Lars G.Westergaard 1, Karin Erb, Steen Laursen, Per E.Rasmussen and Sven Rex

Transcription

1 Human Reproduction vol.11 no.6 pp , 19% The effect of human menopausal gonadotrophin and highly purified, urine-derived follicle stimulating hormone on the outcome of in-vitro fertuization in down-regulated normogonadotrophic women Lars G.Westergaard 1, Karin Erb, Steen Laursen, Per E.Rasmussen and Sven Rex Fertility Clinic, Department of Obstetrics and Gynaecology, Odense University Hospital, DK 5000 Odense C, Denmark 'To whom correspondence should be addressed It has been suggested that the luteinizing hormone (LH) activity of human menopausal gonadotrophin () preparations used for ovarian stimulation in in-vitro fertilization (TVF) may have adverse effects on reproductive outcome. In the present prospective, randomized trial of 218 infertile couples this notion was investigated. A total of 114 women were treated with Pergonal ( group) and 104 with Fertinorm HP ( group). The two groups were comparable with regard to duration of infertility, cause of infertility, age and number of previous FVF attempts and all had normal basal gonadotrophin concentrations before treatment was started. A standard hormonal treatment consisting of pituitary down-regulation with gonadotrophin-releasing hormone analogue (GnRHa) for 14 days starting on cycle day 21, followed by either or highly purified follicle stimulating hormone (), three ampoules (225 IU) per day for 7 days, was used in all cases. The daily hormone dose was thereafter individualized according to the ovarian response. A maximum of two preembryos were transferred after 3 days of culture. Luteal support with progesterone (300 mg per day intravaginally) was used in all cases. Serum concentrations of oestradiol, FSH and LH were measured on days 1 and 8 of stimulation and on the day of oocyte retrieval. The mean number of days of stimulation, mean number of ampoules of or used, mean total motile sperm count on the day of oocyte retrieval and mean numbers of oocytes retrieved (13.4 versus 13.7) or pre-embryos transferred (1.8 versus 1.8) were similar for both groups. Significantly (P < 0.05) more cycles in the group (17 = 16%) were cancelled due to complete failure of fertilization than in the group (7 = 6%). The mean fertilization rate was significantly (P < 0.05) higher in the group (56%) than in the group (50%), and significantly more transferable pre-embryos were obtained in the than in the group (mean: 4.0 versus 3.2; P < 0.01). Serum hormone concentrations were similar in the two groups on stimulation day 1, but differed significantly with regard to FSH, LH and oestradiol on stimulation day 8. The clinical outcome was similar in the two groups, with an ongoing pregnancy rate (>12 weeks of gestation) per started cycle of 33% in the group and 29% in the group. The clinical abortion rates were similar European Society for Human Reproduction and Embryology (10 and 14%), and the implantation rate was 30% in each group. In conclusion, no detrimental effect of the LH activity of on the clinical outcome of FVF in GnRHa down-regulated normogonadotrophic women was found. To the contrary, some beneficial effects of on fertilization rates and pre-embryo development as compared with HP- FSH were demonstrated. These effects, as well as the differences in serum hormone concentrations during ovarian stimulation, may be caused by differences in LH content and/or in the composition of FSH isoforms of the and preparations. Key words: GnRHa down-regulation//ivf/lh content/ ovarian stimulation Introduction Since the earnest days of in-vitro fertilization (TVF) human menopausal gonadotrophins () with or without the addition of clomiphene citrate have been administered to normogonadotrophic women to induce multiple follicular growth in order to maximize the yield of oocytes for fertilization in vitro. However, with these stimulation methods, a considerable number of treatment cycles had to be cancelled due to the frequent occurrence of an untimely, spontaneous luteinizing hormone (LH) surge. In addition, a number of studies have shown that increased concentrations of LH during the follicular phase of the menstrual cycle may be associated with reduced rates of fertilization and implantation and increased miscarriage rates. This applied to patients with polycystic ovarian disease as well as to women with normal cycles, and was found not only in women undergoing treatment with fertility drugs but also in the normal population (Stanger and Yovich, 1985; Howies et al, 1986; Homburg et al., 1988; Punnonen et al, 1988; Regan et al., 1989). Use of gonadotrophin-releasing hormone agonists (GnRHa) to suppress the endogenous gonadotrophin secretion during ovarian stimulation with follicle stimulating hormone (FSH) or proved to be an effective method to lower basal LH secretion during the follicular phase and to prevent an untimely mid-cycle LH surge (Fleming and Coutts, 1986). A large number of studies have shown that this strategy significantly improves clinical pregnancy rates in IVF (Hughes et al., 1992). Whereas the beneficial effect of eliminating spontaneous LH surges in IVF treatment cycles is well established, it is not so certain that the concomitant lowering of basal endogenous LH secretion during the follicular phase also contributes to increased success rates in IVF when using GnRHa downregulation in normogonadotrophic women. However, even 1209

2 L.G.Westergaard et at. after GnRHa down-regulation, resting concentrations of LH seem to be sufficient to provide maximal support for thecal cell androgen synthesis, since <1% of LH receptors need to be occupied to elicit a maximal steroidogenic response (Chappel and Howies, 1991). The many recent reports of good results with the use of pure FSH or even recombinant human FSH in combination with GnRHa down-regulation in FVF support this notion (Scoccia et al, 1987; Edelstein et al, 1990; Devroey et al, 1993). Indeed, it has been proposed that exogenous LH administration for ovarian stimulation may not be required and even may be detrimental to follicle growth and oocyte maturation in normogonadotrophic women (Chappel and Howies, 1991; Daya et al, 1995). Consequently, replacing, which contains equal amounts of FSH and LH-like activity, with preparations containing only FSH would be a rational approach to ovarian stimulation in IVF cycles. The aim of the present study was to answer the question: does the LH component in preparations make a difference with regard to the outcome of IVF treatment in GnRHa downregulated normogonadotrophic women? Materials and methods During the period October 1994 to April 1995, 218 women undergoing IVF treatment at the Fertility Clinic, Odense University Hospital, Denmark, were consecutively recruited to participate in this study. Each individual patient was given oral and written information about the project before consenting to participate by signature. The project was approved by the Ethical Committee of the counties of Fyn and Vejle. The inclusion criteria were (i) age <40 years; (ii) normal menstrual cycle ranging from 26 to 32 days and normal pretreatment scrum concentrations of FSH and LH, i.e. on cycle day 2 <6 months before start of treatment; (iii) standard ovarian stimulation regimen planned and used (see below). The exclusion criteria were (i) infertility caused by endocrine abnormality and (ii) cases in which intracytoplasmic sperm injection (ICSI) or donor semen was used. Study design This was a prospective, randomized study comparing the effects of (Pergonal; Serono Nordic, Some, Sweden) and highly purified (HP)-FSH (Fertinorm HP; Serono Nordic). Method of randomization On their first day of menstruation, patients contacted the clinic to be scheduled to start treatment on day 21. According to the date of first day of menstruation, patients were allocated to either Pergonal (even dates) or Fertinorm HP (odd dates). Information about the project was given 3 weeks later on cycle day 21, when the patient started down-regulation with GnRHa (Suprefact; Hoechst, Copenhagen, Denmark). Consent to participate was given 14 days later, when ovarian stimulation was started. Only after giving consent was a patient informed that the date of first day of bleeding was the factor deciding to which group she was allocated. Patients declining participation were given the standard preparation used by the clinic, i.e. Pergonal. Hormonal treatment The standard hormonal treatment used in this study consisted of the following elements: (i) down-regulation with GnRHa, 0.5 mg s.c. daily for a minimum of 14 days. When down-regulation was confirmed by a serum oestradiol concentration <100 pmol/1 and ultrasound, the 1210 GnRHa was reduced to 0.2 mg s.c. daily until the day of human chorionic gonadotrophin (HCG) injection; (ii) ovarian stimulation with a fixed dose of 225 IU per day for 7 days of either or administered to each patient. On day 8 of stimulation, ovarian response was monitored by ultrasound and serum oestradiol concentrations and the daily dosage of / was individualized accordingly. When at least four follicles with diameter > 17 mm were seen, an ovulatory dose of HCG (Profasi, IU; Serono Nordic) was given by i.m. injection. Blood sampling In all patients blood was sampled routinely firstly on the day downregulation was confirmed and ovarian stimulation initiated (stimulation day 1), before the first injection of or was given; secondly on stimulation day 8 before injection was given and 18 h after the last injection of / and finally on the day of oocyte retrieval, i.e. 36 h after injection of HCG and ~72 h after the last injection of /. The sera were immediately analysed for oestradiol and aliquots frozen at -20 C and stored for later analysis for FSH and LH. Oocyte retrieval, in-vitro culture and pre-embryo transfer At 36 h after HCG injection, oocytes were retrieved by ultrasound guidance per vaginam. Aspirated oocytes were cultured in IVF medium (catalogue no. 1031; Medicult, Copenhagen, Denmark) in air containing 5% COj and 100% humidity. Freshly ejaculated semen was prepared by Percoll gradient centrifugation and, 4 h after retrieval, each oocyte was inseminated by adding spermatozoa to the culture medium. Each oocyte was checked for fertilization 24 h after oocyte retrieval and then transferred to fresh IVF medium. At 48 h after retrieval, the developmental stage of the pre-embryo was checked under an inverted microscope and it was then transferred to M3 medium (Medicult) and cultured for an additional 24 h. At 72 h after oocyte retrieval a maximum of two normally developed preembryos were transferred to the uterine cavity using a Wallace embryo transfer catheter. Only pre-embryos with between 6 and 12 evensized blastomeres and <20% fragmentation as seen under the inverted microscope were considered transferable. Surplus transferable preembryos were cryopreserved using dimethyl sulphoxide (DMSO; Sigma, catalogue no. D 2650) as cryoprotectant. Luteal phase support Each patient was given micronized progesterone pills (Progestane; Organon, Skorlunde, Denmark), 100 mgx3 per day, administered intravaginally starting 2 days before embryo transfer and continuing until the pregnancy test, i.e days after embryo transfer. In cases of a positive pregnancy test, luteal support was continued for 3 weeks more until a viable intrauterine pregnancy could be diagnosed by ultrasound. Hormone assays Oestradiol was measured by fluoroimmunoassay (AutoDELPHIA oestradiol kit; Wallac, Denmark A/S, Allerod, Denmark) in serum routinely on stimulation days 1 and 8 and on the day of oocyte retrieval. FSH and LH were measured on the same days using an automated fluoroimmunoassay system (AutoDELPHIA hfsh and hlh sp kits; Wallac). The intra- and interassay coefficients of variation for oestradiol were 4.2 and 3.6%, for FSH 2.3 and 4.5% and for LH 1.9 and 3.8% respectively. Statistical methods Continuous data were summarized using means and SEM. Differences between the two groups were tested using Student's r-test and % 2 test. A P value <0.05 was considered significant.

3 Effect of and highly purified FSH on IVF Table I. Demographic characteristics of patients receiving either human menopausal gonadotrophin () or highly purified follicle stimulating hormone () for ovanan stimulation Table n. Serum concentrations of oestradiol, follicle stimulating hormone (FSH) and luteinizing hormone (LH) on stimulation day 1, stimulation day 8 and the day of oocyte retrieval. Values are mean ± SEM P values No. of patients (cycles) Median (range) age (years) 32 (24-40) 33 (24-38) Median (range) duration of 5(1-15) 5(1-15) infertility (years) Infertility diagnosis Tubal factor (%) 84 (74) 79 (76) Idiopathic (%) 14(12) 14(13) Male factor (%) 13(11) 10(10) Other (%) 3(3) 1(1) Pretreatment basal concentrations (IU/1) on cycle day 2 FSH (mean ± SEM) 7.7 ± ± 1.2 LH (mean ± SEM) 5.7 ± ± 0.4 LH = luteinizing hormone. Stimulation day 1 Oestradiol (pmol/1) Stimulation day 8 Oestradiol (pmol/1) Day of oocyte retrieval 3.0 ± ± ± ± ± ± ± ± ± 0.4 < ±0.1 < ± 229 < ± ± 0.8 < ± ± 0.02 follicle stimulating hormone; = not significant Results The study comprised a total of 218 patients, of whom 114 were allocated to treatment with in 114 cycles and 104 to treatment with in 104 cycles. The and HP- FSH groups were comparable with regard to age, duration of infertility, infertility diagnosis and pretreatment basal gonadotrophin concentrations (Table I). The numbers of previous IVF attempts were similar in the two groups. Serum concentrations on stimulation day 1 of oestradiol, FSH and LH after 14 days of GnRHa down-regulation were similar in the two groups (Table II). On stimulation day 8, i.e. when all patients had received 21 ampoules of either (21X225 IU FSH and 21X225 IU LH) or (21X225 IU FSH), concentrations of FSH had increased 3- to 4-fold in both groups, but reached a significantly higher value in the group (P < 0.001, Table IT). LH declined in both groups from stimulation day 1 to 8, with the values in the HP- FSH group being significantly lower than in the group (P < 0.02). Oestradiol concentrations on stimulation day 8 were significantly higher in the group than in the HP- FSH group (P < 0.01). On the day of oocyte retrieval all hormone concentrations had declined compared to stimulation day 8. Concentrations of FSH in the group were still significantly higher than those in the group (P < 0.001), whereas LH concentrations were similar in the two groups. The outcome of ovarian stimulation, oocyte retrieval, IVF and pre-embryo development is shown in Table III. Two of the 104 patients (cycles) in the group were cancelled before oocyte retrieval due to poor response (low oestradiol and less than four mature follicles), in contrast to none of the 114 patients in the group. The duration of stimulation (days) and the number of ampoules of or used were identical in the two groups. The harvest of oocytes per cycle was similar in the two groups (13.4 ± 0.6 versus 13.7 ± 0.7). The quality of semen, expressed as total number of normal, motile spermatozoa in the ejaculate obtained after Percoll preparation on the day of oocyte retrieval, was similar in the two groups (Table IE). Significantly (P < 0.05) more oocytes were fertilized in the group (854/1532 oocytes = 56%) compared to the group (694/1397 oocytes = Table HI. Outcome of ovarian stimulation, oocyte retrieval, in-vitro fertilization and pre-embryo development Values are mean ± SEM per cycle No. of cycles started (%) No. of cycles with oocyte retrieval (%) No. of cycles with embryo transfer (%) Duration of stimulation (days) No. of ampoules used No. of oocytes retrieved Semen quality on day of oocyte retrieval 1 Fertilization rate (%) No. of transferable preembryos No. of transferred preembryos P value 114 (100) 114(100) 107 (94) 9.3 ± ± ± ± ± ± (100) 102 (98) 85 (82) 9.3 ± ± ± ± ± ± 0.1 <0.05 <0.005 <0.01 follicle stimulating hormone. "Total number of spermatozoa with normal motility per ejaculate obtained after Percoll gradient centrifugation. 50%; Table DT). In addition, significantly more cycles in the group were cancelled before pre-embryo transfer due to complete failure of fertilization or poor pre-embryo development (18 versus 6%; P < 0.05). Significantly more transferable pre-embryos were obtained after ovarian stimulation with than with (mean ± SEM per cycle: 4.0 ± 0.3 versus 3.2 ± 0.4; P < 0.01, Table HI). The mean number of transferred pre-embryos per cycle was similar in the two groups (1.7 ± 0.1 versus 1.5 ± 0.1; Table IU). The clinical outcome of IVF/embryo transfer did not differ significantly between the two groups as shown in Table IV. A clinical pregnancy was obtained in 36% of the started cycles in the group and in 34% of those in the group, and ongoing pregnancy rates were 32 and 29% respectively. The rate of spontaneous, clinical abortions in the group (4/41 = 10%) was similar to that in the group (5/ 35 = 14%). The rate of transferred pre-embryos which implanted was 30% in both groups (58/195 transferred pre- 1211

4 L.G.Westergaard et al Table IV. Pregnancies, abortions and implantation rate Clinical pregnancy rate 1 Per cycle (%) Per embryo transfer (%) Ongoing pregnancy rate b Per cycle (%) Per embryo transfer (%) Clinical abortion rate (%) Implantation rate c (%) 41/114(36) 41/107 (38) 37/114(32) 37/107 (35) 4/41 (10) 58/195 (30) 35/104 (34) 35/85 (41) 30/104 (29) 30/85 (35) 5/35 (14) 46/154 (30) follicle stimulating hormone. Diagnosed by ultrasound at 7th week of gestation. ''Ongoing intrauterine pregnancy beyond 12th week of gestation. c Number of intrauterine sacs verified by ultrasound (5 weeks after embryo transfer) as a percentage of transferred pre-embryos. embryos in the group and 46/154 in the group). No ectopic pregnancies occurred. Discussion This prospective, randomized study did not demonstrate a detrimental effect of the exogenous LH-like activity contained in on the clinical outcome of FVF in GnRHa downregulated normogonadotrophic women. Although the clinical outcome of FVF in and treated women was similar, our results indicate that exogenous LH does make a difference. Thus, the -treated patients ran a significantly lower risk of complete failure of fertilization and produced significantly more transferable pre-embryos than those treated with. The reason for this difference is not readily explained, but since the two groups were comparable with regard to anamnestic and clinical parameters, and since the quality of semen used for fertilization in vitro was similar, a difference in oocyte quality, i.e. maturity and/or fertilizabiliry, is possible. In contrast to FSH and due to its much shorter half-life, serum concentrations of LH do not increase during treatment (Diczfalusy and Harlin, 1988) and, in GnRHa down-regulated women treated with serum, LH has been reported to remain low and similar to that in women receiving pure FSH (Duijkers et al, 1993). The present results confirm these reports, showing declining LH concentrations in the group from stimulation day 1 through 8 to the day of oocyte retrieval, despite the fact that these patients had, on average, received 28.6 ampoules of or 2145 IU of exogenous LH during that period. This, however, does not imply that growing follicles in women undergoing treatment are not exposed to more LH than in women given pure FSH. As a matter of fact, in the present study on stimulation day 8, i.e. when all patients had received 21 ampoules of either or, we found that serum LH concentrations, although low, were significantly higher in the group than in the group. Despite significantly higher concentrations of FSH in the serum in the HP- FSH group, concentrations of oestradiol were significantly lower in that group compared to the group. This points to a biological effect of exogenous LH-like activity in 1212 stimulating follicular androgen synthesis and aromatization more efficiently (faster?) than. In a recent study by Fleming et al. (1995), which is the only other study known to us that compared the effects of and in GnRHa down-regulated women, similar differences in serum oestradiol and gonadotrophins were found. Moreover, it was found that a subgroup of patients, consisting of nearly half (17 of 40) of those treated with, showed consistently low serum LH (<1 IU/1) and significantly lower serum concentrations of oestradiol. In addition, oestradiol and testosterone concentrations in follicular fluid were lower than those found in -treated patients with serum LH >1 IU/1 (23 of 40) and in all patients {n - 20) treated with (Fleming et al., 1995). The authors concluded from these data that minimal circulating LH concentrations (>1.0 IU/1 by immunoassay) are required for normal oestradiol biosynthesis and secretion, but suppression below 1 IU/1 influences hormone responses. The present results confirm this. Most probably, the process of pre-ovulatory oocyte maturation is not directly coupled to the regulation of follicular steroidogenesis (Westergaard, 1988). However, combining the results of Fleming et al. (1995) with those of the present study, it may be speculated that the explanation for the poorer outcome in the group, with regard to fertilization and pre-embryo development, is that a substantial subgroup of normogonadotrophic women require more than resting concentrations of LH for optimal oocyte maturation, and thus do less well on pure FSH stimulation. Another explanation for the above differences, in theory at least, could be that the composition of FSH isoforms in on the whole is more 'natural' than in preparations and thus more optimal for follicular and oocyte maturation. Previous studies have shown significant changes in the circulating isoforms of FSH throughout the menstrual cycle in women (Padmanabhan et al., 1988; Wide and Bakos, 1993). Moreover, it has recently been shown that the distribution of FSH isoforms in Metrodin-HP (equivalent to Fertinorm HP) deviates significantly from that in Metrodin,. recombinant human FSH and pituitary FSH, being far more acidic in nature than in the latter (Lambert et al., 1995). Since acidic isoforms have a longer half-life than basic isoforms of FSH, it might be expected that stays longer and accumulates to higher concentrations in serum than do other FSH preparations. The present results confirm this, since on stimulation day 8 we found significantly higher concentrations of FSH in the serum of the group than in that of the group, despite the fact that both groups had received equal amounts of exogenous FSH, i.e. 21 ampoules or 1575 IU of FSH. In the study by Lambert et al. (1995), Metrodin-HP was also found to possess a significantly lower biopotency than Metrodin or recombinant FSH as measured by oestradiol production in a rat Sertoli cell bioassay. The results of the present study showing significantly lower oestradiol concentrations on stimulation day 8 in the group despite a significantly higher concentration of FSH in serum than in the group, support the notion that the biopotency of HP- FSH in vivo may be lower than that of other FSH preparations, such as.

5 Effect of and highly purified FSH on IVF In a randomized, prospective study of a similar size to the present study, Daya et al. (1995) compared the effects of FSH (Metrodin) and in IVF and found, in contrast to the present results, a statistically higher fertilization rate in patients treated with FSH compared to. Although, as in the present study, no statistically significant difference in the clinical outcome was found between the groups, it was concluded that the LH activity of accounted for the poorer results in the group compared to the FSH group. However, a number of differences between our study and that of Daya et al (1995) may explain this discrepancy. In the study of Daya et al. GnRHa was used in a flare-up protocol, and down-regulation of LH was assumed but not documented by serum LH measurements as in the present study. Moreover, different FSH preparations were used in the two studies and, in relation to the discussion above on FSH isoforms, this could be an essential difference. Notwithstanding these theoretical considerations, the present results clearly demonstrate that the LH-like activity in preparations does not negatively affect the outcome of IVF in most GnRHa down-regulated, normogonadotrophic women. However, there may be other good arguments for considering replacement of with highly purified urinary FSH or even recombinant human FSH for ovarian stimulation. Currently available gonadotrophin preparations such as and Metrodin are impure (<3% of the total protein content being FSH) and are therefore indicated for i.m. administration only (Howies et al, 1994). By contrast, and recombinant FSH are much more well defined, with >95% of total protein being FSH, and with a specific activity exceeding IU FSH per mg of protein (Howies et al, 1994; le Cotonnec et al, 1994). Thus, unknown proteins in and recombinant FSH are virtually absent and s.c. administration by the patient herself is a possible and attractive option. One serious drawback to the use of and recombinant FSH is that the price of these preparations is two to three times higher than that of. Evidently, the general principle that the purest and most well-defined pharmacological preparations should be administered for medical treatment in humans also applies to gonadotrophins. Therefore, it can be expected that the classical urine-derived preparations will be, and should be, replaced by recombinant human gonadotrophins in the near future. In conclusion, the lesson to be learned from the present study is that just replacing with or recombinant human FSH may not be good enough for ovarian stimulation of GnRHa down-regulated, normogonadotrophic women. Investigations in more sophisticated regimes for ovarian stimulation are needed, with use of recombinant human FSH, with an optimal composition of FSH isoforms, and eventually combined with recombinant human LH or HCG in varying amounts, tailored for the individual patient of follicle stimulating hormone versus human menopausal gonadotrophin in in-vitro fertilization. Hum. Reprod., 10, 392-3%. Devroey, P., Mannaerts, B., Smitz, J. et al. (1993) First established pregnancy and birth after ovarian stimulation with recombinant human follicle stimulating hormone (Org ). Hum. Reprod., 8, Diczfalusy, E. and Harlin, J. (1988) Clinical-pharmacological studies on human menopausal gonadotrophin. Hum. Reprod., 3, Duijkers, U.M., Vemer, H.M., Hollanders, J.M.G. el al. (1993) Different follicle stimulating hormone/luteinizing hormone ratios for ovarian stimulation. Hum. Reprod., 8, Edelstein, M.C., Brzyski, R.G., Jones, G.S. et al. (1990) Equivalency of human menopausal gonadotrophin and follicle stimulating hormone stimulation after gonadotrophin-releasing hormone agonist suppression. FertiL Sleril., 53, Fleming, R. and Coutts, J.R.T. (1986) Induction of multiple follicular growth in normally menstruating women with endogenous gonadotrophin suppression. FertiL Sterii, 45, Fleming, R., Chung, C.C., Yates, R.W.S. et al. (1995) Reduced ovarian hormone production in patients treated with purified FSH and a GnRHanalogue: relationship with LH activity. Hum. Reprod., 10 (Suppl. 1), Abstr. FC 15. Homburg, R., Amar, N.A., Eshel, A. et al. (1988) Influence of serum lulcinizing hormone concentrations on ovulation, conception, and early pregnancy loss in polycystic ovarian syndrome. Br. Med. J., 297, Howies, CM., Macnamee, M.C., Edwards, R.G. et al. (1986) Effect of high tonic levels of luteinizing hormone on outcome of in vitro fertilisation. Lancet, tt, Howies, CM., Loumaye, E., Giroud, D. and Luyet, G. (1994) Multiple follicular development and ovarian steroidogenesis following administration of a highly purified urinary FSH preparation in pituitary desensitized women undergoing FVF: a multicentre European phase in study. Hum. Reprod., 9, 424-^30. Hughes, E.G., Fedorkow, P.M., Daya, S. et al. (1992) The routine use of gonadotropin-releasing hormone agonist prior to in vitro fertilization and gamete intra Fallopian transfer a metaanalysis of randomized controlled trials. FertiL Sterii., 58, Lambert, A., Rodgers, M., Mitchell, R. et al. (1995) In-vitro biopotency and glycoform distribution of recombinant human follicle stimulating hormone (Org 32489), Metrodin and Metrodin-HP. Hum. Reprod, 10, le Cotonnec, J.-Y., Porchet, H.C et al. (1994) Clinical pharmacology of recombinant human follicle-stimulating hormone (FSH). I. Comparative pharmacokinetics with urinary FSH. FertiL SteriL, 61, Padmanabhan, V., Lang, L.L., Sonstein, J. et al. (1988) Modulation of serum follicle stimulating hormone bioactivity and isoform distribution by estrogenic steroids in normal women and in gonadal dysgenesis. J. Clin. Endocrinol. Metab., 67, Punnonen, R., Ashom, R., Vilja, P. et al. (1988) Spontaneous luteinizing hormone surge and cleavage of in vitro fertilised embryos. FertiL SteriL, 49, Regan, L., Owen, EJ. and Jacobs, H.S. (1989) Hypersecretion of luteinizing hormone, infertility and miscarriage. Lancet, 336, Scoccia, B., Blumenthal, P., Wagner, C. et al. (1987) Comparison of urinary human follicle stimulating hormone and human menopausal gonadotrophins for ovarian stimulation in an in vitro fertilization program. FertiL SteriL, 48,446-^49. Stanger, J.D. and Yovich, J.L. (1985) Reduced in vitro fertilisation of human oocytes from patients with raised basal luteinizing hormone levels during the follicular phase. Br. J. ObsteL GynaecoL, 92, Westergaard, L.G. (1988) Intrafollicular factors regulating human ovarian follicular development and oocyte meiosis. PhD thesis. Dan. Med. Bull., 35, Wide, L. and Bakos, A. (1993) More basic forms of both follicle-stimulating hormone and luteinizing hormone in serum at midcycle compared with follicular or luteal phase. /. Clin. EndocrinoL Metab., 76, Received on October 2, 1995; accepted on March References Chappel, S.C. and Howies, C. (1991) Reevaluation of the roles of luteinizing hormone and follicle stimulating hormone in the ovulatory process. Hum. Reprod., 6, Daya, S., Gunby, J., Hughes, E.G. et al. (1995) Randomized controlled trial 1213