l were associated with mortality after correction for gender and PRISM. CONCLUSIONS: Adrenal insufficiency is a frequent finding in children with septic shock. The low-dose corticotrophin stimulation test seems to be an important tool to distinguish between a normal cortisol response to stress and evidence of adrenal failure. Mortality was significantly higher in children that failed to respond to a corticotrophin stimulation test.