Adrenocorticotropic hormone and cortisol response to corticotropin releasing hormone in the critically ill-a novel assessment of the hypothalamic-pituitary-adrenal axis

BACKGROUND: The pathophysiology of adrenal insufficiency, common in surgical intensive care units, has not been fully elucidated. METHODS: Patients at risk (age > 55 years, in the surgical intensive care unit > 1 week, baseline cortisol < 20 mu g/dL) were enrolled. After measuring cortisol and adrenocorticotropic hormone (ACTH), corticotropin-releasing hormone (CRH) was administered. ACTH and cortisol were measured over 120 minutes. Short and long cosyntropin stimulation tests determined adrenal function. Area under the curve (AUC) and mixed linear models were used to compare cortisol and ACTH responses. Patients were grouped according to survival and response to stimulation testing. Chi-square and t tests were performed, and P values < .05 were considered statistically significant. RESULTS: Six of 25 patients responded poorly to cosyntropin, and 5 died compared with 3 after a normal response (P < .01). ACTH (AUC) and ACTH peak were increased in nonsurvivors after CRH administration. Cortisol peak and AUC were not different. CONCLUSIONS: ACTH responsiveness was increased in nonsurvivors and may predict mortality. (C) 2012 Elsevier Inc. All rights reserved.