"Relative" adrenal insufficiency in critical illness

To present a recommended approach to the problem of "relative" adrenal insufficiency (RAI) in the intensive care unit (ICU).

We examine historical data that support the traditional concepts of adrenal insufficiency and the idea that the increase in cortisol secretion during stress is needed to survive the stress. The controversial use of treatment with glucocorticoids (GCs) in patients with sepsis and septic shock in the ICU (and thus survival benefit) is also briefly discussed.

During the past decade, the concept of RAI as the failure of cortisol secretion to increase in response to stress to sustain the patient through that stress has gained strength. In some studies, it has been suggested that as many as 75% of patients in an ICU setting have RAI. Experimental support for the concept is not possible because there is no clinically useful laboratory measure of GC action. Therefore, diagnosis is generally based on interpretation of the cosyntropin stimulation test.

The best clinical judgment should always guide interpretation of any test results, and sharp categorization of patients on the basis of a single cutoff criterion should be avoided. Overall, the concept of RAI has no clinical utility. In these cases, administration of GCs adds cost without benefit and with increased risk.