We sought to determine whether elevated circulating growth hormone (GH) concentrations in uremic prepubertal children are due to an increase in GH secretory activity by the pituitary gland or a decrease in the metabolic clearance of GH consequent to reduced GFR. Deconvolution analysis was applied to the nighttime plasma GH profiles of 1 ) 11 children with preterminal chronic renal failure, 2) 12 children with end-stage renal disease (ESRD), and 3) a control group of matched children with idiopathic short stature (n = 12). Mean (t SEM) half-life of endogenous GH in children with ESRD (27.5 ? 2.7 min) and preterminal chronic renal failure (23.1 ? 2.1 min) was significantly higher than in controls (14.8 ? 1.6 min; p < 0.001). GH half-life correlated inversely with GFR (r = -0.65, p < 0.001).The number of GH secretory bursts110 h in ESRD (8.1 ? 0.4) was amplified compared with preterminal chronic renal failure (6.4 2 0.5) and with controls (5.9 ? 0.4; p < 0.005). GH production rate varied over a broad range in the three groups: It was highest in ESRD (202 ? 56.6 mg/L/10 h; range 36-683), mainly as a result of an increased number of GH secretory bursts, and not statistically different in preterminal chronic renal failure (66.2 ? 11.4 mg/L/lO h; range 25-168) and in controls (129 ? 27.7 mg/L/10 h; range 39-392). Increased GH half-life, in concert with an increased GH production in some individuals with ESRD, leads to a 2.5-fold increase in the mean plasma GH concentration in ESRD compared with the two other groups ( p < 0.005). However, total immunoreactive plasma IGF-I levels were indistinguishable between groups. This disruption of the normal relationship between circulating GH and total plasma IGF-I levels in ESRD suggests a relative insensitivity to the action of GH in uremia, at least in those target organs (e.g. liver) that contribute predominantly to circulating IGF-I levels. regulation of the frequency and amplitude of pulsatile GH secretion, cannot be quantitated accurately in single blood determinations. Measurements of pulsatile plasma GH concentrations have been performed in prepubertal (4) and pubertal (5) children with CRF. Unfortunately, both studies lacked suitable control groups and did not provide insights into neurosecretory events and concurrent hormone clearance that together underlie the plasma GH pulse patterns. Such information is important particularly in patients with CRF, because the kidney is believed to account for a substantial fraction of the total plasma turnover of GH in humans (25-53%) (6) and in rats (67%) (7). Thus, in children with CRF, the question remains unresolved whether increased plasma GH concentrations are primarily due to increased GH secretion or decreased GH elimination by the kidney.