Free Cortisol and Accuracy of Total Cortisol Measurements in the Diagnosis of Adrenal Insufficiency in Brain-dead Patients

Abstract

After brain death, adrenal insufficiency (AI) is very common and may be one of the mechanisms that contributes to hemodynamic instability and loss of potential organ donors. However, when diagnosed by total cortisol measurement, critically ill patients may be overdiagnosed as having AI. The aims of this study were to assess the prevalence of AI when diagnosed using free cortisol measurement and the accuracy of total cortisol measurement to diagnose AI in brain-dead patients. All consecutive brain-dead patients were included in this single-center noninterventional clinical observation study. Assessment of adrenocorticotropin, corticosteroid-binding globulin, baseline and tetracosactin-stimulated serum free and total cortisol concentrations were performed. AI was defined as a baseline free cortisol concentration ≤ 55 nM(-1) and/or Δ free cortisol ≤ 55 nM(-1). Patients were considered to have a low albumin concentration if less than 25 g · L(-1) and a low corticosteroid-binding globulin concentration if less than 27 mg · L(-1) in men or 31 mg · L(-1) in women. Among the 42 included patients, the incidence of AI was 83% (95% CI, 69-93%). Baseline total cortisol was correlated with baseline free cortisol, whatever the albumin or corticosteroid-binding globulin concentration. The area under the receiver operating characteristic curve of baseline total cortisol measurement to diagnose AI was 0.94 (95% CI, 0.81-0.98). The optimal cutoff was 485 nM(-1), providing a sensitivity and a specificity of 89% and 100%, respectively. Total baseline cortisol measurement is accurate and sufficient to diagnose AI in brain-dead patients, even if albumin or corticosteroid-binding globulin concentrations are low.