Ovulation induction in premature ovarian failure: a placebo-controlled randomized trial combining pituitary suppression with gonadotropin stimulation

Size: px

Start display at page:

Download "Ovulation induction in premature ovarian failure: a placebo-controlled randomized trial combining pituitary suppression with gonadotropin stimulation"

Agnes Johnson
6 years ago
Views:

1 FERTILITY AND STERILITY Copyright 1995 American Society for Reproductive Medicine Vol. 64, No.2, August 1995 Printed on acid-free paper in U. S. A. Ovulation induction in premature ovarian failure: a placebo-controlled randomized trial combining pituitary suppression with gonadotropin stimulation Yvonne M. van Kasteren, M.D. * Annemiek Hoek, M.D.t Joop Schoemaker, M.D., Ph.D.t Institute of Endocrinology, Reproduction and Metabolism, Vrije Universiteit, Amsterdam, The Netherlands Objectives: To determine the effect of pituitary suppression with a GnRH agonist (GnRH-a) on the success of ovulation induction with exogenous gonadotropins in patients with premature ovarian failure (POF). Design: Placebo-controlled, randomized, double-blind study. The data were analyzed with a Fisher exact test. Setting: A tertiary care academic center for Reproductive Endocrinology and Fertility. Patients: Thirty patients with POF, 15 in each group. Interventions: The study consisted of four phases: phase 1, no interventions; phase 2, a 4- week period in which the patients received either 1,000 f-tg intranasal buserelin acetate daily or placebo; phase 3, a 3-week period during which the patients additionally received hmg in weekly augmented doses, two, four, and six ampules daily in the first, second, and third weeks, respectively. Ovulation was induced whenever the follicular diameter reached 18 mm and/or total 24-hour estrogen excretion> 140 f-tg (500 nmol). Luteal support was 5,000 IV hcg every 72 hours; phase 4, no interventions. Results: Follicular growth was seen in five patients of the agonist group and in four patients of the placebo group. Three of 15 patients in the agonist group ovulated versus none in the placebo group. The difference was not statistically significant. Conclusions: The fact that 3 of 15 cycles cotreated with a GnRH-a were ovulatory versus none in the placebo-treated group appeared not to be enough evidence to demonstrate that pituitary suppression with a GnRH-a improves the success of ovulation induction with exogenous gonadotropins in patients with POF. Fertil Steril 1995;64:273-8 Key Words: Premature ovarian failure, ovulation induction, pituitary suppression, exogenous gonadotropins Premature ovarian failure (POF) is cessation of ovarian function after puberty before the age of 40 after normal development (1). It is characterized by the occurrence of secondary amenorrhea with elevated gonadotropins and low serum levels of E 2 Approximately 1% of women will experience menopause before the age of 40 (2). The etiology of POF is Received September 19, 1994; revised and accepted March 7, * Reprint requests and present address: Yvonne M. van Kasteren, Department of Obstetrics and Gynecology, Medical Centre Alkmaar, P.O. Box 501, 1800 AM Alkmaar, The Netherlands (FAX: ). t Institute of Endocrinology, Reproduction and Metabolism. diverse but in the majority of patients it is unknown. This idiopathic group probably is heterogeneous, consisting either of patients with early follicular depletion, or of patients with the "resistant ovary syndrome," in whom follicles are still present in the ovaries. In the latter, autoimmune phenomena are thought to playa major role in causing the unresponsiveness of the ovarian follicles to normal gonadotropin stimulation (3). As differentiation between the two is not possible, the group is treated as one from a therapeutic point of view. Because the hypergonadotropic status itself might play a role in the unresponsiveness of the ovary, several authors studied the effect of creating a normogonadotropic status. Menon (4) showed that Vol. 64, No.2, August 1995 van Kasteren et al. GnRH agonist and gonadotropins in PDF 273

2 & MQ_2!212!!E!! GnRH agonists (GnRH-a) could suppress FSH levels to normal. Surrey and Cedars (5) and Check et al. (6), in an open nonrandomized trial, showed that ovulation could be induced in some patients combining pituitary suppression with hmg stimulation. Speculations can be made about the mode of action of GnRH-a. In some patients the initial stimulatory effect of the agonists on the release of gonadotropins (the so-called flare phenomenon) may be responsible for the stimulation of follicular growth. Also, some in vitro data suggest that there is a direct effect of the agonist on the ovary through the aromatase system (Muller H, Rabe T, Kiesel L, Runnebaum B, abstract). The most intriguing idea, however, is that a rise in ovarian sensitivity is induced by a fall in endogenous FSH levels, mediated by an up-regulation offsh receptors in the granulosa cells. Alternatively, a declining susceptibility ofthe ovary for lymphocyte infiltration or a diminished autoantagonistic activity of FSH caused by a change in glycosylation of the FSH molecule induced by the agonist (7) could improve ovarian sensitivity. Before considering any of these possibilities as causative factors for successful induction of ovulation in POF patients, the element of mere coincidence has to be ruled out. We therefore designed a placebo-controlled, doubleblind, randomized trial, hypothesizing that ovulation induction with gonadotropins in patients with POF would be more successful when preceded by and combined with pituitary suppression with a GnRH-a. Patients MATERIALS AND METHODS Thirty women aged 25 to 38 years who desired pregnancy participated in the study. The study was approved by the hospital committee for the ethics of research involving human subjects. Informed consent was obtained from all 30 women. All women were diagnosed to have POF. Premature ovarian failure was defined as a secondary oligomenorrhea or amenorrhea with FSH serum levels> 40 miu/ml (conversion factor to SI unit, 1.00) on two separate occasions in the 2 months preceding admittance to the protocol. All patients had a normal chromosomal pattern and no history of radiotherapy or chemotherapy. All hormonal medication had been discontinued 3 months before entering the protocol. The characteristics of the individual patients are listed in Table 1. The mean age was 30.8 years (range 25 to 38 years). Twenty-two patients had been amenorrheic for >1 year, five patients for >6 months, and three patients were still oligomenorrheic with vaginal bleeding <6 months preceding Table 1 Patient Characteristics and Results Patient characteristics Results Patient Amenor- Family Anti- Follicle Ovulano. Age rhea history bodies growth ReG tion y Buserelin Placebo the protocol. In 14 patients one or more autoantibodies could be demonstrated. Only two patients had a personal history revealing autoimmune disease and two others had a positive family history for this. Six patients had a positive family history for early or premature menopause. Two patients had a history of ovarian surgery for benign cysts and two other patients had gone through a period of severe pelvic inflammation. As would be expected these characteristics were represented equally in both study groups. Study Protocol The study was designed as a prospective, placebocontrolled, double blind, randomized trial. The hospital pharmacy randomized and prepared the medication, which was blinded for the investigator and the patient. The study consisted offour phases. During the first phase of 4 weeks, in which no medication was given, the patients were monitored to evaluate their exact gonadotropic status. In the second phase, the patients received either 1,000 j1g buserelin acetate intranasally per day (Suprefact; Hoechst, Frankfurt, Germany) or placebo for 4 weeks. Monitoring was performed to detect "spontaneous" follicular activity. During the third phase, the pituitary 274 van Kasteren et al. GnRH agonist and gonadotropins in POF Fertility and Sterility

suppression was continued and combined with ovarian stimulation with 150 IU purified urinary FSH (Metrodin; Serono, Aubonne, Switzerland) daily.

3 suppression was continued and combined with ovarian stimulation with 150 IU purified urinary FSH (Metrodin; Serono, Aubonne, Switzerland) daily. If necessary, the dose was raised with 150 IU daily every week until a maximum of 450 IU daily was reached. Ovulation was triggered with 10,000 IU hcg (Profasi; Serono) whenever a follicle had reached a diameter of ~ 18 mm and/or total 24-hour urinary estrogen excretion exceeded 140 /Lg (500 nmol); luteal support consisted of 5,000 IU hcg every 72 hours. When follicular growth was still absent after 3 weeks of stimulation, all medication was discontinued. Monitoring was continued for another 4 weeks during the fourth phase to detect a possible rebound phenomenon. Monitoring consisted of a BBT chart and vaginal ultrasound (US). Twenty-four-hour urine samples were collected three times per week and processed immediately for determination of total urinary estrogens and once per week for pregnanediol excretion, the latter only after ovulation had been documented. Vaginal US was performed once per week in phases 1 and 4 and three times per week in phases 2 and 3. Whenever a follicle reached a diameter of 14 mm, daily scans were performed. Plasma samples were taken three times per week to estimate FSH and LH levels and were analyzed retrospectively. All patients who, in retrospect, had received placebo were offered a three-phase treatment cycle as described for the buserelin acetate group. Hormone Assays Plasma FSH samples were analyzed in duplicate by an immunometric luminescent assay (Amerlite; Kodak Clinical Diagnostics, Amersham, United Kingdom) with a lower detection limit of 0.5 miu/ml (conversion factor to SI unit, 1.00), an intra-assay coefficient of variation (CV) of 5% to 6%, and an interas say CV of 6% to 9%. Plasma LH samples were analyzed by a similar assay with a lower detection limit of 0.3 miu/ml, an intra-assay CV of 3% to 5%, and an interassay CV of 4% to 10%. Total estrogen excretion was measured by an automated rapid fluorometric assay based on the Ittrich reaction. The intra-assay CV was 15%. Pregnanediol was measured by gas chromatography after mild acid hydrolysis, as described by Metcalf(8). A value of>3 /Lmol/ 24 h was considered substantial evidence of ovulation. The intra-assay CV was 12%. mtrasound Follicular diameter was measured by vaginal ultrasonography using a 7.5-MHz probe. In case one or both ovaries could not be visualized, a 5-MHz probe was used. Statistical Analysis The hypothesis was that pituitary suppression with GnRH-a would improve ovulation rate in ovulation induction with gonadotropins in patients with POF. The data were analyzed by the Fisher exact test. In retrospect, a logistic regression analysis was performed to determine whether one or more parameters would positively influence the development of follicular growth. The tested parameters were duration of amenorrhea, positive family history for premature menopause, existence of autoantibodies, and the use of GnRH -a. RESULTS Follicular growth was defined as a rise in total estrogen excretion of 100% above the basal level of the patient and a growing follicle of ~1O mm in diameter on vaginal US. Follicular activity was defined as changes observed in the ovary or estrogen excretion that did not reach the above mentioned criteria. In the placebo group four patients showed follicular growth. Two were oligomenorrheic, one patient had been amenorrheic for 6 months, and one patient had been amenorrheic for 1 year. In all four patients some follicular activity was seen in either phase 1 (no medication) or phase 2 (placebo). One of those showed follicular growth at the end of phase 1 continuing into phase 2. The criteria for hcg were not met (maximum follicular diameter 17 mm) and atresia occurred. During the following stimulation no reaction was seen. In one patient follicular growth started in phase 2 (placebo) and continued into phase 3 (stimulation), but the criteria for hcg were not met (maximum follicular diameter 16 mm) and ovulation did not occur (Fig. 1, patient 9). In the two other patients follicular growth started in phase 3 (stimulation). The follicular diameter exceeded 18 mm and they received hcg. However, ovulation could not be detected. Estrogen excretion remained below 500 nmol/24 h in all patients. In the buserelin acetate group, five patients showed follicular growth. Three patients had experienced amenorrhea for ~6 months, one patient for 1 year, and one patient for 5 years. One patient showed a follicle-like structure with elevated P levels at the beginning of phase 1 (no medication), probably reflecting a luteinized follicle, which disappeared before medication was started. Follicular growth was demonstrated in phase 3 (stimulation), hcg was given, and ovulation occurred. In one patient, follicular growth started at the end of phase 2 (suppression) and continued into phase 3 (stimulation) when hcg was given and ovulation was docu- Vol. 64, No.2, August 1995 van Kasteren et al. GnRH agonist and gonadotropins in PDF 275

4 pt.8 pt.9 pt.12 FSH miulmlr-- -, , r , , r ,lh mlulml ' E t "mo,1 ---,, ---,, ---,P2 umol days Figure 1 Graphic presentation of phase 2 (suppression or placebo) and phase 3 (stimulation). Patient 8, nonresponder in the buserelin acetate group; Patient 9, anovulatory cycle in the placebo group; Patient 12, ovulatory cycle in the buserelin acetate group. ---, FSH miu/ml; -, LH miulml; D, E nmol; total 24-h urinary estrogen excretion,., P2 {Lmol; pregnanediol excretion. mented. In the other three patients follicular growth started in phase 3 (stimulation). One of these patients developed a follicle with a diameter of 18 mm. She received hcg and ovulation occurred (Fig. 1, patient 12). The two other patients did not meet the criteria for hcg (maximum follicular diameter 17 mm) and ovulation did not occur. Again total 24- hour urinary excretion of estrogens did not exceed 500 nmol in anyone patient. The length of the luteal phase was 18, 19, and 21 days, respectively. In one patient total urinary pregnanediol excretion only marginally exceeded 3 {Lmol/24 h. The two other ovulatory cycles showed a normal pattern of pregnanediol excretion. In every patient treated with the agonist, the FSH level was suppressed < 10 miu/ml and the LH level was suppressed <3 miu/ml. Four of five patients showing follicular growth started phase 2 with FSH levels >75 miu/ml. Ovulation occurred exclusively in the agonist group. The difference in ovulation rate of 3/15 in the agonist group and of 0/15 in the placebo group was not statistically significant (P = 0.11). Twelve patients from the placebo group completed a treatment cycle with the agonist after the protocol: one ovulated. Thus, from a total of 27 patients who were treated with a combination of suppression and stimulation, 4 patients responded with an ovulatory cycle. None of the patients became pregnant during the first treatment cycle with buserelin acetate. The four patients who showed ovulatory cycles were treated further with the same medication scheme. One patient ovulated in the second treatment cycle but failed to ovulate in the third cycle. In the fourth cycle, IVF was performed. Of one follicle, one oocyte fertilized and she became pregnant. Unfortunately, she miscarried and she refrained from further treatment. Two patients had two other attempts of ovulation induction without obtaining an ovulatory cycle. In the fourth patient, three more attempts were performed, in all of which the follicle was aspirated. One preovulatory oocyte was obtained only in the last attempt. Although a transfer could be achieved with an embryo of good quality, according to morphological criteria, pregnancy did not occur. A logistic regression analysis was performed to detect whether any parameter would correlate positively with either spontaneous or induced follicular growth. We checked for duration of amenorrhea, existence of autoantibodies, and a positive family history for early menopause. None of these parameters showed any influence on the prediction of follicular growth. Because of the low number of ovulatory cycles, a logistic regression analysis could not be applied with regard to the occurrence of ovulation. However, the three patients that ovulated all had an amenorrhea < 1 year, they showed at least one autoantibody, and they were treated with buserelin acetate. 276 van Kasteren et al. GnRH agonist and gonadotropins in POF Fertility and Sterility

DISCUSSION This study shows no difference in ovulation rate between the group with a GnRH-a-induced normogonadotropic status and the placebo group with a hypergonadotropic status.

5 DISCUSSION This study shows no difference in ovulation rate between the group with a GnRH-a-induced normogonadotropic status and the placebo group with a hypergonadotropic status. However, because of the small number of patients, the power of this study is limited. To detect a statistically significant difference of20% between two groups (type I error = 0.05, type II error = 0.10), 46 patients should have been included in both arms of the study. It is doubtful whether expanding the groups would have given rise to a more distinctive difference, because the 12 patients who were treated with the agonist after having been treated with a placebo showed only one ovulatory cycle. On the other hand, if we had given hcg at a follicular diameter of 16 mm, the difference between the groups might have been more prominent, for the patients in the buserelin acetate group tended to ovulate after hcg and the patients in the placebo group did not. Interpretation ofthe results is complicated further by the assumed heterogeneity of the group. Because differentiation between the follicular and the afollicular form of POF is essential for the evaluation of therapeutical interventions, several attempts have been made to distinguish the two forms by performing ovarian biopsies. The studies concerning ovarian biopsies in POF patients obtained by laparotomy (9-11), culdoscopy (12, 13), or laparoscopy (14-16) all showed that, in 60%, the biopsies were devoid of follicles reflecting complete depletion, whereas, in ±40%, follicles still could be demonstrated. Within the group of patients who had positive biopsies, approximately two thirds showed only a few follicles either primordial and/or in different stages of development and one third showed numerous, predominantly primordial, follicles. The former group also could reflect early depletion of the oocyte storage pool and the latter could reflect the so-called resistant ovary syndrome. These two groups of subjects could respond theoretically to ovulation induction therapy. However, several authors reported pregnancies in patients with negative or empty ovarian laparoscopic biopsies (17-19). Apparently, ovarian biopsies obtained by laparoscopy do not have an acceptable predictive value for the presence of a few follicles. Therefore, no ovarian biopsies were taken in this study. It can be argued that in properly selected patients pituitary suppression might have been able to improve the ovulation rate. Because the resistant ovary syndrome is thought to be associated with autoimmune disease, the presence of autoimmune phenomena might give circumstantial evidence that follicles are present. These autoimmune phenomena may be coexisting autoimmune endocrinopathies (20, 21), infiltration of plasma cells and lymphocytes in the ovaries (22, 23), the presence of autoantibodies (24), or changes in subpopulations of lymphocytes (25). The possibility exists that pituitary suppression can influence only the responsiveness of the ovaries in patients with early follicular depletion who still have some follicles left, whereas the effects of an autoimmune process, creating the resistant ovary syndrome, are not altered. This assumption is supported by the fact that the majority of the patients in this study were amenorrheic for < 1 year, although the duration of amenorrhea did not show any positive predictive value with regard to the occurrence of follicular growth. In contrast, the presence of autoantibodies had no negative predictive influence on the development of follicular growth. Moreover, all three patients that ovulated had autoantibodies. In conclusion, the fact that 3 of 15 cycles cotreated with buserelin acetate were ovulatory versus none in the placebo-treated group appeared not to be enough evidence to demonstrate that pituitary suppression with a GnRH-a improves the success of ovulation induction with exogenous gonadotropins in patients with POF. Acknowledgment. We appreciate greatly the technical assistance of the Endocrine Laboratory, Vrije Universiteit, Amsterdam, The Netherlands (Corry Popp-Snijders, Ph.D.). REFERENCES 1. Coulam CB. Premature gonadal failure. Fertil Steril 1982;38: Coulam CB, Adamson SC, Annegers JF. Incidence of pre mature ovarian failure. Obstet Gynecol 1986;67: Mignot MH, Schoemaker J, Kleigeld M, Rao BR, Drexhage HA. Premature ovarian failure. The association with autoimmunity. Eur J Obstet Gynecol Reprod BioI 1989;30: Menon V, Logan Edwards R, Lynch SS, Butt WR. Luteinizing hormone releasing hormone analog in treatment of hypergonadotropic amenorrhea. Br J Obstet Gynaecol 1983;90: Surrey ES, Cedars MI. The effect of gonadotropin suppression on the induction of ovulation in premature ovarian failure patients. Fertil Steril 1989;52:36-4l. 6. Check JH, Nowroozi K, Chase JS, Nazari A, Shapse D, Vaze M. Ovulation induction and pregnancies in 100 consecutive women with hypergonadotropic amenorrhea. Fertil Steril 1990;53: Mortola JF, Sathanandan M, Pavlou S, Dahl KD, Hsueh AJW, Rivier J, et al. Suppression of bioactive and immunoreactive follicle stimulating hormone and luteinizing hormone levels by a potent gonadotropin-releasing hormone antagonist: pharmacodynamic studies. Fertil Steril 1989; 51: Metcalf MG. Hydrolysis and decomposition of urinary pregnandiol in acid. Steroids 1973:21; Emperaire JC, Audebert A, Greenblatt RB. Premature ovarian failure. Am J Obstet Gynecol 1970; 108: Vol. 64, No.2, August 1995 van Kasteren et al. GnRH agonist and gonadotropins in POF 277

6 10. Starup J, Sele V. Premature ovarian failure. Acta Obstet Gynecol Scand 1973;52: Zarate A, Karchmer S, Gomez E, Castelazo-Ayala L. Premature menopause. Am J Obstet Gynecol 1970; 106: Kinch RAH, Plunkett ER, Smout MS, Carr DH. Primary ovarian failure. Am J Obstet Gynecol 1965;91: Sharf M, Israeli I, Graf G. The value of ovarian biopsy in the diagnosis and treatment of amenorrhea related sterility. Obstet Gynecol 1972;39: Duignan NM. Sex hormone levels and gonadotrophin release in premature ovarian failure. Br J Obstet Gynaecol 1978; 85: Board JA, Redwine AO, Moncure CW, Frable WJ, Taylor JR. Identification of differing etiologies of clinically diagnosed premature menopause. Am J Obstet Gynecol 1979; 134: Muechler EK, Huang K, Schenk E. Autoimmunity in premature ovarian failure. Int J Fertil 1991;36: Shangold MM, Turksoy RN, Bashford RA, Hammond CB. Pregnancy following the "insensitive ovary syndrome." Fertil Steril 1977;28: O'Herhlihy C, Pepperell RJ, Evans JH. The significance of FSH elevation in young women with disorders of ovulation. Br Med J 1980;281: Oshawa M, Wu M, Masahashi T, Asai M, Narita O. Cyclic therapy resulted in pregnancy in premature ovarian failure. Obstet Gynecol 1985;66:64S-7S. 20. Edmonds M, Lamki L, Killinger DW, Volpe R. Autoimmune thyroiditis, adrenalitis and oophoritis. Am J Med 1973;54: Irvine WJ, Barnes EW. Addison's disease, ovarian failure and hypoparathyroidism. Clin Endocrinol Metab 1976;4: Gloor E, Hurlimann J. Autoimmune oophoritis. Am J Clin Pathol 1984;81: Sedmak DD, Hart WR, Tubbs RR. Autoimmune oophoritis: a histopathologic study of involved ovaries with immunologic characterization of the mononuclear infiltrate. Int J Gynecol Pathol 1987;6: Weissenbruch van M, Hoek A, Vliet-Beker van I, Schoemaker J, Drexhage HA. Evidence for existence of immunoglobulins that block ovarian granulosa cell growth in vitro. A putative role in resistant ovary syndrome? J Clin Endocrinol Metab 1991;73: Mignot MH, Drexhage HA, Kleingeld M, Plassche-Boers van de EM, Rao BR, Schoemaker J. Premature ovarian failure: considerations of cellular immunity defects. Eur J Obstet Gynecol Reprod BioI 1989; 30: van Kasteren et al. GnRH agonist and gonadotropins in POF Fertility and Sterility

Criteria for considering studies for this review

Criteria for considering studies for this review Página 1 de 6 Ovulation induction in women with spontaneous premature ovarian failure [protocol] Kalantaridou SN, Calis KA, Nelson LM This protocol should be cited as: Kalantaridou SN, Calis KA, Nelson