We have asked whether or not sleep-associated 12-hour GH profiles were a clinically relevant and practicable tool to identify short children with low spontaneous GH secretion. In 67 prepubertal patients (19 girls and 48 boys, mean age 9.34 years, range 1.99-14.5) sleep-associated 12-hour GH profiles were obtained by drawing peripheral venous blood every 30 min over a 12-hour night period. The diagnosis of GH deficiency (GHD, n = 26), constitutional delay of puberty and growth (CDPG, n = 19), familial short stature (FSS, n = 8), GH neurosecretory dysfunction (GHND, n = 5), and constitutional delay of puberty and growth plus familial short stature (CDPGFSS, n = 9) was made by clinical parameters (SDS height range:-0.69 to -5.59, SDS growth velocity:-4.6 to -2.4) and provocative testing of GH secretion. Integrated GH secretion (area above baseline = AOB, area above zero line = AOOL), peak frequency, area under the peaks, peak amplitude length, peak amplitude height, maximal peak values, and median peak values were calculated using the PULSAR program. Significant differences of GH secretion between patient groups in regard to mean values for area over baseline, area over zero line, amplitude height, maximal peak values, and median peak values of secretion were found. However, there was a large interindividual variation of integrated GH secretion within each patient group and, most importantly, a large overlap between the different patient groups. We conclude that the assessment of pulsatile GH secretion during sleep, even if it can contribute to distinguish between different groups of short children, is not helpful to distinguish between different causes of short stature in an individual child. We suggest that measurement of sleep-associated spontaneous GH secretion needs to be restricted to research facilities.