REPRODUCIBILITY OF 24-H GROWTH-HORMONE PROFILES IN CHILDREN

The rate of GH secretion and the pattern of GH peaks were compared in a group of nine prepubertal children during their prepubertal period in repeated 24-h GH profiles. At investigation, the children were 6-13 years old (at first profile 6-11 years old) and of normal height (+/- 2 SD). Two profiles were obtained per child, (with a mean time interval of 1.5 years, range 0.7 to 3.5 years. The calculated GH secretions of the first and second profiles were compared. As a group, no significant differences in secreted amount of GH, when expressed as data from the second profile as a percentage of data from the first profile (93 +/- 8%), number of peaks (98 +/- 7%) or mean peak amplitudes (92 +/- 11%), were obtained. Between the repeated curves of an individual child, maximal difference in secretion, number of peaks and mean peak amplitudes ranged around +/- 30%, with a mean intraindividual cv of 12%. The reproducibility in the peak distribution for all profiles was also analysed. The relative frequencies is a percentage of the GH peak amplitudes and peak widths were virtually identical in the repeated profiles. Reproducibility of the temporal pattern of profiles was analysed using time-series analysis (Fourier analysis) and showed no difference in rhythmicity between the different occasions. In conclusion, a high reproducibility of both GH secretion and GH pattern was found for the whole group of prepubertal children. The high degree of reproducibility of the 24-h GH profiles of the whole group of children indicated that the information from these curves, in terms of both pattern and total secretion, can be used for clinical as well as for physiological purposes. The intraindividual reproducibility was less pronounced, however, leading to a sound scepticism when relating biological phenomena to a single profile of an individual child.