for administered luteinizing hormone-to-follicle-stimulating hormone ...

Download PDF
9 downloads 74 Views 406KB Size Report
Comment
Aug 20, 2014 - administering a GnRH agonist or a GnRH antagonist during stimulation, premature LH surges should be uncommon and pointed to exogenous ...
Defining the ‘‘sweet spot’’ for administered luteinizing hormone-to-follicle-stimulating hormone gonadotropin ratios during ovarian stimulation to protect against a clinically significant late follicular increase in progesterone: an analysis of 10,280 first in vitro fertilization cycles Marie D. Werner, M.D.,a Eric J. Forman, M.D.,a,b Kathleen H. Hong, M.D.,a Jason M. Franasiak, M.D.,a Thomas A. Molinaro, M.D., M.S.C.E.,a,b and Richard T. Scott Jr., M.D., H.C.L.D.a,b a Division of Reproductive Endocrinology, Department of Obstetrics, Gynecology and Reproductive Sciences, Robert Wood Johnson Medical School, Rutgers University, New Brunswick; and b Reproductive Medicine Associates of New Jersey, Basking Ridge, New Jersey

Objective: To determine whether different ratios of administered LH-to-FSH influence the risk of clinically relevant late follicular P elevations and whether there is an optimal range of LH-to-FSH to mitigate this risk. Design: Retrospective cohort. Setting: Private academic center. Patient(s): A total of 10,280 patients undergoing their first IVF cycle. Intervention(s): None. Main Outcome Measure(s): The ratio of exogenous LH-to-FSH throughout stimulation and association with absolute serum P level R1.5 ng/mL on the day of hCG administration. Result(s): Stimulations using no administered LH (N ¼ 718) had the highest risk of P elevation R1.5 ng/mL (relative risk [RR] ¼ 2.0; 95% confidence interval [CI] 1.8–2.2). The lowest risk of P increase occurred with an LH-to-FSH ratio of 0.30:0.60 (20%; N ¼ 4,732). In contrast, ratios 0.60 (23%, RR 1.1; 95% CI 1.0–1.3). This pattern of lowest risk in the 0.30–0.60 range held true for cycles characterized by low, normal, and high response. When performing a logistic regression to control for multiple confounding variables this relationship persisted.

Received April 3, 2014; revised July 3, 2014; accepted July 7, 2014; published online August 20, 2014. M.D.W. has nothing to disclose. E.J.F. has nothing to disclose. K.H.H. has nothing to disclose. J.M.F. has nothing to disclose. T.A.M. has nothing to disclose. R.T.S. has nothing to disclose. Reprint requests: Marie D. Werner, M.D., Reproductive Medicine Associates of New Jersey, 140 Allen Road, Basking Ridge, New Jersey 07920 (E-mail: [email protected]). Fertility and Sterility® Vol. 102, No. 5, November 2014 0015-0282/$36.00 Copyright ©2014 The Authors. Published by Elsevier Inc. on behalf of the American Society for Reproductive Medicine. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/). http://dx.doi.org/10.1016/j.fertnstert.2014.07.766 1312

VOL. 102 NO. 5 / NOVEMBER 2014

Fertility and Sterility® Conclusion(s): Absent or inadequate LH dosing is associated with a risk for a late follicular elevation in P sufficient to induce suboptimal outcomes. A total LH-to-FSH ratio of 0.30:0.60 was associated with the lowest risk of P elevation. Optimization of this parameter should be considered when making gonadotropin dosing decisions. (Fertil SterilÒ 2014;102:1312–7. Ó2014 by American Society for Use your smartphone Reproductive Medicine.) to scan this QR code Key Words: Gonadotropins, late follicular increase in progesterone, exogenous LH, exogenous and connect to the FSH, stimulation Discuss: You can discuss this article with its authors and with other ASRM members at http:// fertstertforum.com/wernerm-lh-fsh-ratios-elevated-progesterone/

T

he hormonal milieu, which accompanies a supraphysiologic response to controlled ovarian hyperstimulation (COH), has been associated with impaired endometrial receptivity. Much of this diminution has been attributed to significant increases in circulating E2 concentrations; however, other changes that accompany superovulation may also impact endometrial receptivity. One such factor may be subtle increases in P levels during the late follicular phase (1, 2). These P elevations are important as they prognosticate suboptimal clinical outcomes (3–5). Early literature describing these elevations assumed that they were part of the spectrum of early and excessive LH effect on the maturing follicles. As such, they were termed premature luteinization (6, 7). There are two potential sources of LH, either exogenous from injectable gonadotropins or endogenous from the pituitary. Given the near universal practice of administering a GnRH agonist or a GnRH antagonist during stimulation, premature LH surges should be uncommon and pointed to exogenous LH as a possible causative agent. More recently, studies have compared the prevalence of premature P elevations in patients receiving pure FSH stimulations to those receiving hMG alone (8, 9). Given that the hMG group received pharmacologic levels of LH stimulation, it might seem intuitive that they would have had a higher prevalence of premature P elevations. In fact, those women receiving hMG had a lower risk. This suggests that a relationship between LH and premature P elevations is complex and may not be wholly attributed to excessive stimulation. These data suggest that optimizing the effect of LH during COH may be dependent on both the level of exogenous LH and FSH that are administered (10–12). The impact of different administered LH-to-FSH ratios during stimulation have not been studied in detail. To that end, this study seeks to determine whether different ratios of LH:FSH activity in stimulation protocols impact the risk for premature P elevation and whether those differences also apply to different ovarian response groups.

MATERIALS AND METHODS Population In this retrospective cohort study, all patients attempting conception through IVF from October 1999 to May 2013 were reviewed. Patients undergoing their first IVF cycle in this program and whose superovulation protocol used either GnRH agonist down-regulation or a GnRH antagonist were selected VOL. 102 NO. 5 / NOVEMBER 2014

discussion forum for this article now.*

* Download a free QR code scanner by searching for “QR scanner” in your smartphone’s app store or app marketplace.

for further study. Patients using microdose GnRH agonist flare protocols were excluded, as there was no mechanism to quantify the contribution of endogenous LH release on the overall level of LH stimulation. Patient characteristics and demographic information were recorded. Response to stimulation was measured by the number of mature metaphase II oocytes obtained after vaginal oocyte retrieval. This retrospective analysis of data was Institutional Review Board approved by Western Institutional Review Board, protocol 20021333.

Study Design The purpose of this study was to determine whether variations in the relative amounts of exogenous LH and FSH impact the risk for significant P elevations before the administration of hCG to induce final oocyte maturation. The ratio of exogenous LH to FSH was calculated based on the total dose of each medication administered throughout the cycle. Starting dosages and protocol were selected by the primary physician in relation to patient characteristics, such as age, ovarian reserve, and prior history, but were also guided by insurance restrictions. Overall medication dosages maintained a relatively constant ratio throughout the stimulation as per practice standards. Although doses infrequently changed throughout the cycle, this metric was believed to be the most reflective of total LH exposure. Serum levels of LH and FSH were not routinely measured during cycles. The quantity of FSH was expressed in international units and was based on total FSH dose without regard to whether it was from a pure FSH preparation (recombinant or purified), an hMG preparation, or a combination of the two. The quantity of LH was also expressed in international units when using hMG or recombinant LH. One ampule of hMG was considered to have 75 IU of LH activity. In the case of low dose hCG administration, 10 IU was designated to be equivalent to 75 IU of LH. Starting and total doses of exogenous LH and FSH were recorded for each included cycle. The LH-toFSH ratio was calculated by simply dividing the total LH dose by the total FSH dose administered. Serum P levels were measured throughout the cycle, and the P level on the day of hCG administration was also documented to assess for clinically significant late follicular P elevations.

Assay Serum P was determined using the Immulite 2000 immunoassay system (Siemens). The interassay coefficient of variation 1313

ORIGINAL ARTICLE: ASSISTED REPRODUCTION (CV) was 5.58% and intra-assay CV was 5.25% for this system. For the purposes of this study, P levels R1.5 ng/mL on day of hCG administration were characterized as a late follicular increase in P, based on a review of internal data that showed the same diminution in outcome as published literature (1, 13) (data not shown). Cycles were stratified based on the total ratio of exogenous LH to exogenous FSH used. (For example a patint who received a protocol with a starting dose of 150 IU of recombinant FSH and 2 ampules of hMG and maintained this dose for 10 days of stimulation would have received a total exogenous LH dose of 1,500 IU and a total exogenous FSH dose of 3,000 IU and would be categorized as having an LH-to-FSH ratio of 0.5.) A total of 18 groups of exogenous LH-to-FSH exposure were defined to compare meaningful data points in a large population. These 18 groups spanned a ratio from no exogenous FSH (0) to ratios R0.81, with an incremental increase of 0.05 between each group. The data were then stratified relative to ovarian response and the same groups were identified in relation to the number of mature oocytes obtained. This was in an effort to control for the intrinsic differences in response groups, as each group was exposed to varying levels of endogenous LH, which may impact overall outcomes. Low ovarian response was defined as a cycle in which %4 metaphase II oocytes were retrieved. Similarly, a normal response was defined by having 5–19 metaphase II oocytes retrieved and high response by R20 mature oocytes.

Statistical Analyses Statistical analysis was performed using Analyse-it for Excel version 2.26 and STATA version 12. A contingency table was applied for categorical variables and a receiver operator characteristic curve was used to determine the optimal ratio of LHto-FSH administered. Statistical significance was set at P< .05. This analysis was performed for the population as a whole, and then repeated for the group analysis. Logistic regression was used for the entire population to model the relationship between elevated serum P (R1.5 ng/dL) and the starting LH-to-FSH ratio as a continuous variable and as a dichotomous variable using cutoffs of 0.3 and 0.6, as

well as comparing those subjects with the range of 0.3–0.6 to all others. Confounding variables including the number of follicles, age, stimulation protocol, serum E2 at the time of trigger, and diagnosis were controlled for using multivariate logistic regression. This study does not include a direct comparison of implantation and delivery rates in the various LH-to-FSH ratio groups. This reflects the fact that clinical management was not similar in the various groups. During the study interval, transfer timing was influenced by the presence or absence of P levels >1.5 ng/mL on the day of hCG administration. When elevations were detected, embryos were typically cryopreserved and transferred in a subsequent cycle. Thus patients in groups with higher or lower risks for P elevations would have very different transfer strategies preventing meaningful comparison of cycle outcomes such as pregnancy rates (PRs).

RESULTS Population Characteristics A total of 10,280 cycles were included for analysis. There were 5,393 cycles using a GnRH agonist down-regulation protocol and 4,887 using an antagonist protocol. The average age of patients included was 34.7  4.3 years. The average maximum FSH value on day 3 was 6.4  2.4 IU/L. The average body mass index (BMI) was 25.4  5.9 kg/m2. Additional demographic information is provided in Table 1. In the group analysis there were a total of 1,803 low response cycles, with an average age of 36.5  4.2 years, FSH 7.0  2.7 IU/L, and BMI 25.1  5.8 kg/m2. The normal response group included 7,218 cycles with an average age of 34.6  4.2 years, FSH 6.4  2.3 IU/L, BMI 25.5  5.9 kg/ m2. In the high response group, there were a total of 1,259 cycles with an average age of 33.1  4.0 years, FSH 5.6  1.9 IU/L, BMI 25.3  5.8 kg/m2.

Evaluation of the Ratio of Total Exogenous LHto-FSH Dosing in Stimulation Cycles were stratified based on the ratio of LH-to-FSH into the 18 designated small groups. A receiver operator characteristic curve was then created and two critical breakpoints were

TABLE 1 Demographic information. Primary diagnosis category DOR Endometriosis Male Other Ovulatory dysfunction Tubal Unknown Uterine

No. of patients (%)

Age (y) (mean ± SD)

Day 3 FSH (IU/L) (mean ± SD)

BMI (kg/m2) (mean ± SD)

No. of antral follicles (mean ± SD)

Estradiol on day of surge (pg/mL)

No. of M2s (mean ± SD)

436 (4) 666 (6) 3,422 (33) 894 (9) 2,327 (23) 1,118 (11) 1,200 (12) 217 (2)

38.5  3.8 33.7  3.9 33.9  4.2 35.0  4.3 33.4  4.3 34.6  4.0 34.7  4.2 36.2  4.1

7.6  3.0 6.5  2.3 6.5  2.3 6.5  2.4 5.8  2.3 6.7  2.3 6.5  2.2 6.6  2.2

24.3  4.6 24.1  4.6 25.3  5.6 25.0  5.3 26.7  7.2 25.9  5.7 24.0  4.7 25.8  6.0

8.3  4.4 11.5  7.1 13.6  7.4 12.5  7.5 17.6  10.9 12.0  7.1 12.7  8.0 11.6  6.6

1,365.9  718.9 1,843.0  1,004.1 2,046.9  1,121.7 1,954.8  1,096.5 2,254.5  1,217.5 1,995.7  1,117.0 2,042.6  1,047.6 1,964.9  1,020.5

6.2  4.4 9.0  6.7 11.2  6.8 10.8  7.2 12.7  8.1 10.5  7.5 9.9  6.7 9.8  6.9

Note: BMI ¼ body mass index; DOR ¼ diminished ovarian reserve; M2 ¼ metaphase II oocyte. Werner. Gonadotropin ratios alter risk of P increase. Fertil Steril 2014.

1314

VOL. 102 NO. 5 / NOVEMBER 2014

Fertility and Sterility® identified at which the LH-to-FSH ratio was associated with the greatest risk in a premature increase in P, specifically 0.60. Once these breakpoints were determined, groups of exposure were compared to ascertain the optimal range of exogenous LH to exogenous FSH associated with the lowest risk of a premature increase in P. The lowest risk of a premature increase in P was noted in the seven groups spanning the LH-to-FSH ratio of 0.30–0.60 (20%; N ¼ 4,732; P< .001).

Population Outcomes Controlled ovarian stimulations using a ratio of 0.30 to 0.60 had the lowest risk of a premature increase in P with only 20% of cycles exhibiting this pattern. This group was then compared with cycles using no LH, a lower proportion of exogenous LH, or a higher proportion of exogenous LH. Patients using no LH in their stimulation had a 40% chance of exhibiting a premature increase in P and this was statistically higher than the aforementioned group (P< .001, relative risk [RR] ¼ 2.0; 95% confidence interval [CI] 1.8–2.2). Similarly, 32% cycles with an LH-to-FSH ratio of 0.60 exhibited a premature increase in P, which was significantly higher than the 0.30-to-0.60 ratio (P¼ .03; RR ¼ 1.1, 95% CI 1.0–1.3) (Fig. 1).

Response Group Outcomes When the analysis was performed for the 18 groups, differences between each group were also evident. Cycles using absolutely no exogenous LH had the highest risk of a premature increase in P, as mentioned previously, with 40% having a P level R1.5 ng/mL at the end of stimulation, significantly higher than all other subgroups (P< .001; RR ¼ 2.0, 95% CI 1.8–2.2). However, extremes of stimulation, both high and low, were associated with significant risk

FIGURE 1

The incidence of late follicular increase in P is significantly lower in cycles with an administered LH-to-FSH ratio between 0.30 and 0.60. P14 mm, stimulation protocol, serum E2 on the day of trigger, and diagnoses of polycystic ovary syndrome (PCOS) and diminished ovarian reserve, this relationship persisted (OR ¼ 0.42, 95% CI 0.38–0.47; P< .001).

DISCUSSION Increasing evidence during the past several years has confirmed that late follicular elevations in P during IVF stimulation predict suboptimal clinical outcomes after fresh ET (8, 14, 15). For example, it has been recently demonstrated that an increase in P R1.5 ng/mL before hCG administration was the critical threshold at which clinical outcomes were diminished in one large assisted reproductive technology (ART) program (1). The results of the present study provide insight into an iatrogenic cause of a late follicular increase in P, and possibly a way to protect against this adverse effect. The most likely explanation for the adverse impact of this effect relates to advancement of endometrial receptivity resulting from a premature secretory transformation due to supraphysiologic serum P levels (16). The question remains whether clinical management decisions impact these late P increases and whether there are specific interventions that would reduce risk. One option would be to trigger final oocyte maturation early during stimulation (17); however, it is difficult to predict when P will cross a critical threshold that impairs receptivity and this may also result in a suboptimal yield of mature oocytes at retrieval. Although routine cryopreservation of the cohort of embryos has been proposed, this introduces an intervention and delay in pregnancy for a majority of patients who would otherwise have favorable 1315

ORIGINAL ARTICLE: ASSISTED REPRODUCTION

FIGURE 2

The optimal ratio of exogenous LH-to-FSH to prevent a premature increase in P according to response group. Werner. Gonadotropin ratios alter risk of P increase. Fertil Steril 2014.

outcomes with a fresh ET. If there were a way to reduce the risk of late P elevations, this could provide an opportunity for a higher proportion of patients to undergo fresh ET without being subjected to the adverse impact of premature P increases. The purpose of the current study was to determine whether the risk of significant premature P increase is related to one of the most fundamental clinical decisions made by reproductive endocrinologists during IVF stimulation, namely the composition of the gonadotropins administered. When analyzing an extremely large dataset of more than 10,000 IVF cycles, it was determined that in fact the dosing decisions made by clinicians do impact the risk of premature P increase and these data point to a method to reduce this risk: including an adequate proportion of LH activity during stimulation. Because pituitary LH is the signal that initiates follicular luteinization with the accompanying massive production of P, it seemed intuitive that excessive exogenous LH in stimulation may have a similar effect, thus driving up serum P levels. This prompted the development of pure FSH preparations and a shift in clinical practice to prescribe FSH-only protocols. The results of the Menotropin versus Recombinant FSH in vitro Fertilization Group (MERIT) trial, which compared serum and follicular P levels, at first appeared counterintuitive (9). Patients randomized to receive pure FSH actually had higher P levels than those receiving hMG-only protocols. In light of the two-cell, two-gonadotropin theory, as described previously, this should not have been so surprising (18). Excessive FSH stimulation may increase production of P and other precursors from the granulosa cells (GC). Therefore, including some LH in stimulation protocols to counterbalance the effect of FSH may help reduce the risk of late follicular P increases. These analyses reveal that variations in the relative proportion of LH and FSH administered have a substantial impact on the outcomes of ovarian stimulation. A ratio of LH-to-FSH 1316

that falls between 0.30 and 0.60 during the cycle provides the lowest risk of a premature increase in P and represents a target ‘‘sweet spot’’ for clinicians. This relationship holds true for low responders, normal responders, and high responders. However, it is notable that high responders had the greatest risk of a premature increase in P when compared with all other groups. It is important to note that in this analysis, only the absolute level of P was considered, not the relative proportion of E2 to P or P per mature follicle. A likely explanation for this increased risk in high responders relates to a cumulative effect of many follicles producing small amounts of P before hCG administration. Significantly, this population still had the lowest risk of P elevation when their LH-to-FSH ratio decreased within the 0.30–0.60 range. Interestingly, extremes of stimulation that deviated the furthest from the optimal ratio, or sweet spot, of 0.30–0.60, were at the greatest risk of a premature increase in P. The relative risk of premature P increase was more pronounced in the lower ranges of the LH-to-FSH ratio spectrum than the >.60 range. These data suggest that the clinician can influence stimulation outcomes and protect against a premature increase in P by providing an appropriate amount of exogenous LH in stimulation to ensure thecal conversion of P to androgen precursors. Stimulations using no exogenous LH had the highest risk of a premature increase in P and this finding confirms results of previous literature (8, 19). With dose modifications during stimulation, absolute quantities of LH and FSH will vary throughout stimulation. It is important for clinicians to remain cognizant of this need to balance the ratio of LH-toFSH to prevent a late follicular increase in P. Although there are many subtle differences in stimulation style and gonadotropin preferences that can yield excellent ART outcomes, the results of this large dataset may help refine these clinical decisions. As doses are reduced during stepdown protocols, reducing the FSH component would help to keep the total administered LH-to-FSH ratio toward the VOL. 102 NO. 5 / NOVEMBER 2014

Fertility and Sterility® middle-to-upper range of the sweet spot. For example, a starting dose that provides an equivalent dose of pure FSH and hMG (or 10 IU of low dose human chorionic gonadotropin (LDHCG) for every 150 IU of pure FSH) would yield a ratio of 0.50 if the dose was maintained throughout stimulation. A slight reduction in the FSH component toward the end of stimulation, or an equal reduction of FSH and hMG together, would keep the total administered gonadotropin ratio within the desired range. In this analysis both LDHCG and hMG were included as sources of exogenous LH exposure. Very few cycles incorporated recombinant LH. These preparations have different pharmacodynamics and a study comparing individual outcomes would be beneficial. The high risk population of patients with diminished ovarian reserve, who have an increased risk of late follicular P increases due to ‘‘premature luteinization,’’ often receive microdose GnRH agonist flare protocols in this center and, therefore, were not included. Consequently these results may not be generalizable to this population. A limitation of this study relates to its retrospective design. Although differences in risk were found consistently in the different LH-to-FSH ratios and these were consistent across age groups, this type of study cannot definitely prove that changing management in another population prospectively would improve endocrine dynamics during stimulation. The extremely large sample size, however, provides excellent precision regarding the risk of P increase in the different groups of LH-to-FSH ratio and ovarian response. Another confounding variable, which may limit generalized application, is that this analysis only reports total LH-to-FSH ratios during the entirety of a cycle rather than starting or changing doses. This ratio was chosen as the most accurate representation of total gonadotropin dosage due to practice standards where gonadotropin dosages remain relatively constant throughout cycles. In addition, this large dataset could correct for potential sources of bias such as preferences of individual physicians within a large group practice. This study provides insight into the importance of exogenous LH in stimulation and the necessity to include LH activity to balance the deleterious effects of excessive exogenous FSH. The exclusion of exogenous LH in stimulation is iatrogenic, not physiologic, and may impair IVF outcomes by significantly altering the endocrine milieu and placing patients at an increased risk of premature secretory transformation and decreased endometrial receptivity at the time of ET. There appears to be a role for exogenous LH for all types of ovarian responders. Relative doses outside the sweet spot ratio of 0.30-to-0.60 are associated with the highest risk of a premature increase in P that may translate into poorer overall outcomes and diminished clinical PRs.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

Acknowledgments: The authors thank Batsal Devkota for his assistance in data collection of the Reproductive Medicine Associates of New Jersey Bioinformatics team. 18.

REFERENCES 1.

n C, Remohí J, Jenkins J, et al. Circulating Bosch E, Labarta E, Crespo J, Simo progesterone levels and ongoing pregnancy rates in controlled ovarian stim-

VOL. 102 NO. 5 / NOVEMBER 2014

19.

ulation cycles for in vitro fertilization: analysis of over 4000 cycles. Hum Reprod 2010;25:2092–100. Fanchin R, Righini C, Olivennes F, Ferreira AL, de Ziegler D, Frydman R. Consequences of premature progesterone elevation on the outcome of in vitro fertilization: insights into a controversy. Fertil Steril 1997;68:799–805. Shulman A, Ghetler Y, Beyth Y, Ben-Nun I. The significance of an early (premature) rise of plasma progesterone in in vitro fertilization cycles induced by a ‘‘long protocol’’ of gonadotropin releasing hormone analogue and human menopausal gonadotropins. J Assist Reprod Genet 1996;13:207–11. Silverberg KM, Burns WN, Olive DL, Riehl RM, Schenken RS. Serum progesterone levels predict success of in vitro fertilization/embryo transfer in patients stimulated with leuprolide acetate and human menopausal gonadotropins. J Clin Endocrinol Metab 1991;73:797–803. Silverberg KM, Martin M, Olive DL, Burns WN, Schenken RS. Elevated serum progesterone levels on the day of human chorionic gonadotropin administration in in vitro fertilization cycles do not adversely affect embryo quality. Fertil Steril 1994;61:508–13. Hofmann GE, Khoury J, Johnson CA, Thie J, Scott RT Jr. Premature luteinization during controlled ovarian hyperstimulation for in vitro fertilizationembryo transfer has no impact on pregnancy outcome. Fertil Steril 1996; 66:980–6. Ubaldi F, Camus M, Smitz J, Bennink HC, van Steirteghem A, Devroey P. Premature luteinization in in vitro fertilization cycles using gonadotropinreleasing hormone agonist (GnRH-a) and recombinant follicle-stimulating hormone (FSH) and GnRH-a and urinary FSH. Fertil Steril 1996;66:275–80. Andersen AN, Devroey P, Arce JC. Clinical outcome following stimulation with highly purified hMG or recombinant FSH in patients undergoing IVF: a randomized assessor-blind controlled trial. Hum Reprod 2006;21:3217–27. Smitz J, Andersen AN, Devroey P, Arce JC, MERIT Group. Endocrine profile in serum and follicular fluid differs after ovarian stimulation with HP-hMG or recombinant FSH in IVF patients. Hum Reprod 2007;22:676–87. Shoham Z, Howles CM, Zalel Y, Weissman A, Insler V. Induction of follicular growth and production of a normal hormonal milieu in spite of using a constant low dose of luteinizing hormone in women with hypogonadotrophic hypogonadism. Hum Reprod 1994;9:431–6. Martins WP, Vieira AD, Figueiredo JB, Nastri CO. FSH replaced by low-dose hCG in the late follicular phase versus continued FSH for assisted reproductive techniques. Cochrane Database Syst Rev 2013:CD010042. Sullivan MW, Stewart-Akers A, Krasnow JS, Berga SL, Zeleznik AJ. 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 Endocrinol Metab 1999;84:228–32. Hamdine O, Macklon NS, Eijkemans MJC, Laven JSE, Cohlen BJ, Verhoeff A, et al. Elevated early follicular progesterone levels and in vitro fertilization outcomes: a prospective intervention study and meta-analysis. Fertil Steril 2014; 102:448–54.e1. Fanchin R, de Ziegler D, Taieb J, Hazout A, Frydman R. Premature elevation of plasma progesterone alters pregnancy rates of in vitro fertilization and embryo transfer. Fertil Steril 1993;59:1090–4. Fanchin R, Hourvitz A, Olivennes F, Taieb J, Hazout A, Frydman R. Premature progesterone elevation spares blastulation but not pregnancy rates in in vitro fertilization with coculture. Fertil Steril 1997;68:648–52. Harada T, Yoshida S, Katagiri C, Takao N, Ikenari T, Toda T, et al. Reduced implantation rate associated with a subtle rise in serum progesterone concentration during the follicular phase of cycles stimulated with a combination of a gonadotrophin-releasing hormone agonist and gonadotrophin. Hum Reprod 1995;10:1060–4. Kyrou D, Kolibianakis EM, Fatemi HM, Tarlatzis BC, Tournaye H, Devroey P. Is earlier administration of human chorionic gonadotropin (hCG) associated with the probability of pregnancy in cycles stimulated with recombinant follicle-stimulating hormone and gonadotropin-releasing hormone (GnRH) antagonists? A prospective randomized trial. Fertil Steril 2011;96:1112–5. Fleming R. Progesterone elevation on the day of hCG: methodological issues. Hum Reprod Update 2008;14:391–2. Bosch E, Vidal C, Labarta E, Simon C, Remohi J, Pellicer A. Highly purified hMG versus recombinant FSH in ovarian hyperstimulation with GnRH antagonists—a randomized study. Hum Reprod 2008;23:2346–51.

1317