Risk Factors and Outcome of Changes in Adrenal Response to ACTH in the Course of Critical Illness

Abstract

To evaluate the concept of critical illness-related corticosteroid insufficiency (CIRCI) by studying the clinical significance, in terms of risk factors and outcome, of changes in the cortisol response to repeated adrenocorticotropic hormone (ACTH) testing in the course of critical illness. In a retrospective study in a medical-surgical intensive care unit (ICU) of a university hospital, we retrospectively included 54 consecutive patients during a 3-year period, who underwent 2 conventional 250 μg ACTH tests at an interval >24 hours, because of ≥6 hours hypotension requiring repeated fluid challenges or vasopressor/inotropic treatment, while corticosteroid treatment was not (yet) initiated. Serum cortisol was measured immediately before and 30 and 60 minutes after intravenous injection of 250 μg of ACTH. Patients were divided into those with an increase (≥0, n = 27) or a decrease (n = 27) in time in delta (Δ) cortisol in response to ACTH and with a Δcortisol <100 (n = 11) and ≥100 nmol/L (n = 43) at the second ACTH test. Changes in Δcortisol in time were paralleled by changes in Δcortisol/albumin, with a higher frequency of septic shock, persistently high disease severity, increased renal replacement therapy, and decreased platelet counts in the course of disease with a decrease in Δcortisol in time. Similar trends in increased disease severity were observed when Δcortisol remained or fell to <100 nmol/L. A decrease in Δcortisol between the 2 tests, particularly to <100 nmol/L, was associated with increased mortality (18 nonsurvivors in the ICU). The findings favor the concept of dynamic adrenal function rather than poor reproducibility of the ACTH test, so that development of CIRCI, particularly in complicated septic shock and indicated by a fall in Δcortisol (to <100 nmol/L) upon ACTH, correlates to a poor prognosis, independently of baseline cortisol, cortisol binding in blood, and disease severity.