To evaluate the pituitary-adrenal reserve and to standardize the methodology of performing the human CRH (hCRH) stimulation test, we performed the hCRH test on 14 preterm (< 32 gestational weeks), very low birth weight infants, who did not receive antenatal or postnatal corticosteroid treatment, on days 7 and 14 of life. Blood samples were obtained 0 (baseline), 15, 30, and 60 min after an iv dose of hCRH (1 microgram/kg). The plasma ACTH concentration rose from a basal value of 5.7 +/- 0.6 pmol/L (mean +/- SEM) to 11.9 +/- 2.1 pmol/L (P < 0.005), 9.2 +/- 1.2 pmol/L (P < 0.005), and 7.7 +/- 0.8 pmol/L (P < 0.005) at 15, 30, and 60 min, respectively. The corresponding rises in serum cortisol from a basal concentration of 396 +/- 67 nmol/L were 509 +/- 71 nmol/L (P < 0.0001), 647 +/- 62 nmol/L (P < 0.0001), and 578 +/- 60 nmol/L (P < 0.0001). The plasma ACTH concentration consistently peaked early at 15 min, whereas the maximum cortisol response occurred 30 min post-hCRH stimulation. No significant differences were detected between the hCRH tests performed on days 7 and 14 (P > 0.15). Mechanical ventilation, infant gender, and mode of delivery did not significantly influence the hormonal responses (P > 0.25). We have defined in this study the pattern, the magnitude of the pituitary-adrenal response, and the timing of the peak concentrations of plasma ACTH and serum cortisol in relation to a standard iv dose of hCRH. The hCRH test in very low birth weight infants appears to be safe and reproducible, and produces a pituitary-adrenal response comparable to that seen in older children and adults, indicating that pituitary-adrenal function is mature at these early stages of gestation.