Objective: To investigate the adrenal function in non-septic, long-stay critically ill patients.
Design: Prospective, consecutive study.
Setting: General intensive care unit in a university hospital.
Patients: Forty-three non-septic patients with protracted critical illness.
Interventions: A morning blood sample was first obtained to measure baseline plasma cortisol. Subsequently, 1 micro g of corticotropin (ACTH, Synacthene) was injected intravenously and 30 min later a second blood sample was drawn to determine stimulated plasma cortisol. Patients having a stimulated cortisol level of at least 18 micro g/dl were defined as responders. In 36 patients, morning interleukin-6 (IL-6) was also measured.
Measurements and results: Baseline and stimulated plasma cortisol were 16.8+/-4.1 micro g/dl and 21.2+/-5.1 micro g/dl, respectively. Interleukin-6 was high (median 39.3 pg/ml, interquartile range 24.9-86.6 pg/ml) and correlated negatively with stimulated plasma cortisol (r=-0.40, p<0.05). Of the 43 patients studied, 31 patients (72%) were responders and 12 patients (28%) were non-responders to the ACTH stimulation test. Overall, 18 patients died and 25 patients survived to hospital discharge. Non-responders had significantly higher IL-6 levels compared to responders (106+/-73 versus 48+/-42 pg/ml, p<0.05), whereas mortality rate was comparable in the two groups (50% versus 38%, p=0.74).
Conclusions: Circulating plasma IL-6 levels are high during protracted critical illness, and are partially responsible for the relative adrenal insufficiency found in a subset of severely ill patients.