Urinary growth hormone (GH) excretion was quantitated in 12-h overnight urine collections obtained from 31 control children, ages 3 to 17 yr (group 1); 21 children, ages 5 to 19 yr with GH deficiency (group 2), and 30 subjects, ages 10 to 18 yr with idiopathic growth failure and normal GH stimulation tests (group 3). The output of urinary GH was measured in one acromegalic woman. The authenticity of urinary GH, 22 kDa, was confirmed by high-performance liquid chromatography. The elution pattern of urinary GH was identical to that of biosynthetic and pituitary-derived GH. The immunoreactive profiles characterized by monoclonal immunoradiometric GH assay and standard GH radioimmunoassay were identical. The quantity of GH (mean ± SEM per kg body weight) in group 1 (0.27 ± 0.02 ng/kg) was significantly greater than group 2 (0.08 ± 0.02 ng/kg) or group 3 (0.17 ± 0.02 ng/kg,p< 0.01). Approximately 50% of the subjects in group 3 had urinary GH measurements indistinguishable from those observed in the GH-deficient population. Twelve hypopituitary patients (group 2) excreted significantly greater amounts of urinary GH in the first 12 h after GH administration compared to the baseline period (0.41 ± 0.07versus0.12 ± 0.02 ng/kg,p< 0.01). Markedly elevated output of urinary GH (2.0 ng/kg) was documented in one acromegalic patient. The data suggest that measurements of urinary GH may be a useful, simple, and noninvasive screening test for identifying patients with GH deficiency or excess.