The short ACTH test is used in evaluating the hypothalamo-pituitary-adrenal axis (HPA-axis) in preterm neonates after dexamethasone treatment. This test mainly examines primary adrenal suppression but is also used as a method to test secondary adrenal insufficiency because long-term deprivation of ACTH causes atrophy of the adrenal cortex. The CRH test, on the other hand, directly examines the function of the pituitary. We tested 18 infants in the neonatal intensive care unit with both the ACTH test and the CRH test to determine which of these two tests more reliably demonstrates HPA-axis suppression. One patient had normal responses both in the ACTH test and in the CRH test when the limit of 360 nmol/L was used as a sign of proper cortisol secretion. In four cases the patients' cortisol secretion would have been regarded as normal by the low-dose ACTH test, whereas the CRH test did not show an adequate cortisol response. In conclusion, the ACTH test did not reliably indicate HPA-axis suppression after a short (<2 weeks) course of dexamethasone therapy in this study. Therefore, whether the infant is or will be under acute stress after short glucocorticoid treatment, ensuring adequate cortisol secretion with the CRH test should be considered.