Cortisol changes among patients with septic shock and the relationship to ICU and hospital stay

Objective: To evaluate adrenal function in patients with severe sepsis or septic shock. Design: A prospective study of unstimulated and high-dose ACTH stimulated cortisol levels on days 1 and 2 following ICU admission and day 28 or last day of hospitalization (herein day 28). Setting: General intensive care unit. Patients: 34 septic patients. Interventions: On days 1, 2, and 28 of sepsis unstimulated and ACTH stimulated cortisol levels were evaluated. End-points were length of ICU and hospital stay and 28 day all cause mortality. Measurements and results: Eight patients on days 1 and 2 had criteria of adrenal insufficiency defined as unstimulated cortisol levels less than 15 mu g/dl. Patients with unstimulated cortisol levels less than 15 mu g/dl on day 1 or change of 9 mu g/dl or less had longer ICU stays. Patients with unstimulated cortisol levels less than 15 mu g/dl on day 1 also demonstrated longer hospital stay. On day 1 regression analysis revealed that unstimulated cortisol levels had a significant inverse correlation with length of ICU and hospital stay. The proportional change between unstimulated cortisol and post-ACTH cortisol (Delta%) is a new modality; the higher this Delta% value, the longer is the patient's ICU and hospital stay. A significant positive correlation was found in survivors when comparing unstimulated cortisol vs. day 28 changes. Conclusions: The present study highlights the wide range of cortisol levels among patients with sepsis. We observed a difference in cortisol response pattern between survivors and nonsurvivors on day 28. The proportional change between unstimulated cortisol and post-ACTH cortisol was used as a method to evaluate the relative change in cortisol levels between patients.