CIRCADIAN VARIATION OF GROWTH-HORMONE SECRETION IN NORMAL PREPUBERTALCHILDREN - COMPARISON TO CONSTITUTIONAL GROWTH DELAY AND GROWTH-HORMONE DEFICIENCY
Rc. Hermida et al., CIRCADIAN VARIATION OF GROWTH-HORMONE SECRETION IN NORMAL PREPUBERTALCHILDREN - COMPARISON TO CONSTITUTIONAL GROWTH DELAY AND GROWTH-HORMONE DEFICIENCY, Journal of endocrinological investigation, 19(5), 1996, pp. 273-283
When studying short slowly growing children, several investigators hav
e found differences in spontaneous circulating growth hormone (GH) lev
els in some short children as compared with children of normal stature
. The circadian and ultradian characteristics of GH secretion have not
been considered in most of those studies. With the aim to study possi
ble differences in rhythm characteristics of plasma GH with stature, w
e analyzed data from a total of 141 prepubertal children: 1) 13 GH-def
icient children; 2) 36 children with short stature (up to 2 standard d
eviations below their peer group mean); 3) 61 children with very short
stature (up to 4 standard deviations below their peer group mean); an
d 4) a reference group of 31 children with standard stature. Subjects
were living at the hospital setting on a diurnal waking (07:30 a.m. to
10:30 p.m.), nocturnal resting routine during sampling, consuming the
usual hospital diet at fixed times. GH concentrations were determined
by immunoradiometric assay in plasma obtained at about 2-3 hour inter
vals during most of the day and at about half-hour intervals between 9
:00 p.m. and 3:00 a.m. Circadian rhythm characteristics obtained by le
ast-squares estimation were compared between groups divided according
to gender and stature with a parameter test. Results show a statistica
lly significant circadian rhythm in GH secretion for all groups studie
d (p<0.001 for standard, short and very short children; p=0.013 for GH
-deficient children). A comparison of circadian parameters indicates s
imilar characteristics between subjects of short, very short and stand
ard stature. The difference in rhythm-adjusted mean and amplitude is,
however, of borderline statistically significance when comparing non-d
eficient with GH-deficient children.