OBJECTIVES: Adrenal dysfunction has been documented in critically ill adult populations and in children with septic shock and may have serious consequences if undiagnosed. However, there is no information available on adrenal function in critically ill children with various underlying conditions. DESIGN: A prospective, descriptive study was carried out to investigate the occurrence of adrenal dysfunction in pediatric critical illness. PATIENTS: Thirteen patients were consecutively recruited from a pediatric critical care unit with Pediatric Risk of Mortality Scores (PRISM; illness severity scores) >/=10 and hemodynamic instability as defined by the need for inotropes and/or >20 mL/kg fluid in a 24-hr period. INTERVENTIONS: Baseline cortisol and adrenocorticotropic hormone (ACTH) levels were measured followed by post-ACTH stimulation test cortisol levels to determine adrenal reserve. MEASUREMENTS AND MAIN RESULTS: Baseline cortisol levels ranged from 0.10 to 37.4 &mgr;g/dL (mean, 12.2 +/- 9.6). Adrenal dysfunction as defined by a baseline cortisol level of <7 &mgr;g/dL (am cortisol) or a post-ACTH stimulation level of <18 &mgr;g/dL was observed in four of the 13 patients (31%; 95% confidence interval, 5.7%,55.9%). All three of the patients with baseline cortisol levels of <7 &mgr;g/dL had low-normal ACTH levels. There was no difference in baseline PRISM scores (14.7 +/- 4.11, 12.7 +/- 2.65; p =.37) or admitting diagnoses for those with and without adrenal dysfunction. CONCLUSIONS: We conclude that some critically ill children do no mount adequate basal cortisol levels in the face of severe physiologic stress but do respond to external ACTH stimulation to better delineate the hypothalamic-pituitary-adrenal axis in critically ill children and to determine the clinical consequences of these biochemical abnormalities.