Assessment of adrenocortical function in the critically ill.

OBJECTIVE To review current concepts in the diagnosis of adrenocortical disease in the critically ill patient. DATA SOURCES A review of articles reported on adrenocortical insufficiency in the acutely ill patient. SUMMARY OF REVIEW The contribution of adrenal insufficiency to the morbidity of critically ill patients is currently under renewed scrutiny. The debate continues about the role of steroids in sepsis and essentially the question remains unanswered. Central to this debate is the issue of whether adrenal insufficiency is common in the critically ill patient. What is incontrovertible is that adrenocortical function is essential for host survival during critical illness, but what constitutes adrenocortical insufficiency in critically ill patients is not clear. Absolute adrenocortical insufficiency (diagnosed by very low plasma cortisol concentrations) is uncommon in the intensive care population. The diagnosis of relative adrenocortical insufficiency (elevated basal plasma cortisol with a subnormal increase in plasma concentrations following an ACTH stimulus) continues to generate debate. The controversy surrounding the role of steroids in sepsis and the confusion over the criteria for diagnosing adrenal insufficiency in the critically ill are reviewed. CONCLUSIONS We suggest that the following caveats be borne in mind when diagnosing adrenal insufficiency in the critically ill patient. Firstly, the gold standard for the diagnosis has not been established. Secondly, caution must be exercised when interpreting a single plasma cortisol value. In the event of a single result indicating adrenal hypofunction, we suggest repeating the measurements after a 6 to 12 hour interval. The clinician must also be aware of variations in cortisol concentrations induced by the assay. Thirdly, the clinician must be aware of the potential limitations of the conventional high dose corticotrophin test. We also suggest that plasma free cortisol is more relevant than total plasma cortisol in the assessment of adrenal function in critical illness and that the low dose corticotrophin test is more sensitive than the conventional high dose test. These areas should be the subject of further investigations.