Prolonged critical illness is characterized by protein hypercatabolism and preservation of fat depots, associated with blunted GH secretion, elevated serum cortisol levels, and low insulin-like growth factor I (IGF-I) concentrations. In this condition, GH is readily released in response to a bolus of GHRH and GH-releasing peptide-2 (GHRP-2) and, paradoxically, to TRH. We further explored the altered somatotropic axis and cortisol secretion in critical illness by examining the effects of continuous GHRH and/or GHRP-2 infusion. Twenty-six critically ill adults (mean age +/- SEM, 63 +/- 2 yr) were studied during 2 consecutive nights (2100-0600 h). According to a weighed randomization, they received one of four combinations of infusions within a randomized cross-over design for each combination: placebo (one night) and GHRP-2 (the other night; n = 10), placebo and GHRH (n = 4), GHRH and GHRP-2 (n = 6), and GHRP-2 and GHRH plus GHRP-2 (n = 6). The peptide infusions (duration, 21 h) were started after a bolus of 1 microgram/kg at 0900 h and infused (1 microgram/kg/h) until 0600 h. Serum concentrations of GH were determined every 20 min, cortisol every hour, and IGF-I at 2100 and 0600 h on each study night. The placebo profiles showed pulsatile GH secretion with low secretory burst amplitude [0.062 +/- 0.008 microgram/L distribution volume (Lv)/min], high burst frequency (6.6 +/- 0.4 events/9 h), and detectable basal secretion (0.041 +/- 0.009 microgram/L/min) in the face of low serum IGF-I (106 +/- 11 micrograms/L). IGF-I correlated positively and significantly with the basal component, the pulsatile component, and the total amount of nightly GH secretion. GHRH elicited a 2- to 3-fold increase in the mean GH concentration (P = 0.006), the GH secretory burst amplitude (P = 0.007), and basal GH secretion (P = 0.03). GHRP-2 provoked a 4- to 6-fold increase in the mean GH concentration (P < 0.0001), the GH secretory burst amplitude (P = 0.002), and basal GH secretion (P = 0.0007), which were associated with a 61 +/- 13% increase in serum IGF-I within 24 h (P = 0.02). Compared to GHRP-2 alone, GHRH plus GHRP-2 elicited a further 2-fold increase in the mean GH concentration (P = 0.04) and GH basal secretion (P = 0.02), and an additional 40 +/- 6% rise in serum IGF-I (P = 0.04). GHRH and GHRP-2 infusion did not alter elevated cortisol levels. In critically ill adults, low serum IGF-I levels were positively correlated with diminished pulsatile and increased basal GH secretion. Both basal and pulsatile GH secretion were moderately increased by continuous infusion of GHRH, substantially increased by GHRP-2, and strikingly increased by GHRH plus GHRP-2. GHRP-2 alone or combined with GHRH elicited a robust rise in circulating IGF-I levels within 24 h without altering serum cortisol levels. These findings open perspectives for GH secretagogues as potential antagonists of the catabolic state in critical care medicine.