Growth hormone response to clonidine in anovulatory infertile women resistant to clomiphene citrate stimulation

Transcription

1 FERTILITY AND STERILITY VOL. 73, NO. 1, JANUARY 2000 Copyright 1999 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. Growth hormone response to clonidine in anovulatory infertile women resistant to clomiphene citrate stimulation Plinio Rossato, M.D.,* Francesco Minuto, M.D., Simona Garrone, M.D., and Nicola Ragni, M.D. Ospedale Evangelico Internazionale, and University of Genova, Genova, Italy Objective: To evaluate the GH response to the clonidine test in a group of infertile women and to determine their ovulatory response to clomiphene citrate (CC) stimulation. Design: Prospective study. Setting: Reproductive endocrinology unit. Patient(s): Thirty-three anovulatory infertile women (age range, years) and 9 healthy controls with normal ovulation. Intervention(s): In the early follicular phase, 0.3 mg of clonidine was administered between 8:30 and 9:00 A.M. and blood samples were collected for 120 minutes thereafter for measurement of serum GH levels. Plasma levels of insulin and glucose were measured after a 75-g glucose load, and CC was given at a dosage of mg/d for ovulation induction. Main Outcome Measure(s): Serum concentrations of GH, insulin-like growth factor I, insulin, and insulinlike growth factor binding protein-1. Result(s): On the basis of their ovulatory response to CC, 15 patients were considered nonresponsive (group 1) and 18 patients were considered responsive (group 2). Baseline levels of GH, insulin-like growth factor I, and insulin-like growth factor binding protein-1 were similar in the two groups of patients and the controls. The GH response to clonidine was significantly greater in group 2 and in the controls than in group 1. Concentrations of insulin and glucose after the glucose load were not different among the three groups. Conclusion(s): Women who were resistant to CC had a reduced GH response to clonidine. These data suggest that adequate GH secretory capacity is important for CC action. (Fertil Steril 2000;73: by American Society for Reproductive Medicine.) Key Words: GH, clomiphene citrate, clonidine, anovulatory infertility Received March 23, 1999; revised and accepted August 2, Reprint requests: Plinio Rossato, M.D., Ospedale Evangelico Internazionale, Salita Superiore San Rocchino 31/A, Genova, Italy (FAX: ). * Division of Obstetrics and Gynecology, Ospedale Evangelico Internazionale. Department of Endocrinology and Metabolism, University of Genova. Department of Obstetrics and Gynecology, University of Genova /99/$20.00 PII S (99) In infertile, anovulatory women, the initial treatment aimed at inducing ovulation usually involves the administration of an antiestrogen such as clomiphene citrate (CC). Although its precise mechanism of action remains largely unknown, the administration of CC is followed by an increase in the secretion of pituitary gonadotropins that results in follicular recruitment, the growth of one or more follicles, and ovulation (1). The use of CC is associated with ovulation rates of 70% 90% (2). This means that 10% 30% of patients do not respond to treatment with CC. Some of these nonresponding patients have elevated serum FSH, prolactin, or DHEAS levels (3). Patients with polycystic ovary syndrome (PCOS) or obesity appear to be resistant to CC therapy (4). Even nonobese women and women apparently free of other known endocrine abnormalities may be resistant to CC therapy. In most cases, the reason for this lack of response to CC stimulation is not known. Recent studies suggest a relation between GH and gonadal function. Human granulosa cells possess GH binding sites, and GH stimulates E 2 production in cultures of human granulosa cells (5). Menashe et al. (6) observed that the ovarian response to hmg stimulation is correlated with the GH response to clonidine stimulation. Ovesen et al. (7) demonstrated that anovulation and infertility are associated with an impaired secretory capacity for GH. Because there is no discrete evidence that CC activity is dependent on adequate GH se- 78

2 cretion, we designed the present study to assess whether the level of GH reserve is correlated with the ovulatory responsive to CC administration. MATERIALS AND METHODS Subjects Informed written consent was obtained before the study was initiated as part of an Institutional Review Board approved protocol. We recruited 33 women who had acyclic spontaneous bleeding and anovulatory infertility and were seeking ovulation induction. Their cycle length in the previous 6 months varied from days. Anovulation was diagnosed by basal body temperature and persistent low progesterone levels measured at 4- to 6-day intervals beginning on day 14 after the onset of bleeding. Endometrial biopsies performed at least 22 days after the last bleeding episode confirmed anovulation. Each subject underwent clinical and endocrinologic investigations, and her ovarian morphology was evaluated by ultrasonography. All patients had normal thyroid function and normal levels of LH, FSH, prolactin, testosterone, androstenedione, cortisol, and DHEAS as measured by conventional RIA. No patient had taken hormonal preparations for at least 3 months before the study. Women who had evidence of polycystic ovaries at ultrasound examination, hirsutism, hyperandrogenism, or an LH/FSH ratio of 2 were excluded. Nine women with a mean ( SD) age of years (range, years) and normal menstrual cycles served as controls. They all had regular menstrual cycles of days in length. Ovulation was confirmed in the controls by midluteal serum progesterone levels and basal body temperature graphs. All the women had normal ovaries on ultrasound examination. No woman in any of the three groups engaged in aerobic exercise during the study period. Study Design After an overnight fast, each woman received an oral dose of clonidine hydrochloride (0.3 mg, Catapresan; Boehringer, Reggello, Italy) on cycle days 5 7 after a spontaneous bleeding episode. All tests were begun between 8:30 A.M. and 9:00 A.M. and at least 30 minutes after the cannulation of a cubital vein, which was kept patent with a slow infusion of normal saline solution. Blood samples were drawn every 30 minutes starting 30 minutes before the administration of clonidine and continuing for 120 minutes thereafter. One week later, all women underwent an oral glucose tolerance test in which they received a 75-g glucose load after fasting for hours overnight. Blood was sampled at 0, 30, 60, 90, 120, and 180 minutes for determination of plasma glucose and serum insulin levels. Ovulation induction was started with 50 mg/d of CC administered from the fifth to the ninth days of the following cycle. The response was considered ovulatory when growth of a preovulatory follicle was observed by ultrasound, followed by a rise in serum progesterone levels in the midluteal phase. In patients who did not ovulate, the daily dose of CC was increased by 50 mg in each successive cycle up to a total daily dose of 250 mg. If no ovulation occurred, the patient was considered unresponsive to CC. Hormone Assays Insulin-like growth factor-i (IGF-I) was measured with a double-antibody RIA using immunochemicals and a tracer provided by Medgenix (Fleurus, Belgium). The sensitivity of the assay was 150 pg/ml; the intraassay and interassay coefficients of variation were 6% and 7.5%, respectively. Insulin-like growth factor binding protein-1 (IGFBP-1) was measured with an immunoradiometric assay using reagents and a tracer provided by Diagnostic Systems Laboratories, Inc. (Webster, TX). The sensitivity of the assay was 125 pg/ml; the intraassay and interassay coefficients of variation were 2.5% and 4.6%, respectively. Serum concentrations of the other hormones were determined using commercial RIA kits (Sorin, Saluggia, Italy). Intraassay and interassay coefficients of variation were 4.6% and 9.2% for GH, 5.3% and 6.7% for insulin, 5.8% and 7.4% for 17 -E 2, 5.5% and 9.3% for LH, 6.8 and 8.7% for FSH, 6.3% and 9.4% for testosterone, 6.1% and 9.3% for androstenedione, and 6.9% and 8.7% for sex hormone-binding globulin, respectively. Glucose levels were determined using the glucose oxidase technique. Statistical Analysis Each variable was expressed as the mean SD. The release of GH after clonidine administration and the responses of insulin and glucose to the glucose load were expressed as the area under the curve (AUC; calculated by the linear trapezoidal method). The data were analyzed by the Student s t test, the Mann- Whitney U test, analysis of variance for multiple comparisons, and Spearman s rank correlation coefficient. P.05 was considered statistically significant. RESULTS Based on their ovulatory response to CC, the patients were divided into two groups. Group 1 included 15 women who were not responsive to CC stimulation. The patients had a mean ( SD) age of years (range, years). Thirteen of the patients were of normal weight (body mass index [BMI], kg/m 2 ), 2 were overweight (BMI, kg/m 2 ), and none were obese (BMI, 29.5 kg/ m 2 ). The mean ( SD) BMI of the entire group was kg/m 2. Group 2 was composed of 18 patients who ovulated after the administration of CC. They had a mean ( SD) age of FERTILITY & STERILITY 79

3 TABLE 1 Clinical and endocrine characteristics of the anovulatory women and healthy controls. Variable Group 1 (CC-resistant) Group 2 (CC-responsive) Controls P value Age (y) NS BMI (kg/m 2 ) NS Cycle length (d) * GH level ( g/l) NS Glucose level (mmol/l) NS Insulin level (pmol/l) NS E 2 level (pmol/l) NS LH level (miu/ml) NS FSH level (miu/ml) NS Testosterone level (nmol/l) NS Androstenedione level (nmol/l) NS DHEAS level ( mol/l) NS SHBG level (nmol/l) NS Maximum insulin response (pmol/l) NS AUC for glucose (mmol/l 180 min) 1, , , NS AUC for insulin (pmol/l 180 min) 69,251 3,676 66,604 4,474 62,784 3,517 NS IGF-I (ng/ml) NS IGFBP-1 (ng/ml) NS Note: Values are means SD. NS not significant; SHBG sex hormone-binding globulin. * Group 1 vs. group 2 and the control group. Rossato. Growth hormone response. Fertil Steril years (range, years). Twelve of the women were of normal weight, 3 were overweight, and none were obese, according to the criteria outlined previously. The mean BMI of the entire group was kg/m 2. The cycle length during the last 6 months was significantly longer in group 1 (mean SD, days; range, days) than in group 2 (mean SD, days; range, days). The daily doses of CC required to induce ovulation were 50 mg in 4 women, 100 mg in 6 women, 150 mg in 4 women, 200 mg in 3 women, and 250 mg in 1 woman. In the control group, eight subjects were of normal weight and one was overweight. The mean ( SD) BMI of the entire group was kg/m 2. Table 1 shows the clinical and hormonal characteristics of the population studied. There were no differences in levels of gonadotropins, testosterone, androstenedione, or sex hormone-binding globulin between the study groups. Fasting glucose and insulin levels were similar in all three groups. The AUC for insulin and glucose in response to the oral administration of glucose was not significantly different between the groups. In patients who were not responsive to CC, levels of IGF-I were not different from those in patients who were responsive to CC, and they were not related to BMI. Concentrations of IGFBP-1 were normal in all patients. Determination of basal hormone levels showed that the women all had similar levels of E 2 ( pmol/l in group 1, pmol/l in group 2, and pmol/l in the control group). In addition, the mean ( SD) GH levels at 30 minutes and 0 minutes were not significantly different in the three groups ( g/l in group 1, g/l in group 2, and g/l in the control group). The administration of clonidine induced a significant increase in GH levels in all the groups studied (Fig. 1). The mean ( SD) peak GH level after the administration of clonidine was g/l in group 1, which was significantly lower than that in group 2 and in the control group (P.01). The mean ( SD) peak GH level was slightly, but not significantly, lower in group 2 than in the control group ( g/l vs g/l). The AUCs for GH are presented in Figure 2. After clonidine administration, there was a significant difference in the AUC for GH between group 1 and group 2 ( g/l vs g/l at 120 minutes, (P.01). There also was a statistically significant difference in the AUC for GH between group 1 and the control group ( g/l vs g/l at 180 minutes, P.01). There was no statistically significant difference when the AUC for GH was compared between group 2 and the control group. No significant correlation was observed between the GH response to clonidine and the dose of CC needed to induce ovulation (Spearman s rank correlation coefficient: r 0.421, P.116). The GH response to clonidine was not significantly correlated with the mean interval between bleeding episodes (Spearman s rank correlation coefficient: 80 Rossato et al. Secretion of GH in CC-resistant women Vol. 73, No. 1, January 2000

4 FIGURE 1 Growth hormone levels (mean SD) obtained after clonidine administration in two groups of infertile patients and a control group of healthy women with normal ovulation. *P.01, significantly different serum concentrations of GH in group 1 vs. group 2 and the control group. Rossato. Growth hormone response. Fertil Steril r 0.29, P.271 in group 1; and r 0.173, P.482 in group 2). No relevant side effects were observed after the administration of clonidine. A slight decrease in blood pressure was recorded in one patient in group 1, two patients in group 2, and no patients in the control group. No significant changes in heart rate were observed at any time during the study. DISCUSSION Anovulatory women who are resistant to CC therapy are a difficult problem for the clinician. Alternative therapy in these patients usually involves the administration of gonadotropins, which are expensive and associated with the risk of ovarian hyperstimulation syndrome and multiple pregnancy. Therefore, it is important to understand the reasons for resistance to CC so as to overcome them. Roozenburg et al. (8) treated a group of anovulatory subjects who were resistant to CC with naltrexone (an opioid receptor blocker) alone or in combination with an antiestrogen. Ovulation was achieved in 19 of 22 patients, demonstrating that endogenous opioids inhibit GnRH secretion and antagonize the effect of CC. An alternate explanation is that naltrexone reduces circulating insulin levels, thereby improving the ovarian response to gonadotropins. The reason for resistance to CC is not known in most patients, and the predominant sites and mechanisms of action of CC remain uncertain. Clomiphene citrate therapy in women results in elevated FSH and LH levels. However, these changes alone are not sufficient to explain the growth of ovarian follicles and ovulation that occur after treatment. It has been demonstrated that anovulatory patients who show a similar increase in gonadotropin levels after CC treatment have a very different rate of ovulation (9). Moreover, the variable growth of ovarian follicles exposed to comparable gonadotropic stimulation suggests that additional modulatory mechanisms play a role in the local control of follicular development. Evidence from human and in vitro studies suggests an important function for GH along with IGF-I and IGF-II in the regulation of folliculogenesis (10, 11). Our study demonstrated a decreased GH response to clonidine among women who were resistant to CC therapy compared with normal controls and women who were responsive to CC therapy. Clonidine generally is believed to be a specific FERTILITY & STERILITY 81

5 FIGURE 2 The GH response to clonidine as the AUC in two groups of infertile patients and in the control group. *P.01, significantly different AUC in group 1 vs. group 2 and the control group. Rossato. Growth hormone response. Fertil Steril activator of central 2 -adrenergic receptors, with no effect on serotonin or dopamine receptors, and it is used as a test of GH secretory reserve, particularly in children (12). Experimental research indicates that clonidine acts by increasing the secretion of hypothalamic GH-releasing hormone without altering somatostatin release (13). In our study, we excluded women with PCOS because a reduced GH response to levodopa and clonidine previously was demonstrated in patients with PCOS (14). Pyridostigmine significantly increased the GH response to levodopa in these women, demonstrating that reduced GH secretion in patients with PCOS may be associated with increased somatostatin activity. In the same study, Lee et al. (14) observed that IGF-I levels were significantly higher in women with PCOS than in controls. Elevated IGF-I levels may inhibit GH secretion through a negative feedback mechanism (15). Kazer et al. (16) observed normal plasma IGF-I concentrations and low GH levels in patients with PCOS. Women with PCOS frequently have insulin resistance and obesity. Insulin resistance is associated with hyperinsulinemia, and insulin can reduce GH secretion (16). Further, insulin-resistant women with PCOS also show resistance to CC (17). In our study, serum concentrations of IGF-I and IGFBP-1, fasting levels of insulin, and the response of insulin levels to the administration of a glucose load were similar in the anovulatory patients and the normal controls. The fact that insulin resistance does not underlie CC resistance in patients without PCOS is an important finding. Obesity is known to be associated with reduced GH levels (18). The BMI was comparable in the patients and controls in our study. Levels of E 2 can modify the secretion of GH in women. Word et al. (19) observed that an increased GH response to GH-releasing hormone is positively correlated with basal E 2 levels. However, in our study, E 2 levels were similar in CC-resistant and CC-responsive women, and not different from the levels observed in normal controls. These observations indicate that estrogen and body weight cannot explain the differences noted in the response of GH to clonidine. However, another study demonstrated a decreased GH response to arginine infusion in anovulatory women but did not distinguish CC-resistant and CC-responsive patients (8). Several studies indicate that GH plays a significant role in regulating the development and maturation of ovarian follicles by amplifying the effect of gonadotropins on ovarian follicular granulosa cells (20). It is known that 82 Rossato et al. Secretion of GH in CC-resistant women Vol. 73, No. 1, January 2000

6 menarche is delayed in GH-deficient subjects, and it was demonstrated that appropriate reproductive function can be restored by the administration of GH (21). Nevertheless, in Ecuadorian women with Laron-type dwarfism, a condition that is characterized by GH resistance, both the ovarian response to gonadotropin stimulation and fertility are normal (22). Thus, GH is important, but not critical, for follicular recruitment and development. Trials investigating the role of cotreatment with GH in the induction of ovulation have produced conflicting results. When it was given to ovulatory women who previously had been resistant to hmg, adjuvant GH treatment failed to improve follicular development and the ovarian response to gonadotropins. Suikkari et al. (23) observed that poor responders to GnRH agonists and gonadotropins do not benefit from cotreatment with human GH during ovarian stimulation. More favorable results were obtained by administering GH to anovulatory women. Homburg et al. (24) observed that GH administered in a placebo-controlled fashion decreased the required dose of hmg in hypogonadotropic patients. The addition of GH to hmg significantly increased the pregnancy rate in clonidinenegative, but not clonidine-positive, infertile women (25). Patients with PCOS who underwent IVF showed a significantly improved response to hmg when pharmacologic doses of GH were added to their superovulation protocols (26). In our study, there was a statistically significant difference in mean cycle length between groups 1 and 2, with a longer mean interval between bleeding episodes in CCresistant women. Genazzani et al. (27) reported that the administration of GH significantly increased integrated LH plasma levels and the LH pulse frequency and amplitude in normogonadotropinemic amenorrheic patients. Therefore, endogenous GH may act on the gonadotropes and modulate LH responsiveness to GnRH. It can be hypothesized that reduced endogenous GH secretion also may impair the function of the hypothalamicgonadotropic-pituitary axis. In group 1, reduced GH secretory reserve was associated with a defect in the hypothalamic regulation of gonadotropin release. A direct effect of GH on granulosa cells has been demonstrated, although the exact role of GH in follicular growth remains to be clarified. Therefore, another explanation is that the reduced GH response to clonidine and the longer cycle length are related to an improper ovarian response to gonadotropins. Our study showed that anovulatory patients who are resistant to CC therapy have a hypothalamic-pituitary disorder and a relative GH deficiency that may result from increased somatostatinergic tone. Although CC-responsive and CCresistant women had similar basal GH levels, it seems that an adequate secretory capacity for GH is important for CC action. However, further studies are necessary to establish whether basal secretion of GH also is impaired and whether the additional administration of GH might transform CCresistant anovulatory women without hyperandrogenism or polycystic ovaries into CC-responsive women. References 1. Downs KA, Gibson M. Clomiphene citrate therapy for luteal phase defect. Fertil Steril 1983;39: Hammond MG. Monitoring technique for improved pregnancy rates during clomiphene ovulation induction. Fertil Steril 1984;42: Lobo RA, Gysler M, March CM, Goebelsmann M, Mishell DR. Clinical and laboratory predictors of clomiphene response. Fertil Steril 1982;37: Isaacs JD, Lincoln SR, Cowan BD. Extended clomiphene citrate (CC) and prednisone for the treatment of chronic anovulation resistant to CC alone. Fertil Steril 1997;67: Barreca A, Artini PG, Del Monte P, Ponzani P, Pasquini P, Cariola G, et al. In vivo and in vitro effect of growth hormone on estradiol secretion by human granulosa cells. J Clin Endocrinol Metab 1993;77: Menashe Y, Lunenfeld B, Pariente C, Frenkel Y, Mashiach S. Can growth hormone increase, after clonidine administration, predict the dose of human menopausal hormone needed for induction of ovulation? Fertil Steril 1990;53: Ovesen P, Moller G, Moller N, Christiansen JS, Orskov JOL. Growth hormone secretory capacity and serum insulin-like growth factor I levels in primary infertile, anovulatory women with regular menses. Fertil Steril 1992;57: Roozenburg BJ, van Dessel H, Evers J, Bots R. Successful induction of ovulation in normogonadotrophic clomiphene resistant anovulatory women by combined naltrexone and clomiphene citrate treatment. Hum Reprod 1997;12: Quigley MM, Berkowitz AS, Gilbert SA, Wolf DP. Clomiphene citrate in an in vitro fertilization program: hormonal comparisons between 50- and 150-mg daily dosages. Fertil Steril 1984;41: Adashi E. Intraovarian regulation: the proposed role of insulin-like growth factors. Ann NY Acad Sci 1993;697: Barreca A, Del Monte P, Ponzani P, Artini PG, Genazzani AR, Minuto F. Intrafollicular insulin-like growth factor-ii levels in normally ovulating women and in patients with polycystic ovary syndrome. Fertil Steril 1996;65: Saggese G, Cesaretti G, Giannesi N, Brancaloni C, Cinquanta L, Cioni C. Stimulated growth hormone (GH) secretion in children with delays of pubertal development before and after the onset of puberty: relationship with peripheral plasma GH-releasing hormone and somatostatin levels. J Clin Endocrinol Metab 1992;74: Magnan E, Catldi M, Guillaume V. Role of growth hormone (GH)- releasing hormone and somatostatin in the mediation of clonidine induced GH release in sheep. Endocrinology 1994;134: Lee E-J, Lee BS, Lee HC, Park KH, Song CH, Huh KB. Growth hormone response to L-dopa and pyridostigmine in women with polycystic ovarian syndrome. Fertil Steril 1993;60: Yamashita S, Weiss M, Melmed S. Insulin-like growth factor I regulates growth hormone secretion and messenger ribonucleic acid levels in human pituitary tumour cells. J Clin Endocrinol Metab 1986;63: Kazer RR, Unterman TG, Glick RP. An abnormality of the growth hormone/insulin-like growth factor-i axis in women with polycystic ovary syndrome. J Clin Endocrinol Metab 1990;71: Murakawa H, Hasegawa I, Kurabayashi T, Tanaka K. Polycystic ovary syndrome. Insulin resistance and ovulatory response to clomiphene citrate. J Reprod Med 1999;44: Slowinska-Srdeznicka J, Zgliczynsksi W, Wakowska A, Jeske W, Brzezinska A, Soszynski P, et al. An abnormality of the growth hormone/insulin like growth factor-i axis in women with polycystic ovary syndrome due to coexistent obesity. J Clin Endocrinol Metab 1992;74: Word RA, Odom MJ, Byrd W, Carr BR. The effect of gonadotropinreleasing hormone agonists on growth hormone secretion in adult premenopausal women. Fertil Steril 1990;54: Ovesen P, Ingerslev HJ, Orskov H, Ledet T. Effect of growth hormone on steroidogenesis, insulin-like growth factor-1 (IGF-1) and IGF-binding protein-1 production and DNA synthesis in cultured human luteinized granulosa cells. J Endocrinol 1994;140: Katz E, Ricciarelli E, Adashi EY. The potential relevance of growth hormone to female reproductive physiology and pathophysiology. Fertil Steril 1993;59: Rosenfeld RG, Rosenbloom AL, Guevara-Aguirre J. Growth hormone (GH) insensitivity due to primary GH receptor deficiency. Endocr Rev 1994;15: Suikkari A-M, MacLachlan V, Koistinen R, Seppâlä M, Healy DL. Double-blind placebo controlled study: human biosynthetic growth hormone for assisted reproductive technology. Fertil Steril 1996;65: FERTILITY & STERILITY 83

7 24. Homburg R, Levy T, Ben-Rafael Z. Adjuvant growth hormone for induction of ovulation with gonadotrophin-releasing hormone agonist and gonadotrophins in polycystic ovary syndrome: a randomized double blind placebo controlled trial. Hum Reprod 1995;10: Blumenfeld Z, Dirnfeld M, Gonen Y, Abramovici H. Growth hormone co-treatment for ovulation induction may enhance conception in the co-treatment and succeeding cycles, in clonidine negative but not clonidine positive patients. Hum Reprod 1994;9: Volpe A, Artini PG, Barreca A, Coukos G, Minuto F, Genazzani AR. Effects of growth hormone administration in addition to gonadotrophins in normally ovulating women and polycystic ovary syndrome (PCO) patients. Hum Reprod 1992;7: Genazzani AD, Petraglia F, Volponi C. Growth hormone treatment affects plasma LH pulsatile release in women with secondary amenorrhoea. Clin Endocrinol (Oxf) 1993;39: Rossato et al. Secretion of GH in CC-resistant women Vol. 73, No. 1, January 2000