Changes in serum cortisol with age in critically ill patients

Mortality in the intensive care unit (ICU) rises with age, a high basal serum cortisol and a small response to adrenocorticotropin (ACTH) stimulation. Even slight impairment of the adrenal response during severe illness can be lethal.

To determine if age is associated with changes in basal or stimulated serum cortisol in critically ill patients.

We studied 2 groups of surgical ICU patients with hypotension despite > or = 6 h of catecholamine therapy. Group 1 comprised 7 patients aged <30 (mean 22.9 +/- 3.7) years, and group 2 comprised 8 patients aged >60 (mean 75.8 +/- 10.3) years (p < 0.001). We compared baseline serum cortisol levels and the serum cortisol response 30 and 60 min after stimulation with low-dose (1 microg) and 2 h later standard-dose (250 microg) ACTH. We also determined the incidence of adrenal insufficiency in each group using standard criteria and compared selected clinical variables.

There was no significant difference in the mean serum cortisol at baseline although it tended to be higher in older patients. Group 2 patients had a significantly smaller response to the low-dose test at 30 min (p = 0.002), and to the standard-dose test at both 30 (p = 0.02) and 60 min (p = 0.04). There was no significant difference in the incidence of adrenal insufficiency between the 2 groups: 1/7 or 14.3% in group 1 vs. 1/8 or 12.5% in group 2 (p = 1.0). There was no significant difference between the 2 groups in the mean acute physiology score, blood pressure, serum albumin, dopamine, or dobutamine dose (p > 0.05). Creatinine clearance was significantly lower in group 2 (p > 0.001) and endogenous ACTH significantly higher (p = 0.04). Significantly more patients in group 1 (5/7, 72%) than group 2 (1/8, 12.5%) had a diagnosis of trauma on admission (p = 0.04). Seven of the eight patients (88%) in group 2 vs. 1/7 (14%) of patients in group 1 died in the ICU, but this difference was not statistically significant (p = 0.18).

In this small exploratory study, baseline serum cortisol tended to be higher in older patients and older patients had a significantly smaller response to ACTH stimulation on both low-dose and standard-dose tests. In view of the high death rate in the older ICU patients, the findings in this study need to be confirmed in a larger study. The mechanism and clinical significance of these findings remain to be determined but may be related to deterioration in renal function with age.