Relative adrenal insufficiency as a predictor of disease severity, mortality, and beneficial effects of corticosteroid treatment in septic shock

Abstract

Objective: To evaluate the concept of relative adrenal insufficiency necessitating corticosteroid therapy in septic shock.

Design: Retrospective study.

Setting: Medical-surgical intensive care unit of a university hospital.

Patients: We studied 218 consecutive patients with septic shock in a 3-yr period who underwent a short 250-microg adrenocorticotropic hormone test because of >6 hrs of hypotension requiring repeated fluid challenges and/or vasopressor/inotropic treatment.

Interventions: The test was performed by intravenously injecting 250 mug of synthetic adrenocorticotropic hormone and measuring cortisol immediately before and 30 and 60 mins postinjection.

Measurements and main results: Intensive care unit mortality until day 28 was 22%. Nonsurvivors had greater disease severity, as exemplified by higher Simplified Acute Physiology Score II and Sequential Organ Failure Assessment score, on the day of adrenocorticotropic hormone testing. Cortisol levels directly correlated with albumin levels. Simplified Acute Physiology Score II and Sequential Organ Failure Assessment score increased with higher strata of baseline cortisol/albumin or lower cortisol increases/albumin ratios as measures of free cortisol. Baseline cortisol, cortisol increases, and albumin levels did not independently contribute to mortality prediction by disease severity and absence of corticosteroid (hydrocortisone) treatment in a Cox proportional hazard model, although adrenocorticotropic hormone-induced cortisol increase <100 nmol/L (n = 53) predicted mortality (p = .007). Posttest treatment by corticosteroids (n = 161, 74%) was associated with higher survival in patients with cortisol increase <100 nmol/L (p = .0296).

Conclusions: In intensive care unit patients with septic shock, the cortisol response to adrenocorticotropic hormone inversely relates to disease severity, independent of blood cortisol binding. An adrenocorticotropic hormone-induced cortisol increase <100 nmol/L predicts mortality and beneficial effects of corticosteroid treatment. The data favor relative adrenal insufficiency.