Recombinant human LH supplementation during GnRH antagonist administration in IVF/ICSI cycles: a prospective randomized study

I. Cédrin-Durnerin1,4, D. Grange-Dujardin2, A. Laffy2, I. Parneix3, N. Massin1, J. Galey1, L. Théron1, J.P. Wolf1, C. Conord2, P. Clément2, S. Jayot3 and J.N. Hugues1

1 Centre for Reproductive Medicine, Jean Verdier Hospital, A.P.-H.P., 93143 Bondy – University Paris XIII, 2 ART Centre, Clinique du Blanc Mesnil, 93150 Le Blanc Mesnil and 3 IVF Centre, Aquitaine Santé Médecine de la Reproduction, 33 523 Bruges Cedex, France

4 To whom correspondence should be addressed. Email: isabelle.cedrin-durnerin{at}jvr.ap-hop-paris.fr


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: When administered in the late follicular phase to prevent an LH surge, GnRH antagonists induce a sharp decrease in serum LH levels that may be detrimental for assisted reproductive technology cycle outcome. Therefore, a prospective study was designed to assess the effects of recombinant human (r)LH supplementation during GnRH antagonist (cetrorelix) administration. METHODS: The protocol consisted of cycle programming with oral contraceptive pill, ovarian stimulation with rFSH and flexible administration of a single dose of cetrorelix (3 mg). A total of 218 patients from three IVF centres were randomized (by sealed envelops or according to woman's birth date) to receive (n=114) or not (n=104) a daily injection of rLH 75 IU from GnRH antagonist initiation to hCG injection. RESULTS: The only significant difference was a higher serum peak E2 level in patients treated with rLH (1476±787 versus 1012±659 pg/ml, P<0.001) whereas the numbers of oocytes and embryos as well as the delivery rate (25.2 versus 24%) and the implantation rate per embryo (19.1 versus 17.4%) were similar in both groups. CONCLUSIONS: These results show that in an unselected group of patients, there is no evident benefit to supplement GnRH antagonist-treated cycles with rLH.

Key words: assisted reproductive technology/controlled ovarian stimulation/GnRH antagonist/rLH


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
While protocols using GnRH agonists have been considered as the standard protocol for patients undergoing assisted reproductive technologies during the last two decades, the recent introduction of GnRH antagonists offers the opportunity to control the endogenous LH surge in a rapid and more convenient way. Indeed, GnRH antagonists act on gonadotroph secretion through an immediate competitive blockade of GnRH receptors and have induced a marked decrease in serum LH levels and a less pronounced decrease in FSH secretion (Albano et al., 1997Go; Fujimoto et al., 1997Go). Therefore, GnRH antagonists may be administered in the latter stages of the ovarian stimulation to prevent any LH surge (Diedrich et al., 1994Go).

Both multiple dose and single dose protocols proved to be effective in IVF/ICSI cycles (Olivennes et al., 2002Go). However, the optimal timing for GnRH antagonist administration remains to be determined as well as the consequences of this administration on the hormonal ovarian status. Indeed, a ‘minimum threshold’ of LH secretion has to be maintained to ensure adequate steroidogenesis and folliculogenesis (European Recombinant Human LH Study Group, 1998Go). Furthermore, antagonists are usually administered when follicles become increasingly responsive to LH due to the acquisition of LH receptors on granulosa cells and when LH may operate with FSH to achieve full follicular maturity and oocyte competence (Hillier, 2001Go). As a consequence, it may be assumed that the dramatic decrease in LH secretion induced by GnRH administration is likely to be detrimental for the follicle–oocyte complex. As the inhibitory effect of GnRH antagonists upon LH secretion has been shown to be dose dependent (Ganirelix Dose-finding Study Group, 1998Go), the decrease in serum LH values is sharper and more consistent using a single high dose protocol than a multiple low dose one (Reissman et al., 2000Go). Besides, it has been shown in the literature that measurement of the serum LH value is not valid for assessing residual LH bioactivity (Jaakkola et al., 1990Go) or for defining a sub-group of patients that might benefit from exogenous LH administration (Peñarrubia et al., 2003Go). For that reason, supplementation with exogenous recombinant human (r)LH has been suggested as an alternative way to evaluate indirectly the consequences of LH depletion.

Therefore, this study was designed to assess prospectively the effects of recombinant rLH supplementation on the main ovarian parameters and on the outcome of assisted reproduction cycles in patients undergoing IVF/ICSI cycles and using a single dose GnRH antagonist protocol.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Patients
A total of 218 patients from three IVF centres were enrolled in this study. Ovulatory women were included with the following criteria: age 19–38 years; body mass index (BMI) <30 kg/m2; normal ovarian function assessed by hormonal determinations at day 3 of a spontaneous cycle. All patients with a history of low (<5 oocytes) or high (>15 oocytes) ovarian response in a previous IVF/ICSI attempt were excluded from the study. Informed consent was obtained from all patients.

Protocol
Pre-treatment with an oral contraceptive pill (levonorgestrel 0.15 mg + ethynylestradiol 30 µg) was given during the cycle prior to the IVF/ICSI procedure. Three days after pill discontinuation (S1), stimulation was started by daily injection of 150–300 IU of rFSH (Gonal F; Serono SA, France). The starting dose was chosen according to patients' age (150 IU/day, age <35 years; 225 IU/day, age >35 years), BMI (starting dose increased by 75 IU/day if BMI was 27–30 kg/m2) and ovarian responsiveness in previous cycles. This dose was maintained constant for 5 days. From day 6 of the stimulation (S6), rFSH doses were individually adjusted according to hormonal determinations and ultrasound data.

When the leading follicle reached 14–16 mm diameter, patients received a single injection of cetrorelix 3 mg (Cetrotide; Serono SA) and received or not, according to randomization, a daily injection of 75 IU rLH (Luveris; Serono SA) from the time of cetrorelix injection up to hCG administration. In both groups, rFSH dose adjustment was not allowed at the time of cetrorelix administration. If criteria for triggering ovulation were not met within 4 days after cetrorelix administration, additional injections of cetrorelix 0.25 mg per day were performed until hCG administration.

An injection of 10 000 IU hCG (Gonadotrophines Chorioniques Endo; Organon, France) was given when at least three mature (>17 mm) follicles were obtained and oocyte retrieval was performed 36 h later.

Luteal phase was supported by vaginal administration of progesterone 400 mg/day (Utrogestan; Besins International, France) from the day of ovarian puncture to the day of pregnancy test. If a pregnancy occurred, progesterone administration was extended up to the evidence of fetal heart activity at ultrasound.

Objectives and outcomes
The possible deleterious effect of LH depletion induced by antagonist administration on the follicle–oocyte complex was considered to be clinically relevant if a difference in the number of embryos available for transfer could be observed between patients supplemented or not with rLH. Therefore, the primary outcome measure was the number of total embryos per patient. Secondary outcomes were serum estradiol, LH and progesterone values on the day of hCG administration, numbers of total and mature oocytes, pregnancy rate per treated cycle (positive {beta}hCG test), implantation rate per embryo (number of gestational sacs on ultrasound scan/number of transferred embryos) and ongoing pregnancy rate (pregnancy >12 weeks of gestational age).

Sample size
The estimated sample size to detect a difference of 1.5 embryos between groups was 200 patients (one-sided, P<0.05 considered significant). As randomization was not performed on the day of cetrorelix administration, a larger number of patients was included (114 in the group supplemented with rLH and 104 in the group without rLH) to take into account cancellations before cetrorelix administration (Figure 1). Indeed, eight patients were cancelled for a poor ovarian response before cetrorelix administration (six in rLH allocated group and two in the other group) and one for omitting FSH injections for several days (rFSH allocated patient). Six other patients could not be included in the analysis due to the lack of cetrorelix administration (n=4 with two pregnancies in rFSH allocated patients) or to a late administration of cetrorelix on the day of hCG (n=2, one in each group). Therefore, the number of analysed patients was 107 in the group supplemented with rLH and 96 in the group without rLH.



View larger version (25K):
[in this window]
[in a new window]
 
Figure 1. Patient flow through the randomized controlled trial.

 
Randomization
Randomization for rLH supplementation was performed in Centre 1 on day 1 of the stimulation by means of serially numbered, opaque, sealed envelopes. The allocation sequence was generated by a random permutation table and was concealed from clinicians enrolling participants. In Centres 2 and 3, patients were randomized when they received their prescription according to the even or uneven year of the woman's birth. This latter quasi-randomization procedure was applied for clinicians' convenience and explains the unequal number of patients in each group for these two centres (Figure 1). Neither patients nor clinicians were blinded to rLH administration. Patients were included in the study for only one cycle.

Hormonal measurements
Hormonal measurements were carried out using commercially available chemiluminescence immunoassays with an automated Elecsys immunoanalyser (ECLIA; Roche Diagnostics, France). The sensitivity of the assay was 0.1 IU/l for FSH and LH. Intra-assay and inter-assay coefficients of variation (CV) were within 3 and 6% and within 3 and 4% respectively for FSH and LH. The sensitivity of the assay was 5 pg/ml and 0.03 ng/ml for E2 and progesterone respectively. Intra-assay and inter-assay CV were 5 and 10% respectively for E2 and 3 and 5% respectively for progesterone.

Statistical analysis
Results are expressed as mean±SD. Statistical analysis was performed using Statview 4.5 (Abacus concept software). Nominal or continuous variables were analysed with {chi}2-test or analysis of variance adjusted for centre as required. P<0.05 was considered as statistically significant.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Patients were recruited in the study between April 2002 and June 2003. As shown in Table I, both groups were comparable as regards characteristics of patients and ovarian function assessed on day 3 of the cycle.


View this table:
[in this window]
[in a new window]
 
Table I. Characteristics of patients

 
Stimulation phase was started with similar doses of rFSH in both groups (Table II). Cetrorelix was administered at about day 9 of the stimulation when the leading follicle was ~15 mm in diameter and after the same requirement of rFSH in both groups. Three cycles were cancelled after cetrorelix administration in patients not supplemented with rLH for the following reasons: absence of follicular growth (n=1) or drop of serum E2 levels associated with poor follicular response (n=2); whereas no rLH-supplemented cycle was cancelled. When hormonal values were available on the day of cetrorelix administration, serum E2, LH and progesterone values were comparable between groups. In contrast, at the time of hCG administration, patients supplemented with rLH had significantly higher serum E2 values (P<0.001) than patients treated with rFSH alone, whereas serum LH and progesterone values were identical.


View this table:
[in this window]
[in a new window]
 
Table II. Parameters during ovarian stimulation

 
Moreover, no difference was observed between groups as regards the biological parameters (oocytes and embryos) or the outcome of IVF/ICSI procedures (Table III).


View this table:
[in this window]
[in a new window]
 
Table III. Outcome parameters of IVF/ICSI procedures

 
Figure 2 shows the respective patterns of serum LH and E2 values in both groups. Serum LH values decreased from ~3 IU/l at the time of cetrorelix administration to <1 IU/l on the day of hCG. Serum LH levels on day 1 of the stimulation (3 days after pill discontinuation) were available only for patients from centre 1: the mean LH value was 2.1±1.82 IU/l.



View larger version (14K):
[in this window]
[in a new window]
 
Figure 2. Serum estradiol ( ) and LH ( ) patterns in patients treated with rLH + rFSH (•) or rFSH alone ({circ}) from the time of cetrorelix to hCG administration. Results are expressed as mean±95% confidence interval. Data were available on the day of cetrorelix administration (n=114) or 1 day (n=63) or 2 days (n=26) before cetrorelix administration.

 
As final E2 values are not only dependent on the residual LH secretion but also on the number of follicles, it has been suggested that the E2:oocyte ratio can be a valuable index to assess LH requirement. Indeed, the E2:oocyte ratio was significantly higher in patients treated with rLH than in untreated patients (166±93 versus 124±89, P=0.0018). According to the median values in both groups (150 and 100), we analysed the E2:oocyte ratio by splitting values with a 50 pg interval. Figure 3 shows the percentage of patients supplemented or not with rLH in each category. For the lower values of the ratio, most patients were not supplemented with rLH. Considering pregnancy rates, there was only a trend for lower pregnancy rate when the ratio was <150.



View larger version (24K):
[in this window]
[in a new window]
 
Figure 3. Distribution of estradiol (E2):oocyte ratio (50 pg intervals) according to rLH supplementation. The y-axis gives the percentage of patients in each treatment group according to E2:oocyte ratio. Absolute numbers are indicated for each group (198 cycles, two missing data). Ongoing pregnancy rates per puncture and per transfer for the overall number of patients (supplemented or not with rLH) are given in each E2:oocyte ratio category.

 
The timing of cetrorelix administration has also been considered as a determining factor in the successful outcome of GnRH antagonist protocols. However, we did not find any influence of the day of cetrorelix administration related to day 1 of stimulation upon the pregnancy rate (Table IV). Similarly, the time interval between the cetrorelix injection and the hCG administration did not influence the pregnancy rate (data not shown).


View this table:
[in this window]
[in a new window]
 
Table IV. Pregnancy rate (PR) according to the duration of stimulation before cetrorelix administration

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
This study shows that supplementation with rLH in patients treated for IVF/ICSI procedure with a GnRH antagonist protocol is associated with a significant increase in serum peak estradiol values. Indeed, as the time interval between the cetrorelix injection and the hCG administration, and the doses of rFSH used during this period, were similar in both groups, the higher E2 values can be attributed to the use of rLH. These data confirm the key role of LH to ensure adequate steroidogenesis and are consistent with the two gonadotrophins–two cells theory (Short, 1962Go). They also indicate that the administration of a single high dose of cetrorelix induces a marked decrease in bioactive LH. In both groups, serum immunoreactive LH values on the day of hCG administration were below the LH threshold reported in WHO group I anovulatory patients. This absence of difference in serum immunoreactive LH values between groups may be explained by the short half-life of the product, as exogenous LH is no longer measurable in serum 12 h after injection (Porchet et al., 1994Go). However, in patients without rLH supplementation, the lower serum E2 values observed mean that the residual bioactive LH is actually below the minimal threshold required to ensure optimal steroidogenesis. The dose of 75 IU of rLH per day is sufficient to draw E2 towards higher values in most cases but seems insufficient in a small number of patients, as was reported to be the case in WHO group I anovulatory patients (Burguès et al., 2001Go). The contribution of a decreased serum bioactive FSH to the reduced E2 secretion has also been suggested after administration of GnRH antagonist in the late follicular phase of normally cycling women (Thompson et al., 1995Go). Therefore, higher doses of rLH might be necessary to sustain follicular E2 secretion in the presence of falling serum FSH concentrations (Sullivan et al., 1999Go). Conversely, supplementation with rLH does not affect the number of retrieved oocytes and obtained embryos. These results are in line with the concept that folliculogenesis and oogenesis are mainly dependent on FSH supply and that the role of LH as well as estradiol is not primordial in these processes. It is unlikely that tailoring the FSH starting dose could have biased the final of number of retrieved oocytes and obtained embryos between groups. Indeed, similar numbers of rFSH units were used when evaluating on the day of cetrorelix and hCG administration. As the same number of embryos was transferred in both groups, rLH supplementation does not seem to modify the pregnancy rate or implantation rate per embryo. Similar results were reported in studies comparing the effects of hMG versus rFSH concomitantly with cetrorelix in both fresh embryo transfers (Olivennes et al., 1999Go) and freeze–thaw embryo transfer cycles (Asimakopoulos et al., 2002Go). Therefore, our study shows that, in patients included in a GnRH antagonist protocol, the estradiol secretory pattern is not predictive of the IVF/ICSI cycle outcome. It also suggests that changes in estradiol levels induced by GnRH antagonist administration cannot account for the slightly lower pregnancy rates reported in a recent meta-analysis (Al-Inany and Aboulghar, 2002Go).

Supplementation with exogenous rLH has been chosen as an alternative way to evaluate indirectly the consequences of LH depletion because the interpretation of LH values raises many concerns. First, as it is not usual in most assisted reproduction teams to perform hormonal measurements on a daily basis, and as LH release is pulsatile, the interpretation of a single LH determination is limited. Second, it has been shown that measurement of serum immunoreactive LH value is not valid for assessing residual LH bioactivity (Jaakkola et al., 1990Go). For that reason the E2:oocyte ratio has been proposed as a biological measurement of LH requirement (Loumaye et al., 1997Go). Finally, no clear threshold values of deleterious LH levels could be established during assisted reproduction cycles. In anovulatory patients with hypogonadotrophic hypogonadism treated by rFSH alone, no pregnancy can be obtained without any LH supply while follicular growth and ovulation are preserved (European Recombinant Human LH Study Group, 1998Go) except if basal LH values are >1.2 IU/l (O'Dea et al., 2000Go). In GnRH agonist down-regulated cycles, the impact of low LH levels on outcome is still a matter of debate despite numerous studies (Filicori et al., 2002Go) and no obvious criteria could be established to define whose patients may benefit from adding LH to the stimulation protocol (Shoham, 2002Go). In GnRH antagonist-treated cycles where the period of LH decrease is restrained to the end of the follicular phase, the impact of low residual LH levels could be less pronounced. However, low residual LH concentrations and impaired estradiol secretion with increasing doses of antagonist were associated with low implantation rate despite the similar number of oocytes and embryos obtained (Ganirelix Dose-finding Study Group, 1998Go). As a direct effect of the antagonist seems uncertain because antagonist concentrations are not detectable at the time of transfer, an endometrial impact of low LH levels has been suggested. The direct action of LH on uterine receptivity has recently been hypothesized in the model of oocyte donation. In spite of adequate estrogen and progesterone preparation of the endometrium, a beneficial effect of hCG administration could be observed in patients with low LH levels (Tesarik et al., 2003Go). The trend to lower pregnancy rate we have observed in patients with LH deficiency attested by low E2:oocyte ratio could be explained by this endometrial effect. Therefore, further studies are needed to determine better whether a specific sub-group of patients could benefit from rLH supplementation after GnRH antagonist initiation, although it is not evident to define prospectively the criteria to select these patients.

On the contrary, other authors have suggested that high serum LH levels at the early stage of stimulation might be responsible for advanced endometrial maturation and lower pregnancy rates (Kolibianakis et al., 2002Go, 2003aGo). This observation leads the authors to recommend an early initiation of the antagonist on day 1 of the stimulation (Kolibianakis et al., 2003bGo). However, many factors may influence the serum LH values during stimulation protocols. We observed that pre-treatment with oral contraceptive pill decreases basal LH values at the beginning of the stimulation. Therefore, while FSH administration has been demonstrated to lower endogenous LH levels (Hull et al., 1998Go), increasing E2 levels after a few days of stimulation may trigger a premature LH surge. At the time of GnRH antagonist administration, we observed lower serum LH levels in our study than those reported in the flexible protocol without pre-treatment. Therefore, oral contraceptive pill pre-treatment could be another way to decrease the area under the curve of LH secretion before GnRH antagonist introduction, but whether it may improve implantation remains to be investigated. Finally, the use of a high dose, long-acting GnRH antagonist could prevent an LH surge more efficiently and help to control further LH secretion. So, both the use of a hormonal pre-treatment and a long-acting antagonist could explain why we did not find any influence of the timing of cetrorelix introduction upon the pregnancy rate.

In conclusion, despite some weaknesses in the study design (two different randomization procedures, no blinding for rLH administration), our findings do not support a systematic supplementation with rLH at the time of GnRH antagonist injection in a flexible protocol. In patients pre-treated with an oral contraceptive pill, the duration of stimulation prior to GnRH antagonist introduction does not seem to be related to the outcome of the cycle.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Albano C, Smitz J, Camus M, Riethmuller-Winzen H, Van Steirteghem A and Devroey P (1997) Comparison of different doses of gonadotropin-releasing hormone antagonist Cetrorelix during controlled ovarian hyperstimulation. Fertil Steril 67, 917–922.[CrossRef][Medline]

Al-Inany H and Aboulghar M (2002) GnRH antagonist in assisted reproduction: a Cochrane Review. Hum Reprod 17, 874–885.[Abstract/Free Full Text]

Asimakopoulos B, Nikolettos N and Al-Hasani S (2002) Outcome of cryopreserved pronuclear oocytes obtained after stimulation with either HMG or recFSH and the GnRH antogonist cetrorelix. RBM Online 5 (Suppl 1), 52–56.[Medline]

Burguès S and the Spanish Collaborative Group on Female Hypogonadotropic Hypogonadism (2001) The effectiveness and safety of recombinant human LH to support follicular development induced by recombinant human FSH in WHO group I anovulation: evidence from a multicentre study in Spain. Hum Reprod 16, 2525–2532.[Abstract/Free Full Text]

Diedrich K, Diedrich C, Santos E, Zoll C, al-Hasani S, Reissmann T, Krebs D and Klingmuller D (1994) Suppression of the endogenous luteinizing hormone surge by the gonadotrophin-releasing hormone antagonist Cetrorelix during ovarian stimulation. Hum Reprod 9, 788–791.[Abstract]

European Recombinant Human LH Study Group (1998) Recombinant human luteinizing hormone (LH) to support recombinant human follicle-stimulating hormone (FSH)-induced follicular development in LH- and FSH-deficient anovulatory women: a dose-finding study. J Clin Endocrinol Metab 83, 1507–1514.[Abstract/Free Full Text]

Filicori M, Cognigni G, Samara A, Melappioni S, Perri T, Cantelli B, Parmigiani L, Pelusi G and DeAloysio D (2002) The use of LH activity to drive folliculogenesis: exploring uncharted territories in ovulation induction. Hum Reprod Update 8, 543–557.[Abstract/Free Full Text]

Fujimoto VY, Monroe SE, Nelson LR, Downey D and Jaffe RB (1997) Dose-related suppression of serum luteinizing hormone in women by a potent new gonadotrophin-releasing hormone antagonist (Ganirelix) administered by intranasal spray. Fertil Steril 67, 469–473.[Medline]

Ganirelix Dose-finding Study Group (1998) A double-blind, randomized, dose-finding study to assess the efficacy of the gonadotrophin-releasing hormone antagonist ganirelix (Org 37462) to prevent premature luteinizing hormone surges in women undergoing ovarian stimulation with recombinant follicle stimulating hormone (Puregon). Hum Reprod 13, 3023–3031.[Abstract]

Hillier SG (2001) Gonadotropic control of ovarian follicular growth and development. Mol Cell Endocrinol 179, 39–46.[CrossRef][Medline]

Hull Ml, Livesey JH, Evans JJ and Benny PS (1998) The effects of recombinant follicle stimulating hormone (Gonal-F) on endogenous luteinizing hormone secretion in women. Hum Reprod 13, 1139–1143.[Abstract]

Jaakkola T, Ding Y, Valavaara R, Martikainen H, Tapanainen J, Ronnberg L and Huthtaniemi I (1990) The ratios of serum bioactive/immunoreactive luteinizing hormone and follicle-stimulating hormone in various clinical conditions with increased and decreased gonadotrophin secretion: re-evaluation by a highly sensitive immunoassay. J Clin Endocrinol Metab 70, 1496–1505.[Abstract]

Kolibianakis EM, Bourgain C, Albano C, Osmanagaoglu K, Smitz J, Van Steirteghem AC and Devroey P (2002) Effect of ovarian stimulation with recombinant follicle-stimulating hormone, gonadotropin-releasing hormone antagonists, and human chorionic gonadotropin on endometrial maturation on the day of oocyte pick-up. Fertil Steril 78, 1025–1029.[CrossRef][Medline]

Kolibianakis EM, Albano C, Kahn J, Camus M, Tournaye H, Van Steirteghem AC and Devroey P (2003a) Exposure to high levels of luteinizing hormone and estradiol in the early follicular phase of gonadotropin-releasing hormone antagonist cycles is associated with a reduced chance of pregnancy. Fertil Steril 79, 873–880.[CrossRef][Medline]

Kolibianakis EM, Albano C, Kahn J, Camus M, Tournaye H, Van Steirteghem AC and Devroey P (2003b) Initiation of gonadotropin-releasing hormone antagonist on day 1 as compared to day 6 of stimulation: effect on hormonal levels and follicular development in in vitro fertilization cycles. J Clin Endocrinol Metab 88, 5632–5637.[Abstract/Free Full Text]

Loumaye E, Engrand P, Howles C and O'Dea L (1997) Assessment of the role of serum luteinizing hormone and estradiol response to follicle-stimulating hormone on in vitro fertilization treatment outcome. Fertil Steril 67, 889–899.[CrossRef][Medline]

O'Dea L and the U.S. Recombinant Human LH Study Group (2000) Recombinant human LH in support of recombinant FSH in female hypogonadotropic hypogonadism—evidence of threshold effect. Fertil Steril 74 (Suppl 1), S36.

Olivennes F, Belaisch-Allart J, Emperaire JC, Alvarez S, Bouchard P and Frydman R (1999) The use of HMG versus rec-FSH with the single dose GnRH antagonist (cetrorelix) protocol in IVF-ET: a prospective randomized study. Fertil Steril 72, (Suppl 1; 55th annual meeting of the American Society for Reproduction Medicine), S114.

Olivennes F, Cunha-Filho JS, Fanchin R, Bouchard P and Frydman R (2002) The use of GnRH antagonists in ovarian stimulation. Hum Reprod Update 8, 279–290.[Abstract/Free Full Text]

Peñarrubia J, Fabregues F, Creus M, Manau D, Casamitjana R, Guimera M, Carmona F, Vanrell JA and Balasch J (2003) LH serum levels during ovarian stimulation as predictors of ovarian response and assisted reproduction outcome in down regulated women stimulated with recombinant FSH. Hum Reprod 18, 2689–2697.[Abstract/Free Full Text]

Porchet HC, Munafo A, Beltrami V and Le Cotonnec JV (1994) Clinical pharmacology of rhLH (LHadi). Hum Reprod 9 (Suppl 1.A), 23.

Reissman T, Schally AV, Bouchard P, Reithmüller H and Engel J (2000) The LHRH antagonist Cetrorelix: a review. Hum Reprod Update 6, 322–331.[Abstract/Free Full Text]

Shoham Z (2002) The clinical therapeutic window for luteinizing hormone in controlled ovarian stimulation. Fertil Steril 77, 1170–1177.[CrossRef][Medline]

Short RV (1962) Steroids in the follicular fluid and the corpus luteum of the mare: a two-cell type theory of ovarian steroids synthesis. J Endocrinol 2, 59–63.

Sullivan MW, Stewart-Akers A, Krasnow JS, Berga SL and Zeleznik AJ (1999) Ovarian responses in women to recombinant follicle-stimulating hormone and luteinizing hormone (LH): a role for LH in the final stages of follicular maturation. J Clin Endrocrinol Metab 84, 228–232.[Abstract/Free Full Text]

Tesarik J, Hazout A and Mendoza C (2003) Luteinizing hormone affects uterine receptivity independently of ovarian function. RBM Online 7, 59–64.[Medline]

Thompson KA, Henderson G, LaPolt PS, Dahl KD, Rivier J and Meldrum D (1995) Gonadotropin requirements of the developing follicle. Fertil Steril 63, 273–276.[Medline]

Submitted on February 26, 2004; accepted on May 19, 2004.