The utility of a random cortisol level in determining neonatal central adrenal insufficiency.

Abstract

While it has been established within the first 4 months of life that there is no circadian rhythm, what is unclear is the usefulness of a random serum cortisol (rSC) level in determining neonatal central adrenal insufficiency (CAI). The objective of the study is to determine the utility of using rSC in infants less than 4 months old in the evaluation of CAI. Retrospective chart review of infants who underwent a low dose cosyntropin stimulation test ≤4 months of life with rSC taken as baseline cortisol before stimulation. Infants were divided into three groups: those diagnosed with CAI, those at risk for CAI (ARF-CAI) and a non-CAI group. Mean rSC for each group was compared, and ROC analysis was used to identify rSC cut-off for the diagnosis of CAI. Two hundred and fifty one infants with the mean age of 50.5 ± 38.08 days, and 37% of these were born at term gestation. The mean rSC were lower in the CAI group (1.98 ± 1.88 mcg/dl) as compared to the ARF-CAI (6.27 ± 5.48 mcg/dl, p = .002) and non-CAI (4.6 ± 4.02 mcg/dl, p = .007) groups. ROC analysis identified a cut-off of rSC level of 5.6 mcg/dL is associated with 42.6% sensitivity and 100% specificity for the diagnosis of CAI in term infants. This study demonstrates that though anrSC can be used within the first 4 months of life, its value is best when done ≤30 days of life. Moreover, a diagnostic cut-off for CAI using rSC levels was identified for term infants.