PULSATILE GROWTH-HORMONE RELEASE IN NORMAL WOMEN DURING THE MENSTRUAL-CYCLE

OBJECTIVE We sought to characterize pulsatile growth hormone (GH) release in normal women during the menstrual cycle and to document possible relationships between such characteristics and concentrations of 17 beta-oestradiol and progesterone. SUBJECTS Fifteen women with ostensibly normal menstrual function were studied during the early follicular phase, 15 during the late follicular phase and 15 during the mid-luteal phase of the menstrual cycle. DESIGN The phase of the menstrual cycle having been documented, blood samples were obtained from each woman every 10 minutes for 24 hours. MEASUREMENTS Serum GH was measured in each sample by immunoradiometric assay. Pulsatile GH release was appraised utilizing the objective, statistically-based pulse detection algorithm Cluster. RESULTS The mean (+/-SEM) integrated serum GH concentration (mU/l min) in late follicular phase women (5335+/-848) was higher than that observed in early follicular phase women (3156+/-322; P=0.032). The integrated GH concentration calculated for mid-luteal phase women (3853+/-788) was intermediate between but not statistically different from that observed in early follicular (P=0.48) and late follicular (P=0.14) phase women. No differences in GH pulse frequency (pulses/24 hours) were found among early follicular (8.27+/-0.55), late follicular (7.93+/-0.91) or mid-luteal (8.47+/-0.66) phase women. Mean maximal GH pulse amplitude (mU/l) was higher in late follicular phase (8.93+/-1.00) than early follicular phase (5.74+/-0.67; P=0.008) and mid-luteal phase (5.76+/-0.74; P = 0.008) women. Similarly, incremental GH pulse amplitude (mU/l) was higher in late follicular phase (7-33+/-0.83) than early follicular phase (4.68+/-0.58; P=0.005) and mid-luteal phase (4.36+/-0.39; P=0.002) women. No differences in mean pulse widths or in the interpeak valley mean GH concentrations were found among the groups. Multiple regression of each pulse parameter against serum concentrations of testosterone, 17-beta-oestradiol and progesterone revealed a significant (P=0.045) positive correlation between maximum GH pulse amplitude and oestradiol and a significant (P=0.04) negative correlation between maximal GH pulse amplitude and progesterone (r = 0.41). CONCLUSION These results suggest that late follicular phase concentrations of oestradiol may enhance circulating GH via an amplitude-modulated rather than a frequency-modulated effect on the endogenous GH pulse. Progesterone may blunt this oestrogen-associated effect, thus resulting in the observed mid-luteal phase concentrations of GH. Whether these gonadal hormones act primarily at the hypothalamus and/or anterior pituitary gland remains to be clarified, but the present observations indicate that pulsatile GH release throughout the normal menstrual cycle is significantly amplitude regulated.