Modulatory role of estrogens and progestins on growth hormone episodic release in women with hypothalamic amenorrhea *

Transcription

1 FERTILITY AND STERILITY Copyright 1993 The American Fertility Society Vol. 60, No.3, September 1993 Printed on acid-free paper in U. s. A. Modulatory role of estrogens and progestins on growth hormone episodic release in women with hypothalamic amenorrhea * Alessandro D. Genazzani, M.D.t:j: Felice Petraglia, M.D.t Cristina Volpogni, M,D,t Mario Gastaldi, M.D.t Francesco Pianazzi, M.D.t Vanna Montanini, M.D. Andrea R. Genazzani, M.D. t University of Modena, Modena, Italy Objective: To define the characteristics of spontaneous GH episodic secretion and the modulatory role of gonadal steroids in patients with hypothalamic amenorrhea associated with weight loss. Design: Women were studied for 8 hours, sampling every 10 minutes, and plasma G H levels were measured by RIA. Subjects: Fifteen patients with weight-loss-related amenorrhea were studied in baseline conditions. Five out of 15 patients underwent two cycles of hormonal replacement therapy with E2 patches (100 Jl,g every 3 days for 24 days) and medroxyprogesterone acetate (MPA) (10 mg/d, from day 12 to day 24). On the second cycle of therapy, the pulsatility study was repeated twice: after only estrogen (day 11) and after E2 plus progestin (day 22). Four normally cycling women were studied as a reference group during midfollicular and midluteal phases. Results: Amenorrheic patients showed mean plasma GH levels similar to healthy women during the follicular phase but significantly lower than those observed during the luteal phase. GH pulse frequency was higher in patients than in controls, whereas pulse amplitude was comparable with the follicular phase but lower during the luteal phase. During the hormonal replacement therapy, when only E2 was administered, GH pulse frequency decreased, whereas GH integrated plasma concentrations and GH pulse amplitude increased significantly. After MPA and E2 administration, GH pulse amplitude and GH plasma levels decreased, which was similar to pretreatment condition. Conclusions: The present study demonstrated that in amenorrhea associated with weight loss the frequency of GH episodic release is significantly higher than in normally cycling women. Moreover, a different modulatory role of estrogen (increased amplitude) and P (decreased amplitude) on the episodic release of GH in amenorrheic women undergoing a replacement treatment was shown by the present data. Fertil Steril 1993;60: Key Words: GH, pulsatile secretion, hypothalamic amenorrhea, hormonal replacement therapy Hypothalamic amenorrhea is a model of hypogonadism, which is characterized by several neuroendocrine aberrations, affecting the release of several hypophyseal hormones (1). The possible Received February 8, 1993; revised and accepted June 1, * Supported in part by the CNR (National Research Council, Rome, Italy) program "AItri Interventi" AI (A.D.G.). t Department of Obstetrics and Gynecology. :j: Reprint requests: Alessandro D. Genazzani, M.D., Department of Obstetrics and Gynecology, University of Modena, Via del Pozzo 71, Modena, Italy. Department of Endocrinology. dependence of these disregulations from abnormal activity of some central neuromodulators is suggested. The use of naloxone, an opioid receptor blocker, increases LH pulse frequency and amplitude in healthy women but not in amenorrheic patients (2-4). In addition, the use of me to clop rami de determined the increase of LH plasma levels in patients affected by amenorrhea but not in normally cycling women (2), suggesting that opiatergic and dopaminergic pathways are probably involved in such neuroendocrine disregulations. Gonadal steroids are also important in the modulation of the Vol. 60, No.3, September 1993 Genazzani et al. GH and amenorrhea 465

2 release of hypophyseal hormones in women, as reported for gonadotropins (5,6), GH (7,8), and PRL (9). Recent reports demonstrated that amenorrheic patients have altered LH (10) and PRL (1) secretory patterns. The present study aimed to evaluate GH pulse characteristics in a group of patients with weight- loss-related amenorrhea. These patients were also studied under hormonal replacement therapy to define the modulatory role of gonadal steroids on GH episodic release. Subjects MATERIALS AND METHODS Fifteen amenorrheic patients were enrolled for this study. Criteria for the inclusion were as follows: [1] absence of menstrual cycles for ~6 months before the study, [2] normal plasma levels of adrenal cortex and thyroid hormones, [3] history of weight loss in the last 18 months, [4] body weight lower than the ideal body weight (IBW) (-21% below IBW) and no weight gain during the 6 weeks preceding the study, [5] absence of depression or psychiatric diseases assessed according to Diagnostical Statistical Manual-III (Revised) criteria. Four normally cycling women were used as a control group. None ofthem showed any weight loss in the last 6 months before the study, and their weight was 57 ± 2.8 kg (means ± SEM), 4.5% above the estimated IBW. Five out of 15 amenorrheic patients underwent two cycles of hormonal replacement therapy (24 days each at 7-day intervals) with the following schedule: E2 patches (100 Jlg, Estroclim; SigmaTau, Pomezia, Italy) replaced twice a week for 24 days (day 1 to day 24) and medroxyprogesterone acetate (MPA) 10 mg/d from day 12 to day 24. All amenorrheic subjects underwent a pulsatility study after an overnight fast. The five patients undergoing hormonal replacement therapy repeated a pulsatility study on day 11 and on day 22 of the second cycle of substitutive therapy. Controls were studied twice: during the midfollicular (days +6 and + 10) and the midluteal (days + 19 and +23) phases. In all subjects a heparin well was inserted into an antecubital vein 1 hour before commencing venous sampling at 10-minute intervals for 8 hours (from 8:00 A.M. to 4:00 P.M.). Blood samples were immediately centrifuged and stored frozen at - C until assayed. All subjects gave informed consent, and the study 466 Genazzani et at. GH and amenorrhea protocol was approved by the Human Investigation Committee of the University of Modena. Assays All samples from the same subject were analyzed by RIA in duplicate in the same assay. Plasma GH concentrations were determined using a commercially available kit (Radim; Pomezia, Rome, Italy). Intra-assay and interassay variability were 5% and 8.9%, respectively. The evaluation ofthe variability of G H assays was determined on replicates from a serum pool from the same individual assayed together with the time series, as previously described (10, 11). Prolactin, E2, and P plasma concentrations were determined in two different plasma specimens using commercially available RIAs whose intra-assay and interassay coefficients of variation were 4.1 % and 6.8% for PRL, 4.9% and 8% for E2, and 5% and 7.8% for P, respectively. Statistical Analysis Presence of significant difference between groups was tested, after one-way analysis of variance, using Student's t-test for paired and unpaired data, as appropriate. Variance Model and Pulse Detection The measurement error was evaluated from the best of five variance models estimated separately by the program PREDETEC.WK1 (12) on the duplicates for each time series. PREDETEC also provides the coefficients of the best variance model to be used in program DETECT (13), as previously described (12). Both programs have been developed at the Laboratory of Theoretical and Physical Biology, National Institutes of Health, Bethesda, Maryland. The presence of the significant GH pulses was determined using the program DETECT (13), whose detection logic and characteristics already have been described (12, 13). Data from each subject were processed with P set to 0.01 (1 %) for the nominal false-positive rate. RESULTS Hormonal characteristics of all subjects participating in the study are summarized in Table 1. When comparing integrated mean plasma concentrations, amenorrheic subjects showed plasma GH Fertility and Sterility

3 Table 1 Hormonal Characteristics of Study Subjects* GH No. peaks in 8 hours Duration Amplitude E2 P Amenorrhea (n = 15) Follicular phase (n = 4) Luteal phase (n = 4) "gil 3.5 ± ± ± ± 0.3t 4.0 ± ± 0.6 min "gil pmolll nmolll 61.5 ± ± ± ± ± ± ± 1.61 ± ± ± ± ± * Values are means ± SEM. t P < 0.05 vs. follicular and luteal phases. + P < 0.05 vs. follicular phase and amenorrheic patients. P < 0.01 vs. amenorrheic patients. II P < 0.01 vs. follicular phase and amenorrheic patients. levels similar to the follicular phase but lower than during the luteal phase (P < 0.01) of eumenorrheic women (Table 1). Both healthy women and amenorrheic patients demonstrated distinct episodic secretion of GH. Figures 1 and 2 show the GH secretory profiles for two amenorrheic patients and for one normal control, respectively. Amenorrheic patients showed a GH pulse frequency (5.8 ± 0.3 peaks/8 hours [means ± SEM], P < 0.05) higher than controls in both phases of the cycle (follicular phase 4 ± 0.4, luteal phase 4 ± 0.6 peaks/8 hours). Growth hormone pulse amplitude in amenorrheic patients was similar to that observed in controls during the follicular phase but significantly lower than during the luteal phase (P < 0.01; Table 1). During E2 administration amenorrheic patients showed a decreased G H pulse frequency (from 5.7 ± 0.2 to 4 ± 0.6 peaks/8 hours; P < 0.05). When MPA was associated to E2, GH pulse frequency did not show any change (Fig. 3). Integrated GH concentrations and GH pulse amplitude during the E2 replacement therapy was higher (P < 0.05 and P < 0.01, respectively) than in baseline conditions and during E2 plus MPA administration (Table 2). DISCUSSION The present study demonstrated the presence of a modified episodic release of GH in women with amenorrhea associated with weight loss. Moreover, we demonstrated the modulatory role of gonadal steroids on GH pulsatile secretion in amenorrheic women under hormonal replacement therapy. Previous studies reported that hypothalamic amenorrhea has a modified LH pulsatile release (10, 14), and when weight loss is associated, an abnormal thermoregulation (15), reduced and delayed response to GnRH and thyrotropin-releasing hormone standard tests (16), and lack of diurnal variation of cortisol (17) have been shown. These abnormalities are correlated with the severity of weight Vol. 60, No.3, September 1993 loss, and a more profound neuroendocrine dysfunction occurs in women with anorexia nervosa (16). In agreement with a previous report (1), the integrated mean G H plasma concentrations of HYPOTHALAMIC AMENORRHEA GH IIgIL, o GH I19/L, o Figure 1. Plasma GH pulsatile profile of two amenorrheic patients. Arrows indicate significant secretory episodes found by program DETECT. Genazzani et al. GH and amenorrhea 467

4 FOLLICULAR PHASE GH /loll , 8 A stress-induced phenomenon. In fact, stress increases G H release in humans, and it also might affect GH pulse frequency. However, it cannot exclude a reduction of the inhibitory tone of somatostatin on GH secretion. HYPOTHALAMIC AMENORRHEA o A GH "gil + LUTEAL PHASE 30 GH /loll B... o B 30.-~ ~ o Figure 2. Plasma GH pulsatile profile of one normally menstruating woman during the follicular (A) and luteal phases (B). Arrows indicate significant secretory episodes found by program DETECT. amenorrheic patients were similar to the values observed in controls during the follicular phase but significantly lower than during the luteal phase. Even if characterized by a significant weight reduction, our patients were not anorexic. This can explain why they showed a normal rather than an increased GH integrated plasma concentrations, as has been reported for anorexic women (18) as well as after fasting (19). Amenorrheic women showed a GH pulse frequency higher than eumenorrheic women. Because GH pulsatile release depends on GH- releasing hormone (RH) secretion from the hypothalamus, our data support the hypothesis of an increased frequency of endogenous G H -RH discharge from the hypothalamus, which might be dependent on a c 5 o GH "gil -UNDER E ~ o UNDER E2+MAP Figure 3. Plasma GH secretory pattern for one amenorrheic patient in baseline conditions (A), during E2 administration (B), and during E2 plus MPA administration (C). GH pulse frequency decreases and amplitude increased under E2 treatment. When MPA was added to E 2, GH pulse frequency remained unaltered whereas pulse amplitude decreased. Arrows indicate significant GH pulses found by program DETECT. 468 Genazzani et al. GH and amenorrhea Fertility and Sterility

5 Table 2 GH Pulsatile Characteristics in Five Amenorrheic Patients in Baseline Conditions and Under Hormonal Replacement Therapy* GH I'lI/L Before 3.3 ± 0.5 UnderE2 4.5 ± 0.6t Under E2 + MPA 3.0 ± 0.6 Values are means ± SEM. t p < 0.05 vs. other groups. Peaks in 8 hours 5.7 ± 0.2t 4.0 ± ± 0.2 Duration Amplitude E2 min I'lI/L pmol/l 56.9 ± ± ± ± ± 1.6:1: 7.7 ± ± ± ± 23.8 :1: P < 0.01 vs. other groups. When considering the amplitude, amenorrheic subjects showed an amplitude similar to the eumenorrheic women during the follicular phase but significantly lower than the values observed during the luteal phase ofthe menstrual cycle. In fact, control women during the luteal phase had a significant increase of GH pulse amplitude, with no changes in pulse frequency. These data confirm previous studies that reported the change of amplitude but not of the frequency of G H episodic secretion throughout the menstrual cycle (7) and the presence of a significant correlation between G H pulse amplitude and estrogen plasma levels during the late follicular phase, the periovulatory period, and the luteal phase (7, 8). In the present study we tested the modulatory role of gonadal steroids on GH episodic discharge in amenorrheic patients. Our present data demonstrated that E2 determines a significant increase in both the integrated mean GH plasma concentrations and G H pulse amplitude, reducing the pulse frequency. These results suggested that E2 modulates GH episodic release, probably either enhancing the response of G H secreting cells to endogenous G H -RH or increasing the amount of G H -RH released per single secretory burst. This last hypothesis could also reflect the reduced frequency of GH pulses observed in amenorrheic patients. However, in vitro experiments (, 21) and data obtained in humans (22) support the hypothesis of specific direct modulatory role of E2 on pituitary cells. Therefore, the lack of appropriate estrogen plasma levels in amenorrheic women may explain the low GH pulse amplitude observed in these patients. Accordingly, oral hormonal replacement therapy determines an increase of GH pulse amplitude and mean plasma concentrations in postmenopausal women (23, 24). However, our results differ from those reported by Bellantoni et al. (), who observed neither the increase of GH nor the decrease of insulin-like growth factor I (IGF-I) plasma concentrations after transdermal estrogen replacement schedule in postmenopausal women. Because this therapeutic schedule did not modify plasma IGF-I and GH levels (), they concluded that this result was due to the fact that transdermal substitutive therapy delivers E2 directly into the peripheral circulation and does not pass through the liver. Our data seem to exclude this hypothesis, even if IGF-I plasma levels were not assayed. When a progestin (i.e., MPA) was associated with estrogens, G H pulsatile secretion was significantly reduced in pulse amplitude and in plasma mean integrated concentrations. MP A, as progestin, counteracts the effects induced by E 2, determining the reduction of GH pulse amplitude and thus reducing GH integrated plasma concentrations. This does not correspond to the increased GH plasma levels and GH pulse amplitude observed in healthy controls during the midluteal phase when compared with the midfollicular phase. The discrepancy between exogenous progestin effects in hypothalamic amenorrhea and endogenous P effects in healthy controls on integrated GH plasma concentrations and G H pulse characteristics may represent a further neuroendocrine aberration of hypothalamic amenorrhea. Moreover, this fact also may become more apparent by the constant levels of estrogens observed in amenorrheic women during the hormonal therapy when compared with the menstrual phases of normal women. In conclusion, hypothalamic amenorrhea associated with weight loss is characterized by an increased frequency and a reduced amplitude of GH pulsatile release. Moreover, transdermal estrogen administration reversed this situation, suggesting the positive role of E2 in the modulation of endogenous GH secretion. The finding that MPA counter acted in amenorrheic patients the positive effects induced by E2 differenciates these patients from healthy controls where luteal phase is characterized by an enhancement of both GH integrated plasma levels and GH pulse amplitude. These data suggest Vol. 60, No.3, September 1993 Genazzani et al. GH and amenorrhea 469

6 the presence of an abnormal modulation of GH discharge in hypothalamic amenorrhea. Acknowledgments. We are grateful to David Rodbard, M.D., Division of Computer Research and Technology, National Institutes of Health, Bethesda, Maryland, for kindly providing the program DETECT. REFERENCES 1. Berga SL, Mortola SF, Girton L, Suh B, Laughlin G, Pham P, et al. Neuroendocrine aberrations in women with functional hypothalamic amenorrhea. J Clin Endocrinol Metab 1989;68: Quigley ME, Sheehan KL, Casper RF, Yen SSC. Evidence for increased dopaminergic and opioid activity in patients with hypothalamic hypogonadotropic amenorrhea. J Clin Endocrinol Metab 1980;50: Khoury SA, Reame NE, Kelch RP, Marschall JC. Diurnal patterns of pulsatile luteinizing hormone secretion in hypothalamic amenorrhea: reproducibility and responses to opiate blockade and an a,-adrenergic agonist. J Clin Endocrinol Metab 1987;64: Wildt L, Leyendecker G. Induction of ovulation by the chronic administration of naltrexone in hypothalamic amenorrhea. J Clin Endocrinol Metab 1987;64: Genazzani AD, Rodbard D, Forti G, Petraglia F, Baraghini GF, Genazzani AR. Estimation of instantaneous secretory rate of luteinizing hormone in women during the menstrual cycle and in men. Clin Endocrinol (Oxf) 1990;32: Cagnacci A, Melis GB, Paoletti AM, Gambacciani M, Soldani R, Spinetti A, et al. Influence of oestradiol and progesterone on pulsatile LH secretion in postmenopausal women. Clin Endocrinol (Oxf) 1989;31: Faria ACS, Bekenstein LW, Booth RA, Vaccaro VA, Asplin CM, Veldhuis JD, et al. Pulsatile growth hormone in normal women during the menstrual cycle. Clin Endocrinol (Oxf) 1992;36: Genazzani AR, Lemarchand-Beraud T, Aubert ML, Felber JP. Pattern of plasma ACTH, hgh, and cortisol during menstrual cycle. J Clin Endocrinol Metab 1975;41: Clifton DK, Aksel S, Bremner WJ, Steiner RA, Soules MR. Statistical evaluation of coincident prolactin and luteinizing hormone pulses during the normal menstrual cycle. J Clin Endocrinol Metab 1988;67: Genazzani AD, Petraglia F, Fabbri G, Monzani A, Montanini V, Genazzani AR. Evidence of luteinizing hormone secretion in hypothalamic amenorrhea associated with weight loss. Fertil Steril 1990;54: Genazzani AD, Petraglia F, Benatti R, Montanini V, Algeri I, Volpe A, et al. Luteinizing hormone (LH) secretory burst duration is independent from LH, prolactin, or gonadal steroid plasma levels in amenorrhei women. J Clin Endocrinol Metab 1991;72: Genazzani AD. Application of peak-detection programs to clinical data. In: Guardabasso V, Rodbard D, Forti G, editors. Computers in endocrinology: recent advances. New York: Raven Press, 1990: Oerter KE, Guardabasso V, Rodabrd D. Detection and characterization of peaks and estimation of instantaneous secretory rate for episodic pulsatile hormone secretion. Computers Biomed Res 1986;19: Crowley WF, Filicori M, Spratt DI, Santoro NF. The physiology of gonadotropin -releasing hormone (GnRH) secretion in men and women. Recent Prog Horm Res 1985;41: Vigersky RA, Andersen AE, Thompson RH, Lauriaux DL. Hypothalamic dysfunction in secondary amenorrhea associated with simple weight loss. N Engl J Med 1977;297: Vigersky RA, Lauriaux DL, Andersen AE. Anorexia nervosa: behavioral and hypothalamic aspects. Clin Endocrinol (Oxf) 1976;5: Wennik JMB, Delemarre Van De Waal HA, Van Kassel H, Mulder GH, Foster JP, Shoemaker J. Luteinizing hormone secretion patterns in boys at the onset of puberty measured using a highly sensitive immunoradiometric assay. J Clin Endocrinol Metab 1988;67: Muller EE, Locatelli V. Undernutrition and pituitary function: relevance to the pathophysiology of some neuroendocrine alterations of anorexia nervosa. J Endocrinol 1992;132: Hartman ML, Veldhuis JD, Johnson ML, Lee MM, Alberti KGMM, Samojlik E, et al. Augmented growth hormone (GH) secretory burst frequency and amplitude mediate enhanced GH secretion suring a two- day fast in normal men. J Clin Endocrinol Metab 1992;74: Evans WS, Krieg RJ, Limber ER, Kaiser DL, Thorner MO. Effect of in vivo gonadal hormone environment on in vitro hgrf-40-stimulated GH release. Am J Physiol 1985;249 :E Simard J, Hubert JF, Hosseinzadech T, Labrie F. Stimulation of growth hormone release and synthesis by estrogens in rat anterior pituitary cells in culture. Endocrinology 1986;119: Ho KY, Evans WS, Blizzard RM, Veldhuis JD, Merriam G R, Samokli KE, et al. Effects of sex and age on the 24-hour profile of growth hormone secretion in man: importance of endogenous estradiol concentrations. J Clin Endocrinol Metab 1987;64: Dawson-Hughes B, Stern D, Goldman J, Rechlin S. Regulation of growth hormone and somatomedin-c secretion in postmenopausal women: effect of physiological estrogen replacement therapy. J Clin Endocrinol Metab 1986;63: Mercuri N, Petraglia F, Genazzani AD, Amato F, Maietta Latessa A, De Leo V, et al. Hormonal treatments modulate plasma growth- hormone, gonadotropin and osteocalcin levels in postmenopausal women. Maturitas. In press.. Bellantoni MF, Harman SM, Cho DE, Blagman MR. Effects of progestin -opposed transdermal estrogen administration on growth hormone and insuline-like growth factor-i in postmenopausal women of different ages. J Clin Endocrinol Metab 1991;72: Genazzani et al. GH and amenorrhea Fertility and Sterility