Gonadotropin-releasing hormone agonist reduces the miscarriage rate for pregnancies achieved in women with polycystic ovarian syndrome

Size: px

Start display at page:

Download "Gonadotropin-releasing hormone agonist reduces the miscarriage rate for pregnancies achieved in women with polycystic ovarian syndrome"

Oliver Carson
5 years ago
Views:

1 FERTILITY AND STERILITY Copyright e 1993 The American Fertility Society Vol. 59, No.3, March 1993 Printed on acid-free paper in U.S.A. Gonadotropin-releasing hormone agonist reduces the miscarriage rate for pregnancies achieved in women with polycystic ovarian syndrome Roy Homburg, M.B., B.S. * Tally Levy, M.D. Drorit Berkovitz, M.D. Jacob Farchi, M.D. Dov Feldberg, M.D. Jacob Ashkenazi, M.D. Zion Ben-Rafael, M.D. Fertility Unit, Department of Obstetrics and Gynecology, Golda Medical Center, and Sackler Medical School, Tel Aviv University, Petah Tikva, Israel Objective: To compare the effect of treatment with gonadotropin-releasing hormone agonist (GnRH-a) and human menopausal gonadotropins (hmg) with that of gonadotropins only, on the cumulative livebirth rate and miscarriage rate of pregnancies achieved in women with polycystic ovarian syndrome (PCDS). Design: Retrospective analysis of the outcome of 97 pregnancies according to the treatment protocol, with or without GnRH -a. Calculation of miscarriage rate and cumulative livebirth rate by life-table analysis. Setting: Infertility clinic and in vitro fertilization (IVF) unit. Patients: Women with polycystic ovaries (n = 239) who were clomiphene citrate failures and received either GnRH-a/hMG (n = 110) or gonadotropins only (n = 129) for ovulation induction (n = 138) or superovulation for IVF (n = 101). Interventions: For ovulation induction, hmg was given in a step-up, individually adjusted dose scheme. For IVF, three ampules of pure follicle-stimulating hormone were given for 3 days followed by three ampules per day hmg and then individual dose adjustment. Gonadotropin-releasing hormone agonist (Decapeptyl, D-Trp6, microcapsules, 3.75 mg) was given in a single dose 2 weeks before gonadotropin treatment. Main Outcome Measures: The rate of early miscarriages «12 weeks) per pregnancies achieved was analyzed, and the cumulative livebirth rate for each treatment group was calculated by lifetable analysis. Results: Miscarriage rates after treatment in ovulation induction with (16.7%) and without GnRH-a (39.4%) and in IVF with (18.2%) and without GnRH-a (38.5%) were almost identical and were therefore analyzed together. Dfpregnancies achieved with GnRH-a, 17.6% miscarried compared with 39.1 % of those achieved with gonadotropins alone. Cumulative livebirth rate after four cycles for GnRH-a was 64% compared with 26% for gonadotropins only. Conclusions: Cotreatment with GnRH-a/hMG for anovulatory women with PCDS reduces the miscarriage rate and improves the livebirth rate compared with treatment with gonadotropins alone. Fertil Steril 1993;59: Key Words: Polycystic ovarian syndrome, induction of ovulation, in vitro fertilization, gonadotropin-releasing hormone agonist, gonadotropin treatment, miscarriage rate Received September 16, 1992; revised and accepted November 16,1992. * Reprint requests: Roy Homburg, M.B., B.S., Golda Medical Center, Petah Tikva 49372, Israel. Polycystic ovarian syndrome (PCOS) is a very common cause of female anovulatory infertility. If treatment with clomiphene citrate (CC) fails, gonadotropin treatment is usually attempted but is Vol. 59, No.3, March 1993 Homburg et al. GnRH-a reduces miscarriage rate in PCOS 527

2 ,.. accompanied by difficulties such as premature luteinization, a relatively low pregnancy yield, and complications of ovarian hypersensitivity such as multiple follicular development, ovarian hyperstimulation syndrome, and multiple pregnancies. After six ovulatory cycles with gonadotropin induction of ovulation but no resulting pregnancy, in vitro fertilization (IVF) may be recommended (1). However, once a pregnancy is achieved, the high miscarriage rate is extremely irritating and heartbreaking for the infertile couple. Several studies have indicated that high concentrations of luteinizing hormone (LH) in the late follicular phase may be responsible for the high miscarriage rate (2-4). The use of gonadotropin-releasing hormone agonist (GnRH-a) throughout induction of ovulation or superovulation for IVF would seem to be a possible solution to the problem and a test of the hypothesis. This study compares the miscarriage rate of pregnancies achieved and cumulative livebirth rates in a large number of women with PCOS who were treated with gonadotropins only or in combination with GnRH-a in ovulation induction and IVF programs. Definitions MATERIALS AND METHODS The diagnosis of PCOS was based on the typical ultrasound (US) findings (a hyperechogenic central stroma and at least 10 follicles < 9 mm in diameter) (5) accompanied by anovulation and infertility for at least 1 year and either oligomenorrhea/amenorrhea and/or hirsutism. They were referred for treatment after at least six cycles of CC in a dose of up to 1,000 mg per cycle during which they had failed to conceive. A miscarriage was defined as thespontaneous abortion (SAB) of a clinical pregnancy, up to 12 full weeks of gestation, previously diagnosed on US examination by the identification of an intrauterine gestational sac. Treatment Protocol for Ovulation Induction Women in the ovulation induction group received a standard individually adjusted dose scheme of treatment with human menopausal gonadotropin (hmg, Pergonal; Teva, Petah Tikva, Israel) starting with one ampule per day by intramuscular injection. The dose was raised by one ampule per day every 5 days until a follicular growth response and an in- crease in 17-(3 estradiol (E2) levels were noted whereupon this daily effective dose was continued. Human chorionic gonadotropin (10,000 IU 1M, hcg, Chorigon; Teva) was administered when at least one follicle with diameter> 16 mm was seen on US and E2 levels were in the range of 600 to 2,000 pg/ml (2200 to 7,300 pmol/l). Only for those women receiving GnRH-a was luteal phase support given in the form of injections of hcg (2,500 IU 1M) on days 3, 7, and 11 after the original injection of hcg. If ovarian hyperstimulation was considered a likely outcome, progesterone (P) treatment was substituted for hcg. Treatment Protocol for IVF Women in the IVF program received a standard stimulation protocol consisting of three ampules per day of purified follicle-stimulating hormone (pfsh, Metrodin, Teva) for 3 days followed by three ampules per day of hmg. Adjustments of the dose of hmg were based on the individual dose-response scheme as described above. Human chorionic gonadotropin (10,000 IU 1M) was administered when serum E2 concentrations reached a mean of 150 to 200 pg/ml (530 to 700 pmol/l) per preovulatory follicle of 15 to 16 mm or above and there were at least two follicles> 15 mm in diameter. Ovum pickup was performed 32 to 36 hours after hcg administration by vaginal US and embryo transfer (ET) 48 hours later. Hydrogenized P vaginal suppositories (privately manufactured, 50 mg, two times per day) were given throughout the luteal phase and were continued until a pregnancy test was negative or up to the 7th week of pregnancy. The GnRH-a employed was Decapeptyl (D-Trp6; Ferring, Malmo, Sweden) in the form of microcapsules in a single injection of 3.75 mg 1M given 3 weeks after a progestin-induced withdrawal bleed. Two weeks after this injection the serum concentration of E2 was measured, and a US examination of the ovaries was performed. If the E2 concentration was <20 pg/ml «70 pmol/l) and there were no follicles or cysts> 10 mm in diameter, gonadotropin treatment was started. If this ovarian quiescence had not been achieved, a further examination was performed 1 week later. Patients and Study Design This is a retrospective analysis comparing the outcome of 97 pregnancies achieved from the treatment of 239 women with PCOS through 484 cycles 528 Homburg et al. GnRH-a reduces miscarriage rate in peas Fertility and Sterility

3 f Table 1 Basal Clinical and Laboratory Parameters of Women With PC os Receiving Treatment With or Without GnRH-a for Ovulation Induction or IVF Ovulation induction IVF GnRH-a HMG GnRH-a FSH/hMG No. of patients No. of cycles Age (y) BMI (kg/m 2 ) LH (lull) FSH (lull) T (ng/ml):j: * Values are means ± SD. t NA, not available ± ± ± ± 5.9* 28.3 ± ± ± ± ± ± ± 4.7 NAt 18.3 ± ± ± 0.27 :j: Conversion factor to SI unit, ± 4.2 NA 16.4 ± ± ± 0.3 by ovulation induction (n = 138,295 cycles) or by IVF-ET (n = 101,189 cycles) (Table 1). For ovulation induction, 111 cycles with GnRH-a/hMG and 184 cycles with hmg only were given. In the IVF group, GnRH-a was administered in 105 cycles and gonadotropins only in 84 cycles. Women in the IVF treatment group had failed to conceive in six ovulatory cycles after ovulation induction. They had all been thoroughly investigated, but no other cause for their infertility was discovered over and above the anovulation associated with PCOS. There were no preselection criteria for the allocation of patients to treatment with or without GnRH-a, and both treatment alternatives were used concurrently throughout the study. Within the ovulation induction and IVF groups the distribution of patients according to mean age, body mass index (BMI), and mean serum concentrations of LH, FSH, and testosterone (T) did not differ between those receiving GnRH-a and those who did not, confirming the basal similarity of the groups (Table 1). The LH:FSH ratio of the whole group was 2.5:1. Of those receiving ovulation induction, 47 women had previous miscarriages: 25 in the GnRH -a group and 22 of those who received gonadotropins only. Of the women undergoing IVF, 35 had previous miscarriages: 17 received GnRH-a and 18 received gonadotropins only. None had suffered more than two previous miscarriages. In this report, particular emphasis has been placed on a comparison of miscarriage rates and cumulative livebirth rates between the agonist- and non-agonisttreated women. To compare differences in the frequency of miscarriage X2 analysis was employed. Cumulative livebirth rates were analyzed by life-table analysis (6) on all treatment cycles leading to the first livebirth. Thus a cycle that resulted in a conception that terminated in a miscarriage was recorded as a failed treatment cycle. Statistical significance of differences in rates among the treatment groups was calculated by the Lee-Desu statistic (7) for comparison of life-table analyses. RESULTS A total of 97 pregnancies were achieved by ovulation induction (n = 51, 295 cycles) and by IVF (n = 46, 189 cycles). Twenty-seven pregnancies (27.8%) were aborted spontaneously before the 12th week of gestation: 16 in the ovulation induction group, 11 from IVF. Table 2 details the miscarriage rate according to the program and treatment protocol employed. The miscarriage rate per pregnancy in the ovulation induction program with GnRH -a was 3 of 18 (16.7%) compared with 13 of 33 (39.4%) without GnRH -a. The miscarriage rate in the IVF program with GnRH-a was 6 of 33 (18.2%) compared with 5 of 13 (38.5%) without GnRH-a. Because the miscarriage rates in each program with or without Table 2 Number of Pregnancies, Livebirths, and Miscarriages With and Without GnRH-a for Ovulation Induction and IVF Pregnancies Livebirths Miscarriages Ovulation induction GnRH-a No GnRH-a IVF GnRH-a No GnRH-a Total GnRH-a (17.6)* No GnRH-a (39.1)* X2 4.54, P = Values in parentheses are percents. Vol. 59, No.3, March 1993 Homburg et al. GnRH-a reduces miscarriage rate in peas 529

4 GnRH-a are so similar, they have been pooled. Thus, the miscarriage rate when employing GnRH -a was 9 of 51 (17.6%) compared with 18 of 46 (39.1%) without GnRH -a. This is a statistically significant difference: X2 4.54, P = 0.03,95% confidence limits 0.12 and 0.926, odds ratio The cumulative livebirth rates for treatment with and without GnRH-a are shown in Figure 1. A total of 484 cycles have been analyzed: 216 with GnRH-a and 268 with gonadotropins only. After four treatment cycles, the cumulative livebirth rate with GnRH-a was 64% compared with 26% without GnRH -a. A statistical analysis of the difference in the two curves over four and six treatment cycles, employing the Lee-Desu statistic (7), revealed a significantly improved outcome with the use of GnRH-a (P < 0.05 and P = 0.06, respectively). Multiple pregnancies, which may be associated with reduced livebirth rates, occurred on 10 occasions with GnRH-a (9 sets of twins, 1 quadruplet) and twice without GnRH-a (1 twin, 1 triplet). All resulted in livebirths. DISCUSSION The high rate of miscarriage after the induction of ovulation in women with PCOS is a troublesome and frustrating complication. After the initial hypothesis proposed by Jacobs et al. (8) that inappropriate secretion of LH impairs fertility, an adverse effect of high tonic levels of LH in the follicular phase on pregnancy rates (PRs) in IVF has been reported (9-11). Homburg et al. (2) found a similar association during ovulation induction of women with PC OS and, in addition, that the occurrence of miscarriage was also associated with elevated, tonic follicular phase serum LH concentrations. This finding has been strengthened by several clinical correlations. Sagle et al. (12) reported that ultrasonically diagnosed polycystic ovaries were present in some 80% of women with recurrent SABs. Regan et al. (4) found a striking association of raised follicular phase serum LH concentrations with the miscarriage rate in a study of 194 women with regular menstrual cycles. The miscarriage rate in women with normal LH levels «10 lull) was 12% compared with 64% in those with elevated LH concentrations. The obvious test of this clinical association and the most appropriate proposed treatment available is GnRH -a to block endogenous gonadotropin secretion, followed by gonadotropic stimulation of the o..l11jletl\19 11\19 birth rate '11 80'11 70'11 60'11 50'11 40'11 30'11 20'11 10'11. " <' ",,-,,,,-f'"... 0'1I -----~----~----~-----L ~ o 234 Treatment cycle - GnRH-a.+. hmgifsh 5 8 Figure 1 Cumulative livebirth rates, calculated by life table analysis, for treatment in cycles (n = 484) with (n = 216) and without GnRH a (n = 268). ovaries, in an attempt to reduce unscheduled LH release. A previous, randomized, controlled study in patients with PCOS compared results of induction of ovulation with hmg or pfsh with or without buserelin acetate (13). The number of pregnancies in the group of 46 patients was too small to assess the effect of buserelin acetate on the miscarriage rate. A more encouraging study was that of Johnson and Pearce (3) in a group of women who had PCOS associated with recurrent SAB after CC treatment. They reported that 48% of women randomized to receive CC had a repeat spontaneous miscarriage compared with 9% who received buserelin acetate and FSH. In the present study, we have demonstrated that the cotreatment of GnRH-a and hmg is capable of reducing the miscarriage rate by more than half, compared with the use of gonadotropins only, in a large group of patients with PCOS undergoing induction of ovulation or IVF. It is noteworthy that the miscarriage rates in each program with or without GnRH-a were remarkably similar. The improved outcome of pregnancy with the use of GnRH-a in this group of difficult patients is prominently demonstrated by the cumulative livebirth rates. The innovative use of this method rather than the PR per cycle or, indeed, the cumulative conception rate, which gives couples knowledge of the probability of conception only, allows an assessment of the most important outcome to the infertile couple, i.e., the probability of their treatment resulting in the birth of a baby. This method of expressing results and comparing treatment modes therefore reflects a more realistic and practical knowledge to patients 530 Homburg et a1. GnRH-a reduces miscarriage rate in peas Fertility and Sterility

5 and physicians alike, and its use should be encouraged. Although the present study was retrospective, no selection criteria were used for the allocation ofpatients into groups of treatment with and without GnRH-a, and the basal clinical and laboratory parameters of the groups compared did not differ. In particular, between the groups compared, there was no difference in possible predisposing factors for miscarriage such as the number of previous miscarriages, age, and BMI, and there were more multiple pregnancies in the GnRH-a patients. In the ovulation induction group, luteal phase support was given only to the GnRH -a patients, and it may be argued that perhaps it was the luteal support that caused this group to have a lower rate of miscarriage. However, our previous experience in women with PCOS has shown no advantage of giving luteal phase support after induction of ovulation with hmg alone. Should this observation that GnRH-a is capable of significantly improving the livebirth rate of women with PCOS by reducing the number of miscarriages be supported by prospective, randomized studies, its addition to treatment protocols for both ovulation induction and IVF may alleviate a serious clinical problem. Acknowledgment. We thank Navah Jelin, M.A., Central Office, Kupat Holim, Israel, for her statistical advice. REFERENCES 1. Urman B, Fluker MR, Ho Yuen B, Fleige-Zahradka BG, Zouves CG, Moon YS. The outcome of in vitro fertilization and embryo transfer in women with polycystic ovary syndrome failing to conceive after ovulation induction with exogenous gonadotropins. Fertil SterilI992;57: Homburg R, Armar NA, Eshel A, Adams J, Jacobs HS. Influence of serum luteinising hormone concentrations on ovulation, conception and early pregnancy loss in polycystic ovary syndrome. Br Med J 1988;297: Johnson P, Pearce MJ. Recurrent spontaneous abortion and polycystic ovarian disease: comparison of two regimens to induce ovulation. Br Med J 1990;300: Regan L, Owen EJ, Jacobs HS. Hypersecretion ofluteinising hormone, infertility and miscarriage. Lancet 1990;336: Adams J, Polson DW, Abdulwahid N, Morris DV, Franks S, Mason HD, et al. Multifollicular ovaries: clinical and endocrine features and response to pulsatile gonadotropin-releasing hormone. Lancet 1985;2: Cooke ID, Sulaiman RA, Lenton EA, Parsons J. Fertility and infertility statistics: their importance and application. Clin Obstet GynaecolI981;8: Lee E, Desu M. A computer program for preparing K samples with right censored data. Comput Programs Biomed 1972;2: Jacobs HS, Porter RN, Eshel A, Craft I. Profertility uses of LHRH analogues. In: Vickery BH, Nestor JJ Jr, editors. LHRH and its analogues: contraceptive and therapeutic applications, part 2. Lancaster: MTP Press, 1987: Stanger JD, Yovich JL. Reduced in vitro fertilisation of hu man oocytes from patients with raised basalluteinising hormone levels during the follicular phase. Br J Obstet Gynaecol 1985;92: Howles CM, MacNamee MC, Edwards RG, Goswamy R, Steptoe PC. Effect of high tonic levels of luteinising hormone on outcome of in vitro fertilisation. Lancet 1986;2: Punnonen R, Ashorn R, Vilja P, Heinonen PK, Kujansuu E, Tuohimaa P. Spontaneous luteinizing hormone surge and cleavage of in vitro fertilized embryos. Fertil Steril 1988;49: Sagle M, Bishop K, Ridley N, Alexander FM, Michel M, Bonney RC, et al. Recurrent early miscarriage and polycystic ovaries. Br Med J 1988;297: Homburg R, Eshel A, Kilborn J, Adams J, Jacobs HS. Combined luteinising hormone releasing hormone analogue and exogenous gonadotropins for the treatment of infertility associated with polycystic ovaries. Hum Reprod 1990;5:32-5. Vol. 59, No.3, March 1993 Homburg et al. GnRH-a reduces miscarriage rate in peds. 531

In vitro fertilization and embryo transfer for the treatment of infertility associated with polycystic ovary syndrome

Assisted reproductive techno.logy FERTILITY AND STERILITY Vol. 60, No.5, November 1993 Copyright 199a The American Fertility Society Printed on acid-free paper in U. S. A. In vitro fertilization and embryo