Gonadotropins and Gonadal Steroid Feedback

Recent studies indicate that the midcycle gonadotropin surge in the human occurs without an increase in hypothalamic gonadotropin-releasing hormone (GnRH) pulse frequency. In addition, previous studies employing a GnRH antagonist to provide a semiquantitative estimate of endogenous GnRH secretion suggest that the overall amount of GnRH secreted is decreased at the time of the surge. To investigate the hypothesis that a normal gonadotropin surge can be generated in the human with a decreased amount of GnRH at the midcycle, 7 GnRH-deficient subjects underwent two cycles of a physiologic regimen of intravenous pulsatile GnRH therapy. In the control cycle, 75 ng/kg/ bolus of GnRH, a dose known to be sufficient for folliculogenesis, was administered throughout the cycle, using physiological frequencies. In a second cycle, the bolus dose of GnRH was decreased by one-half log order to 25 ng/kg just prior to the luteinizing hormone surge and returned to 75 ng/kg after documented ovulation. All cycles were ovulatory. The peak luteinizing hormone level (77.4B9.7 vs. 67.5B17.6 IU/l) did not differ between the control and decreased GnRH cycles. There was no difference in the peak serum estradiol level (475.8B144.1 vs. 493.2B93.0 pg/ml), follicular phase length (15.0B1.3 vs. 14.8B0.6 days), or progesterone level (22.4B5.1 vs. 34.8B5.7 ng/ml) on day 6 of the luteal phase in the control and decreased GnRH cycles, respectively. Three pregnancies were achieved in each of the control and reduced GnRH cycles. We conclude that a decreased overall amount of GnRH generates a normal midcycle gonadotropin surge and has no significant impact on luteal phase adequacy or fertility. These results provide further evidence that a decrease in endogenous hypothalamic GnRH secretion may occur at the midcycle in normal women. This study also provides evidence that the GnRH requirements for normal follicular and luteal phase dynamics may well be greater than those required for generation of a normal midcycle gonadotropin surge and ovulation in women. OOOOOOOOOOOOOOOOOOOOOO Received: January 16, 1998 Accepted after revision: February 19, 1998 Kathryn A. Martin Reproductive Endocrine Unit, Bartlett Hall Massachusetts General Hospital, Fruit Street Boston, MA 02114 (USA) Tel. +1 (617) 726 3038, Fax +1 (617) 726 5357, E-Mail Martin.Kathryn@mgh.harvard.edu ABC Fax + 41 61 306 12 34 E-Mail karger@karger.ch www.karger.com © 1998 S. Karger AG, Basel This article is also accessible online at: http://BioMedNet.com/karger 364 Neuroendocrinology 1998;67:363–369 Martin/Welt/Taylor/Smith/Crowley/Hall Introduction The midcycle gonadotropin surge is the critical neuroendocrine event responsible for final maturation and rupture of the dominant follicle. It requires a complex coordination of signals from hypothalamus, pituitary, and ovary. There is overwhelming support for the presence of estrogen-induced augmentation of the pituitary response to gonadotropin-releasing hormone (GnRH) at the time of the preovulatory surge in all species examined [1–7]. However, the contribution of changes in hypothalamic GnRH secretion to the surge is less clear. In the rat [8, 9] ewe [10, 11], and nonhuman primate [12], a spontaneous preovulatory increase in GnRH secretion has been demonstrated at the time of the luteinizing hormone (LH) surge. Whereas this increase is essential for induction of an LH surge in the ewe [13] and the rat [14], it may play only a permissive role in the nonhuman primate [1–3, 15]. In the human, direct measurement of GnRH is impossible, and indirect techniques have been used to assess the hypothalamic changes in GnRH secretion at the time of the surge. To date, studies using a combination of these indirect techniques in normal and GnRH-deficient women provide evidence that the GnRH secretion is not increased at the midcycle as compared with other cycle stages [16–27]. Moreover, studies using a GnRH antagonist to provide a semiquantitative estimate of endogenous GnRH secretion suggest that GnRH may, in fact, be decreased at the surge relative to other cycle stages [28]. We have previously determined that 75 ng/kg/bolus GnRH, administered intravenously at frequencies mimicking those occurring in the normal menstrual cycle [18], replicates normal menstrual cycle gonadotropin and sex steroid dynamics as well as fertility in GnRH-deficient women [17] and that a dose of 25 ng/kg/bolus is clearly inadequate in reliably recreating normal ovulatory cycle dynamics [17, 23]. To test the hypothesis that a normal midcycle surge can be achieved with a decrease in the amount of hypothalamic GnRH secreted, GnRH-deficient women underwent two cycles of pulsatile GnRH treatment in which the dose of GnRH was varied. GnRH was administered at a dose of 75 ng/kg/bolus through the follicular and luteal phases of all cycles with a one-half log order decrease to 25 ng/kg/bolus through the midcycle in the experimental cycle. Our results demonstrate that in GnRH-deficient women treated with pulsatile GnRH in a manner designed to replicate normal follicular and luteal phase dynamics, a normal midcycle surge is generated despite a decrease in the overall amount of exogenous GnRH administered at the midcycle. Materials and Methods