- Email: [email protected]
Pretreatment with estrogen does not affect IVF-ICSI cycle outcome compared with no pretreatment in GnRH antagonist protocol: a prospective randomized trial
Pretreatment with estrogen does not affect IVF-ICSI cycle outcome compared with no pretreatment in GnRH antagonist protocol: a prospective randomized trial drin-Durnerin, M.D.,a Anne Guivarc'h-Leve ^ que, M.D.,b and Jean-Noe €l Hugues, M.D., Ph.D.,a on behalf of the Isabelle Ce decine et Endocrinologie de la Reproduction Groupe d'Etude en Me a decine de la Reproduction, Ho ^ pital Jean Verdier, Assistance Publique—Ho ^ pitaux de Paris, Universite Paris XIII, Bondy, France; Service de Me and b Clinique Mutualiste la Sagesse, Rennes, France
Objective: To assess effects of estrogen pretreatment in GnRH antagonist protocol. Design: Prospective, randomized multicenter study. Setting: Ten private or university-based centers. Patient(s): A total of 472 patients undergoing IVF/ICSI. Intervention(s): Randomization by sealed envelopes to receive 17b-estradiol (4 mg/d) or no pretreatment before daily recombinant FSH administration started on the first day of estrogen discontinuation or on cycle day 2 in nonpretreated women. Main Outcome Measure(s): The primary outcome measure was the number of retrieved oocytes. Secondary efficacy variables included total FSH dose, cycle duration, and outcome. Result(s): The mean numbers of retrieved oocytes (10.9 5.7 vs. 10.2 5.6) and obtained embryos (5.5 3.7 vs. 4.8 3.7) were not significantly different between women allocated to estrogen pretreatment (n ¼ 238) and no pretreatment (n ¼ 234). Total FSH amount (1,557 408 vs. 1,389 347 IU) and stimulation duration (10.8 1.4 vs. 10.0 1.5 days) were slightly but significantly increased in pretreated patients. Positive pregnancy tests, ultrasound pregnancy rate, and delivery rate per cycle were similar (36%, 33%, and 26.6%, respectively, vs. 38.2%, 35.4%, and 30%). Conclusion(s): These data confirm that estrogen pretreatment is associated with requirement of higher FSH doses and longer duration of stimulation without any significant increase in the number of retrieved oocytes. However, estrogen does not affect cycle outcome and therefore might be used in clinical practice for programming IVF retrievals during working days. Clinical Trials Registration Number: NCT01489852. (Fertil SterilÒ 2012;97:1359–64. Ó2012 by American Society for Reproductive Medicine.) Key Words: ART, GnRH antagonist, estrogen pretreatment, ovarian stimulation
G
nRH antagonist protocol offers several important advantages compared with long agonist protocol (1), namely: a simple way to prevent premature LH surge without any risk of inadvertent administration at the beginning of pregnancy; no
flare-up effect on gonadotropin secretion; no risk of ovarian cyst formation; no burden of desensitization symptoms, a reduced duration of treatment; and a lower consumption of gonadotropins. Furthermore, a recent meta-analysis (2) concluded that their efficacy is similar
Received December 13, 2011; revised February 17, 2012; accepted February 21, 2012; published online March 28, 2012. I.C.-D. has nothing to disclose. A.G.-L. has nothing to disclose. J.-N.H. has nothing to disclose. Schering Plough–Organon Pharmaceutical Co. (now MSD) provided logistic support (room rental, transportation) for study investigator meetings. decine et Endocrinologie de la Reproduction: Isabelle Other members of the Groupe d'Etude en Me –Me decine de la Reproduction, Bruges), Claudine Parneix (Clinique Jean Villar, Aquitaine Sante ^ le Sante Le onard de Vinci, Chambray les Tours), Bettina Bsta €ndig (Centre Matisse, Vasseur (Po ^ pital Paul de Viguier, Toulouse), Ge rard Marcollin (Clinique du Nice), Florence Lesourd (Ho s Colombel (Centre Hospitalier Universitaire, Nantes), Sophie Dubourdieu Bois, Lille), Agne re se Le Martellot (Ho ^ pital Augustin Morvan, Brest). (Nantes), and Marie-The drin-Durnerin, M.D., Service de Me decine de la Reproduction, Ho ^ pital Jean Reprint requests: Isabelle Ce Verdier, Av. du 14 Juillet, Bondy 93143, France (E-mail: [email protected]). Fertility and Sterility® Vol. 97, No. 6, June 2012 0015-0282/$36.00 Copyright ©2012 American Society for Reproductive Medicine, Published by Elsevier Inc. doi:10.1016/j.fertnstert.2012.02.028 VOL. 97 NO. 6 / JUNE 2012
to that of long agonist protocol in terms of live birth rate, but associated with a highly significant reduction in the incidence of severe ovarian hyperstimulation. For these reasons, it seems justified to move away from the standard long agonist protocol to a GnRH antagonist protocol as first-line choice for in vitro fertilization (IVF)/intracytoplasmic sperm injection (ICSI) procedure in good-prognosis patients. However, many teams remain reluctant to use GnRH antagonist protocol or reserve it for poor-prognosis patients (3). One of the main drawbacks that could discourage a widespread use of this protocol is the absence of cycle programming. In clinical practice, programming is very convenient both for physicians by facilitating work 1359
ORIGINAL ARTICLE: ASSISTED REPRODUCTION study conducted in ten IVF centers. The inclusion criteria were: regular normo-ovulatory cycles (28–35 days), age <38 years, body mass index (BMI) 18–30 kg/m2, and first or second IVF/ICSI attempt. Exclusions were based on the following criteria: high basal levels of serum FSH (>12 IU/L) or E2 (>80 pg/mL), fewer than five follicles at the antral follicular count (AFC) performed on day 3 of a spontaneous cycle, or history of high (>20 oocytes) or low (<5 oocytes) ovarian response in an earlier IVF attempt. Selected women were randomly assigned to receive an estrogen pretreatment or no pretreatment. Random allocation sequence was generated from a table of random numbers and concealed from each physician who enrolled and randomized patients. Randomization was stratified by center and performed by means of sealed envelopes. This study was not blind. The study was approved by the local Ethical Committee, and each of the patients provided informed written consent.
schedule whether from week to week or within weeks on working days, and for patients by planning appointments for ultrasound examination and blood sampling. Another potential advantage of programming antagonist protocol is the possibility to synchronize follicle growth. Indeed, FSH administration is commonly started on day 2 or 3 of a spontaneous menstrual cycle at a time the follicular cohort has already begun to grow under the influence of the FSH intercycle rise starting 5 days before the onset of menses. This actually contrasts with long GnRH agonist protocol, where ovarian resting and homogenization of the follicular cohort are obtained after pituitary desensitization. Consequently, the slight reduction in the oocyte yield reported with antagonist protocol compared with long agonist protocol (4, 5) could be explained by the heterogeneity of the follicular cohort before stimulation. For these reasons, more attention has been paid to the potential interest of steroid pretreatments for scheduling GnRH antagonist cycles. Three main options have been used to achieve cycle programming. First the oral contraceptive pill (OCP) seems to be effective in avoiding ovarian retrievals during weekends (6) but is not associated with increased oocyte yield (7). Furthermore, a recent meta-analysis pointed out that OCP pretreatment may be associated with fewer clinical pregnancies (8). Second, although they have not been widely used for programming antagonist cycles, synthetic progestogens are good candidates because of their potent suppressive effect on pituitary gonadotropin secretion, as demonstrated with short GnRH agonist protocols (9). Finally, luteal administration of natural estrogens was reported to improve the synchronization of early antral follicle growth and to allow retrieval of two additional oocytes (10). In a previous study focused on hormonal and ultrasound events during the wash-out period after pretreatment administration (11), we observed that the hormonal environment actually differs according to the estrogenic or progestogen composition of the product used. OCP and synthetic progestogen sharply decreased both serum FSH and LH values with recovery of normal basal levels after a 5-day wash-out period. In contrast, natural estrogen decreased only FSH levels to a lower extent than OCP and progestogen, and the recovery of basal levels was quicker within 3 days. Whatever the pretreatment used, endogenous FSH suppression at the end of the recovery period was associated with a more homogeneous cohort of follicles before stimulation. However, the overall impact of steroid pretreatments on the subsequent cycle outcome remains to be determined in a larger study. Because OCP seems to be deleterious to cycle outcome (8) and the previously used progestogen product was withdrawn from the market, we chose to focus our clinical research on the consequences of estrogen pretreatment. For that purpose, we conducted a multicenter study in France to assess the effectiveness of estrogen pretreatment on ovarian stimulation and cycle outcome compared with no pretreatment.
Pretreatment with 17b-estradiol (Provames; Aventis) was administered during the cycle preceding the IVF/ICSI cycle. A daily dose of 4 mg (2 mg twice a day) was given orally, started 7 days before the presumed onset of menses and administered up to the next Thursday after the occurrence of menstrual bleeding. The control group did not receive any pretreatment. Hormonal and ultrasound assessments were performed before the start of ovarian stimulation on Friday in the estrogen-pretreated group and on cycle day 2 after spontaneous menses in the control group. Ovarian stimulation cycle was canceled in case of ovarian cysts >20 mm in diameter or with serum progesterone (P) levels >3 ng/mL. The starting dose of FSH (Puregon; Organon) was 150 IU per day with a 50-IU increment in women aged >35 years or with BMI >25 kg/m2. This dose was kept constant for 5 days and then adjusted according to monitoring of ovarian response. A daily administration of ganirelix (0.25 mg Orgalutran; Organon) was introduced in a fixed way on day 6 and was repeated up to the time of hCG administration. Ovulation was triggered by recombinant hCG (250 mg Ovitrelle; Merck Serono) when at least three mature (R17 mm) follicles were obtained, and oocyte pick-up was performed 36 hours later. ICSI or IVF procedures were applied as required. Embryo transfers were performed on day 2 or 3 according to center practice. Good-quality embryos were defined by the presence of 3–5 blastomeres on day 2, or 6–10 blastomeres on day 3 with amount of fragmentation <20%. Luteal phase was supported by vaginal administration of micronized P (400 mg/ d Utrogestan; Besins International) and micronized 17bestradiol (2 mg/d Provames; Aventis) from the day of ovarian puncture to the day of pregnancy test. If a pregnancy occurred, P administration was extended until evidence of fetal heart activity at ultrasound examination.
MATERIALS AND METHODS
Hormonal and Ultrasound Monitoring
Subjects
Both serum samples and vaginal ultrasound examinations were performed on days 1 (S1), 6 (S6), and 8 (S8) of ovarian stimulation and on the day of or the day preceding hCG
From December 2006 to July 2010, 472 women undergoing IVF/ICSI cycles were enrolled in this prospective randomized 1360
Protocols
VOL. 97 NO. 6 / JUNE 2012
Fertility and Sterility® administration (DhCG or DhCG1, respectively). Serum E2, LH, and P concentrations were measured at each visit, whereas serum FSH concentrations were determined only at S1. Hormonal measurements were carried out in each center with the use of commercially available chemoluminescence immunoassays with automated Elecsys immunoanalyzer (ECLIA; Roche Diagnostic; or Advia Centaur; Bayer Diagnostics). The sensitivity of the assay was 5 pg/mL for E2, 0.03 ng/ mL for P, 0.3 IU/L for FSH, and 0.07 IU/L for LH. Intra- and interassay coefficients of variation were, respectively, 5% and 10% for E2, 3% and 5% for P, 1.4% and 2.4% for FSH, and 2.3% and 2.6% for LH. Ultrasound assessments were performed with a 6-Mhz vaginal transducer. The number of follicles sized 2–5 mm, 6–9 mm, and >10 mm in both ovaries, as well as the endometrium thickness, were recorded. Follicle diameter was calculated as the mean diameter measured in two dimensions.
Sample Size Estimate The primary end point was the number of retrieved oocytes. Based on our previous study, the mean SD of the number of retrieved oocytes is 6.5 and a difference of two oocytes was expected between pretreated patients and patients without pretreatment. The calculated sample size with a ¼ b ¼ .05 was 225 subjects per group. Ten more patients per group were planned to be included to take cancellations into account. The major secondary end points were the consumption of gonadotropins, the duration of treatment, and implantation and pregnancy rates.
Statistical Analysis Statistical analysis was performed using Statview (Abacus Concepts). Nominal or continuous variables were analyzed with c2, analysis of variance (ANOVA), or ANOVA for repeated measures as required. Analyses were adjusted for center if necessary. A P value of < .05 was considered to be statistically significant.
RESULTS Baseline Characteristics A total of 472 patients were randomized, 238 allocated to estrogen pretreatment and 234 to no pretreatment. The flow chart of the patients during the study is given in Supplemental Figure 1 (available online at www.fertstert.org). Patients from the two groups were similar for age, BMI, indications of assisted reproduction procedure, rank of attempt, cycle length, basal hormonal levels of FSH, LH, or E2, and AFC (Table 1).
Hormonal and Ultrasound Assessments Before Ovarian Stimulation Sixteen randomized patients did not start IVF procedure: discontinuation for reasons unrelated to protocol (3 patients in each group) and spontaneous pregnancies (1 in estrogenpretreated patients and 9 in patients without pretreatment). Therefore, 233 pretreated patients and 220 nonpretreated VOL. 97 NO. 6 / JUNE 2012
TABLE 1 Patient characteristics.
Age (y) BMI (kg/m2) Primary infertility Duration (y) Indications (%) Male Tubal Endometriosis Unexplained Attempt rank Cycle length (d) Day 3 FSH (IU/L) LH (IU/L) E2 (pg/mL) AFC
Estrogen (n [ 238)
No pretreatment (n [ 234)
P value
31.1 3.6 22.3 3.2 74% 3.3 1.7
31.2 3.7 22.3 3.0 75% 3.4 2.0
NS NS NS NS
67 16 9 23 1.2 0.5 29.3 2.6
57 19 8 15 1.2 0.5 29.3 2.2
NS NS NS NS NS NS
6.8 5.1 40 16.3
6.6 4.9 40 16.1
NS NS NS NS
Note: Results are expressed as mean SD. AFC ¼ antral follicle count. C edrin-Durnerin. Estradiol treatment and antagonist protocol. Fertil Steril 2012.
patients were evaluated before starting ovarian stimulation. After a mean duration of 11 days of estrogen pretreatment, serum FSH and LH levels were significantly lower compared with the control group (Table 2) and, as expected, serum E2 levels were significantly higher because of the short free interval after discontinuation of estrogen pretreatment. Serum P levels were not significantly different between groups. At ultrasound examination, AFC was similar between the two groups, but endometrium was slightly thicker in estrogenpretreated patients.
Ovarian Stimulation Parameters Three patients did not start ovarian stimulation, one in the pretreated group (no period) and two in nonpretreated patients (1 cyst, 1 protocol change). In relation to lower endogenous gonadotrophin levels at the start of ovarian stimulation, the serum E2 level and the number of recruited and large follicles were significantly lower on day 6 of ovarian stimulation in patients pretreated with estrogen (Table 3). At this time, no difference was observed between groups in LH
TABLE 2 Hormonal measurements and ultrasound assessment before the start of ovarian stimulation.
Treatment duration (d) FSH (IU/L) LH (IU/L) E2 (pg/mL) P (ng/mL) AFC Endometrium (mm)
Estrogen (n [ 233)
No pretreatment (n [ 220)
P value
10.7 3.1 4.5 2.2 4.8 3.1 187 179 0.6 0.6 16.3 5.9 5.1 2
– 7 2.1 5.7 2.4 42 20 0.7 0.5 16.1 6.1 4.3 2
.0001 .018 < .0001 NS NS .015
Note: Results are expressed as mean SD. AFC ¼ antral follicle count. C edrin-Durnerin. Estradiol treatment and antagonist protocol. Fertil Steril 2012.
1361
ORIGINAL ARTICLE: ASSISTED REPRODUCTION
TABLE 3 Ovarian stimulation parameters. Parameter
Estrogen (n [ 233)
No pretreatment (n [ 220)
P value
158 23
156 24
NS
530 330 3.6 4 6.3 0.6 0.3 4.8 3.3 21 (9) 7.3 2 16 (6.9) 4.5 1.4 1,557 408 162 31 10.8 1.4
673 358 3.8 4.2 6.7 0.7 0.4 5.1 3.4 68 (30.9) 7.6 1.6 16 (7.3) 3.9 1.2 1,389 347 158 30 10.0 1.5
.002 NS NS NS .005 < .0001 NS NS < .0001 < .0001 NS < .0001
1,890 906 2 1.6 0.9 0.5 6.0 3.5 3.5 1.9 4.7 2.8 10 2
1,544 740 1.3 1.2 0.9 0.4 5.7 4.1 3.2 2.1 4.2 2.8 10 2
< .0001 < .0001 NS NS NS NS NS
FSH starting dose (IU/d) At stimulation day 6 E2 (pg/mL) LH (IU/L) Patient with LH >10 IU/L (%) P (ng/mL) Foll. 11–14 mm (n) Patient with foll. R15 mm, n (%) Endometrium (mm) Cancellations, n (%) Antagonist treatment (d) Total FSH dose (IU) Daily FSH dose (IU) hCG administration (d) At day of hCG administration E2 (pg/mL) LH (IU/L) P (ng/mL) Foll. 11–14 mm (n) Foll. 15–16 mm (n) Foll. >17 mm (n) Endometrium (mm) Note: Results are expressed as mean SD. Foll. ¼ follicles.
Cedrin-Durnerin. Estradiol treatment and antagonist protocol. Fertil Steril 2012.
levels or the percentage of patients with LH >10 IU/L. The cancellation rate was similar (poor response: 10 and 9; high response: 2 and 3; other reasons: 4 and 4; in pretreated and nonpretreated patients, respectively). Despite similar starting doses, the total amount of FSH used during stimulation was significantly higher in estrogen-pretreated patients than in patients without pretreatment. Furthermore, the duration of ovarian stimulation was significantly longer, by almost 1 day, in estrogen-pretreated patients. As a consequence of both higher FSH amount and longer duration of stimulation, serum E2 and LH levels were significantly higher on the day of hCG administration in estrogen-pretreated patients.
Biological Parameters and Cycle Outcome No significant difference was observed in the mean number of retrieved or mature oocytes, in the fertilization rate, in the number of obtained embryos, and the percentage of goodquality embryos (Table 4). The mean number and quality of transfered embryos were similar between the two groups. No significant difference was observed in pregnancy rate regarding positive pregnancy tests or ultrasound pregnancies. The implantation rates also were not significantly different. Additionally, no difference was observed in miscarriage, multiple pregnancy, and delivery rates. In each group, one supplemental delivery (not included in Table 4) was obtained resulting from spontaneous pregnancy in one patient canceled for spontaneous ovulation and in one patient with fertilization failure.
DISCUSSION Our results did not confirm the potential beneficial effects of estrogen pretreatment in GnRH antagonist cycles on the 1362
number of retrieved oocytes but showed that this pretreatment, which is convenient for scheduling IVF, does not negatively affect cycle outcome. These data challenge the concept of the synchronization of the follicular cohort by the use of a steroid pretreatment. Similar lack of difference was seen with OCP pretreatment. Nevertheless, we can extrapolate from our previous study (11) that follicle size within the recruitable cohort should have been smaller in estrogen-pretreated patients than in patients without pretreatment. Indeed, in contrast to the design of our previous study, ovarian stimulation was started on the first day of estrogen discontinuation after a short free interval similar to that used in the study by Fanchin et al. (10). At that time, the FSH rebound has not yet occurred, as shown in our previous study (11) and therefore FSH levels are significantly lower than in nonpretreated patients. This observation leads us to suggest that the duration of the wash-out period after E2 pretreatment should be extended by 1 day to allow recovery of baseline FSH levels before starting ovarian stimulation. Indeed, the duration of the wash-out period is an important determining factor of FSH consumption, as outlined for OCP pretreatment, where a 5-day wash-out seems to be a better compromise than a 3-day wash-out between homogenization of the cohort and FSH consumption. Despite the use of similar starting doses of exogenous FSH, we observed on day 6 of ovarian stimulation a significantly lower number of medium-size follicles and a lower proportion of cycles with follicles >15 mm in patients pretreated by estrogen, suggesting a delay at the initiation of follicular growth. Furthermore, the duration of ovarian stimulation was nearly one day longer and total FSH requirement was significantly increased by 150 IU in estrogen-pretreated patients. VOL. 97 NO. 6 / JUNE 2012
Fertility and Sterility®
TABLE 4 Biologic parameters and cycle outcome. Parameter Oocytes retrieval (n) ICSI/IVF (n) Oocytes M2 oocytes Fertilization rate (%) Embryos Good-quality embryos (%) Transfers (n) D2/D3/D5 Replaced embryos Good-quality embryos (%) Positive b-hCG (n) US pregnancy (n) IR (yolk sacs/embryos) IR (fetal heart/embryos) Miscarriages or losses (n) Singleton (n) Twin (n) Total delivery (n) Delivery rate/cycle Delivery rate/retrieval Delivery rate/transfer
Estrogen (n [ 233)
No pretreatment (n [ 220)
P value
217 144/73 10.9 5.7 8.0 4.6 71 5.5 3.7 59 197 111/86/0 1.7 0.6 84.9 84 77 89/337 (26.4%) 84/337 (24.9%) 15 52 10 62 26.6% 28.6% 31.5%
204 147/57 10.2 5.6 7.6 4.7 68.1 4.8 3.2 65 192 99/92/1 1.7 0.6 88.8 84 78 95/323 (29.4%) 88/323 (27.2%) 12 52 14 66 30% 32.3% 34.3%
NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS
Note: Results are expressed as mean SD. D2/D3/D5 ¼ days 2, 3, and 5; ICSI ¼ intracytoplasmic sperm injection; IR ¼ implantation rate; IVF ¼ in vitro fertilization; M2 ¼ metaphase II. Cedrin-Durnerin. Estradiol treatment and antagonist protocol. Fertil Steril 2012.
The increased homogeneity of the follicular cohort induced by steroid pretreatment is supposed to improve the synchrony of follicular growth during stimulation and is expected to result in an increase in both oocyte yield and gamete quality. But we did not observe any significant difference in the number of medium, intermediate, and large follicles on the day of hCG administration, nor in the number of retrieved or mature oocytes. Our data contrast with those reported by Fanchin et al., who observed on the day of hCG an increased number of large-size follicles and mature oocytes, but difference in FSH requirement between groups was not assessed. In studies that showed a higher oocyte yield after OCP pretreatment, a high concomitant increase in total amount of gonadotropin was reported (12, 13), which biased the analysis of the specific effect of pretreatment. Finally, compelling evidence for an effect of pretreatment on the follicular cohort only came from the analysis of the kinetics of follicular growth during stimulation. The initiation of follicular growth was delayed, but a higher number of large-size follicles on the day of hCG administration was reported for both OCP (13) and E2 (10) pretreatments compared with no pretreatment. This pattern of follicular growth closely mimics the kinetics observed following down regulation in long GnRH agonist protocols (13–15). However, this change in follicular growth does not seem to result in better oocyte quality after OCP pretreatment, as reflected by the number of good-quality embryos obtained (13). In contrast, promising results were obtained after estrogen pretreatment with an increased number of good-quality embryos (10). Nevertheless, this was not confirmed by the present study performed in an adequately sized population. Finally, it seems that both OCP and E2 pretreatments, when administered to a selected population of young normal responders, are unable to VOL. 97 NO. 6 / JUNE 2012
increase quantity and quality of yielded oocytes, although the follicular cohort has been homogenized. This is not surprising, because follicular cohorts of normal size are rarely disorganized, in contrast to what can be observed in patients with decreased ovarian reserve. In the latter population, it is likely that the benefit of steroid pretreatment in terms of oocyte yield could be more easily demonstrated. Another important issue is the potential impact of steroid pretreatments on cycle outcome. Some concerns regarding the negative effect of OCP on the likelihood of pregnancy have been raised by randomized studies (13, 16, 17). Although those conclusions were not confirmed in retrospective analyses of large number of patients (18, 19), this major drawback has been reemphasized by successive metaanalyses of randomized studies (7, 8). Two hypotheses have been advanced to explain the underlying mechanism for this deleterious effect: a decrease in LH exposure or an effect on the endometrium. First, a decrease in serum LH levels induced by OCP pretreatment has been reported to impair ovarian response and implantation in a subset of patients (20). However, the impact of low LH levels on success rate in GnRH antagonist protocols is still a quite controversial issue, with some studies showing an improved outcome (21) and others an impaired outcome (22) or no consequence (23). Similarly, a slight but significant decrease in serum LH levels was observed in the present study after discontinuation of estrogen pretreatment. It was shown that these residual LH levels increased with the length of exposure to estrogen pretreatment after the occurrence of menses but were not significantly correlated with cycle outcome (24). On the other hand, an increased incidence of premature LH rise before the introduction of GnRH antagonist could impair outcome of cycles programmed by 1363
ORIGINAL ARTICLE: ASSISTED REPRODUCTION estrogen pre-treatment (15). Nevertheless, our results do not corroborate any significant difference in the incidence of premature LH surge in estrogen-pretreated patients compared with patients without pretreatment. Second, OCP was reported to negatively affect implantation by lowering endometrial thickness (16), or by altering E2 and P endometrial receptors (17). Estrogen pretreatment could offer an interesting alternative, because the endometrium is thicker at the initiation of ovarian stimulation. However, the trend to higher pregnancy rate previously reported with E2 pretreatment (10) has not been confirmed by our results. Therefore, in the absence of any deleterious effect on cycle outcome, estrogen pretreatment should be used in clinical practice for programming antagonist cycles to favor oocyte retrievals during working days. In conclusion, our data confirm that estrogen pretreatment is associated with requirement of higher FSH doses and longer duration of ovarian stimulation, whereas we did not observe any significant increase in the number of retrieved oocytes and good-quality embryos. In addition, estrogen pretreatment did not affect cycle outcome. The impact of increased FSH requirement and slightly longer duration of treatment in terms of cost and discomfort for patients is balanced by the improvement in scheduling allowed by estrogen pretreatment.
9.
10.
11.
12.
13.
14.
15.
16.
Acknowledgments: The authors thank Dr. C. Perves for her help in organizing investigator meetings.
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
Devroey P, Aboulghar M, Garcia-Velasco J, Griesinger G, Humaidan P, Kolibianakis E, et al. Improving the patient's experience of IVF/ICSI: a proposal for an ovarian stimulation protocol with GnRH antagonist co-treatment. Hum Reprod 2009;24:764–74. Al-Inany HG, Youssef MAFM, Aboulghar M, Broekmans F, Sterrenburg M, Smit J, et al. Gonadotrophin-releasing hormone antagonists for assisted reproductive technology. Cochrane Database Syst Rev 2011;(5):CD001750. Griesinger G, Felderbaum R, Diedrich K. GnRH antagonists in ovarian stimulation: a treatment regimen of clinicians'second choice? Data from the german national IVF registry. Hum Reprod 2005;20:2373–5. Al-Inany HG, Abbou Setta AM, Aboulghar M. Gonadotrophin-releasing hormone antagonists for assisted conception: a Cochrane review. Reprod Biomed Online 2007;14:640–9. Kolibianakis EM, Collins J, Tarlatzis BC, Devroey P, Diedrich K, Griesinger G. Among patients treated for IVF with gonadotrophins and GnRH analogues, is the probability of live birth dependent on the type of analogue used? A systematic review and meta-analysis. Hum Reprod Update 2007;12: 651–71. Barmat LI, Chantilis SJ, Hurst BS, Dickey RP. A randomized prospective trial comparing gonadotropin-releasing hormone (GnRH) antagonist/recombinant follicle-stimulating hormone (rFSH) versus GnRH-agonist/rFSH in women pretreated with oral contraceptives before in vitro fertilization. Fertil Steril 2005;83:321–30. Griesinger G, Venetis CA, Marx T, Diedrich K, Tarlatzis BC, Kolibianakis EM. Oral contraceptive pill pre-treatment in ovarian stimulation with GnRH antagonists for IVF: a systematic review and meta-analysis. Fertil Steril 2008; 90:1055–63. Smulders B, van Oirschot SM, Farquhar C, Rombauts L, Kremer JAM. Oral contraceptive pill, progestogen or estrogen pre-treatment for ovarian stimulation protocols for women undergoing assisted reproductive techniques [review]. Cochrane Database Syst Rev 2010;(1):CD006109.
1364
17.
18.
19.
20.
21.
22.
23.
24.
Cedrin-Durnerin I, Bulwa S, Herve F, Martin-Pont B, Uzan M, Hugues JN. The hormonal flare-up following GnRH agonist administration is influenced by a progestogen pre-treatment. Hum Reprod 1996;11:1859–63. Fanchin R, Cunha-Filho JS, Schonaeur LM, Kadoch IJ, Cohen Bacri P, Frydman R. Coordination of early antral follicles by luteal estradiol administration provides a basis for alternative controlled ovarian hyperstimulation regimens. Fertil Steril 2003;79:316–21. Cedrin-Durnerin I, Bst€andig B, Parneix I, Bied-Damon V, Avril C, Decanter C, et al. Effects of oral contraceptive, synthetic progestogen or natural estrogen pre-treatments on the hormonal profile and the antral follicle cohort before GnRH antagonist protocol. Hum Reprod 2007;22:109–16. Huirne JA, Hugues JN, Pirard C, Fischl F, Sage JC, Pouly JL, et al. Cetrorelix in an oral contraceptive-pretreated stimulation cycle compared with buserelin in IVF/ICSI patients treated with r-hFSH: a randomized, multicentre, phase IIIb study. Hum Reprod 2006;21:1408–15. Rombauts L, Healy D, Norman RJ, Orgalutran Scheduling Study Group. A comparative randomized trial to assess the impact of oral contraceptive pre-treatment on follicular growth and hormone profiles in GnRH antagonist-treated patients. Hum Reprod 2006;21:95–103. Cunha Filho JS, Terres LF, Holanda F, Freitas F, Glitz C, Genro VK, et al. Luteal phase administration in ovarian stimulation cycles with antagonist is comparable to the GnRH agonist (long) protocol. J Assist Reprod Genet 2007;24: 326–30. Ye H, Huang GN, Zeng PH, Pei L. IVF/ICSI outcomes between cycles with luteal estradiol (E2) pre-treatment before GnRH antagonist protocol and standard long agonist protocol: a prospective and randomized study. J Assist Reprod Genet 2009;26:105–11. Kolibianakis EM, Papanikolaou EG, Camus M, Tournaye H, van Steirteghem AC, Devroey P. Effect of oral contraceptive pill pre-treatment on ongoing pregnancy rates in patients stimulated with GnRH antagonists and recombinant FSH for IVF. A randomized controlled trial. Hum Reprod 2006;21:352–7. Nyboe Andersen A, Witjes H, Gordon K, Mannaerts B, Xpect Investigators. Predictive factors of ovarian response and clinical outcome after IVF/ICSI following a rFSH/GnRH antagonist protocol with or without oral contraceptive pre-treatment. Hum Reprod 2011;12:3413–23. Pinkas H, Sapir O, Avrech OM, Ben-Haroush A, Ashkenzi J, Fisch B, et al. The effect of oral contraceptive pill for cycle scheduling prior to GnRH-antagonist protocol on IVF cycle parameters and pregnancy outcome. J Assist Reprod Genet 2008;25:29–33. Bellver J, Albert C, Labarta E, Pellicer A. Early pregnancy loss in women stimulated with gonadotropin-releasing hormone antagonist protocols according to oral contraceptive pill pre-treatment. Fertil Steril 2007;87: 1098–101. Meldrum DR, Scott RT, Levy MJ, Alper MM, Noyes N. Oral contraceptive pretreatment in women undergoing controlled ovarian stimulation in ganirelix acetate cycles may, for a subset of patients be associated with low serum luteinizing hormone levels, reduced ovarian response to gonadotropins, and early pregnancy loss. Fertil Steril 2009;91:1963–5. Kolibianakis EM, Zikopoulous K, Schiettecatte J, Smitz J, Tournaye H, Camus M, et al. Profound LH suppression after GnRH antagonist administration is associated with a significantly higher ongoing pregnancy rate in IVF. Hum Reprod 2004;19:2490–6. Huirne JAF, van Loenen ACD, Schats R, McDonnell J, Hompes PGA, Schoemaker J, et al. Dose-finding study of daily GnRH antagonist for the prevention of premature LH surges in IVF/ICSI patients: optimal change in LH and progesterone for clinical pregnancy. Hum Reprod 2005;20:359–67. Doody KJ, Devroey P, Leader A, Witges H, Mannaerts BM. No association between endogenous LH and pregnancy in a GnRH antagonist protocol: part1, corifollitropin alpha. RBM Online 2011;23:449–56. Guivarc'h-Lev^eque A, Homer L, Arvis P, Broux PL, Moy L, Priou G, et al. Programming in vitro fertilization retrievals during working days after gonadotropin releasing hormone antagonist protocol with estrogen pretreatment: does the length of exposure to estradiol impact on controlled ovarian hyperstimulation outcomes? Fertil Steril 2011;96:872–6.
VOL. 97 NO. 6 / JUNE 2012
Fertility and Sterility®
SUPPLEMENTAL FIGURE 1 Enrolment Assessed for eligibility (n=not available) Randomized (n=472) Allocation Allocated to estrogen (n=238) Received estrogen (n=234) Did not receive estrogen (n=4) Spontaneous pregnancy (n=1) Discontinued (n=3)
Allocated to no pre-treatment (n=234) Received no pre-treatment (n=222) Did not receive no prettt (n=12) Spontaneous pregnancy (n=9) Discontinued (n=3)
Stimulation Start stimulation (n=233) No stimulation (n=1) No periods (n=1)
Start stimulation (n=220) No stimulation (n=2) Ovarian cyst (n=1) Protocole exchange (n=1)
Oocyte retrieval Retrieval (n=217) No retrieval (n=16) Poor response (n=10) High response (n=2) Other reason (n=4) 1 preg.
Retrieval (n=204) No retrieval (n=16) Poor response (n=9) High response (n=3) Other reason (n=4)
Transfer Transfer (n=197) No transfer (n=20) No mature oocyte (n=4) Fertilization failure (n=9) OHSS risk (n=2) Other (n=5)
Transfer (n=192) No transfer (n=12) No mature oocyte (n=2) Fertilization failure (n=5) 1 preg. OHSS risk (n=1) Other (n=4)
Study flow chart. Cedrin-Durnerin. Estradiol treatment and antagonist protocol. Fertil Steril 2012.
VOL. 97 NO. 6 / JUNE 2012
1364.e1
Objective: To assess effects of estrogen pretreatment in GnRH antagonist protocol. Design: Prospective, randomized multicenter study. Setting: Ten private or university-based centers. Patient(s): A total of 472 patients undergoing IVF/ICSI. Intervention(s): Randomization by sealed envelopes to receive 17b-estradiol (4 mg/d) or no pretreatment before daily recombinant FSH administration started on the first day of estrogen discontinuation or on cycle day 2 in nonpretreated women. Main Outcome Measure(s): The primary outcome measure was the number of retrieved oocytes. Secondary efficacy variables included total FSH dose, cycle duration, and outcome. Result(s): The mean numbers of retrieved oocytes (10.9 5.7 vs. 10.2 5.6) and obtained embryos (5.5 3.7 vs. 4.8 3.7) were not significantly different between women allocated to estrogen pretreatment (n ¼ 238) and no pretreatment (n ¼ 234). Total FSH amount (1,557 408 vs. 1,389 347 IU) and stimulation duration (10.8 1.4 vs. 10.0 1.5 days) were slightly but significantly increased in pretreated patients. Positive pregnancy tests, ultrasound pregnancy rate, and delivery rate per cycle were similar (36%, 33%, and 26.6%, respectively, vs. 38.2%, 35.4%, and 30%). Conclusion(s): These data confirm that estrogen pretreatment is associated with requirement of higher FSH doses and longer duration of stimulation without any significant increase in the number of retrieved oocytes. However, estrogen does not affect cycle outcome and therefore might be used in clinical practice for programming IVF retrievals during working days. Clinical Trials Registration Number: NCT01489852. (Fertil SterilÒ 2012;97:1359–64. Ó2012 by American Society for Reproductive Medicine.) Key Words: ART, GnRH antagonist, estrogen pretreatment, ovarian stimulation
G
nRH antagonist protocol offers several important advantages compared with long agonist protocol (1), namely: a simple way to prevent premature LH surge without any risk of inadvertent administration at the beginning of pregnancy; no
flare-up effect on gonadotropin secretion; no risk of ovarian cyst formation; no burden of desensitization symptoms, a reduced duration of treatment; and a lower consumption of gonadotropins. Furthermore, a recent meta-analysis (2) concluded that their efficacy is similar
Received December 13, 2011; revised February 17, 2012; accepted February 21, 2012; published online March 28, 2012. I.C.-D. has nothing to disclose. A.G.-L. has nothing to disclose. J.-N.H. has nothing to disclose. Schering Plough–Organon Pharmaceutical Co. (now MSD) provided logistic support (room rental, transportation) for study investigator meetings. decine et Endocrinologie de la Reproduction: Isabelle Other members of the Groupe d'Etude en Me –Me decine de la Reproduction, Bruges), Claudine Parneix (Clinique Jean Villar, Aquitaine Sante ^ le Sante Le onard de Vinci, Chambray les Tours), Bettina Bsta €ndig (Centre Matisse, Vasseur (Po ^ pital Paul de Viguier, Toulouse), Ge rard Marcollin (Clinique du Nice), Florence Lesourd (Ho s Colombel (Centre Hospitalier Universitaire, Nantes), Sophie Dubourdieu Bois, Lille), Agne re se Le Martellot (Ho ^ pital Augustin Morvan, Brest). (Nantes), and Marie-The drin-Durnerin, M.D., Service de Me decine de la Reproduction, Ho ^ pital Jean Reprint requests: Isabelle Ce Verdier, Av. du 14 Juillet, Bondy 93143, France (E-mail: [email protected]). Fertility and Sterility® Vol. 97, No. 6, June 2012 0015-0282/$36.00 Copyright ©2012 American Society for Reproductive Medicine, Published by Elsevier Inc. doi:10.1016/j.fertnstert.2012.02.028 VOL. 97 NO. 6 / JUNE 2012
to that of long agonist protocol in terms of live birth rate, but associated with a highly significant reduction in the incidence of severe ovarian hyperstimulation. For these reasons, it seems justified to move away from the standard long agonist protocol to a GnRH antagonist protocol as first-line choice for in vitro fertilization (IVF)/intracytoplasmic sperm injection (ICSI) procedure in good-prognosis patients. However, many teams remain reluctant to use GnRH antagonist protocol or reserve it for poor-prognosis patients (3). One of the main drawbacks that could discourage a widespread use of this protocol is the absence of cycle programming. In clinical practice, programming is very convenient both for physicians by facilitating work 1359
ORIGINAL ARTICLE: ASSISTED REPRODUCTION study conducted in ten IVF centers. The inclusion criteria were: regular normo-ovulatory cycles (28–35 days), age <38 years, body mass index (BMI) 18–30 kg/m2, and first or second IVF/ICSI attempt. Exclusions were based on the following criteria: high basal levels of serum FSH (>12 IU/L) or E2 (>80 pg/mL), fewer than five follicles at the antral follicular count (AFC) performed on day 3 of a spontaneous cycle, or history of high (>20 oocytes) or low (<5 oocytes) ovarian response in an earlier IVF attempt. Selected women were randomly assigned to receive an estrogen pretreatment or no pretreatment. Random allocation sequence was generated from a table of random numbers and concealed from each physician who enrolled and randomized patients. Randomization was stratified by center and performed by means of sealed envelopes. This study was not blind. The study was approved by the local Ethical Committee, and each of the patients provided informed written consent.
schedule whether from week to week or within weeks on working days, and for patients by planning appointments for ultrasound examination and blood sampling. Another potential advantage of programming antagonist protocol is the possibility to synchronize follicle growth. Indeed, FSH administration is commonly started on day 2 or 3 of a spontaneous menstrual cycle at a time the follicular cohort has already begun to grow under the influence of the FSH intercycle rise starting 5 days before the onset of menses. This actually contrasts with long GnRH agonist protocol, where ovarian resting and homogenization of the follicular cohort are obtained after pituitary desensitization. Consequently, the slight reduction in the oocyte yield reported with antagonist protocol compared with long agonist protocol (4, 5) could be explained by the heterogeneity of the follicular cohort before stimulation. For these reasons, more attention has been paid to the potential interest of steroid pretreatments for scheduling GnRH antagonist cycles. Three main options have been used to achieve cycle programming. First the oral contraceptive pill (OCP) seems to be effective in avoiding ovarian retrievals during weekends (6) but is not associated with increased oocyte yield (7). Furthermore, a recent meta-analysis pointed out that OCP pretreatment may be associated with fewer clinical pregnancies (8). Second, although they have not been widely used for programming antagonist cycles, synthetic progestogens are good candidates because of their potent suppressive effect on pituitary gonadotropin secretion, as demonstrated with short GnRH agonist protocols (9). Finally, luteal administration of natural estrogens was reported to improve the synchronization of early antral follicle growth and to allow retrieval of two additional oocytes (10). In a previous study focused on hormonal and ultrasound events during the wash-out period after pretreatment administration (11), we observed that the hormonal environment actually differs according to the estrogenic or progestogen composition of the product used. OCP and synthetic progestogen sharply decreased both serum FSH and LH values with recovery of normal basal levels after a 5-day wash-out period. In contrast, natural estrogen decreased only FSH levels to a lower extent than OCP and progestogen, and the recovery of basal levels was quicker within 3 days. Whatever the pretreatment used, endogenous FSH suppression at the end of the recovery period was associated with a more homogeneous cohort of follicles before stimulation. However, the overall impact of steroid pretreatments on the subsequent cycle outcome remains to be determined in a larger study. Because OCP seems to be deleterious to cycle outcome (8) and the previously used progestogen product was withdrawn from the market, we chose to focus our clinical research on the consequences of estrogen pretreatment. For that purpose, we conducted a multicenter study in France to assess the effectiveness of estrogen pretreatment on ovarian stimulation and cycle outcome compared with no pretreatment.
Pretreatment with 17b-estradiol (Provames; Aventis) was administered during the cycle preceding the IVF/ICSI cycle. A daily dose of 4 mg (2 mg twice a day) was given orally, started 7 days before the presumed onset of menses and administered up to the next Thursday after the occurrence of menstrual bleeding. The control group did not receive any pretreatment. Hormonal and ultrasound assessments were performed before the start of ovarian stimulation on Friday in the estrogen-pretreated group and on cycle day 2 after spontaneous menses in the control group. Ovarian stimulation cycle was canceled in case of ovarian cysts >20 mm in diameter or with serum progesterone (P) levels >3 ng/mL. The starting dose of FSH (Puregon; Organon) was 150 IU per day with a 50-IU increment in women aged >35 years or with BMI >25 kg/m2. This dose was kept constant for 5 days and then adjusted according to monitoring of ovarian response. A daily administration of ganirelix (0.25 mg Orgalutran; Organon) was introduced in a fixed way on day 6 and was repeated up to the time of hCG administration. Ovulation was triggered by recombinant hCG (250 mg Ovitrelle; Merck Serono) when at least three mature (R17 mm) follicles were obtained, and oocyte pick-up was performed 36 hours later. ICSI or IVF procedures were applied as required. Embryo transfers were performed on day 2 or 3 according to center practice. Good-quality embryos were defined by the presence of 3–5 blastomeres on day 2, or 6–10 blastomeres on day 3 with amount of fragmentation <20%. Luteal phase was supported by vaginal administration of micronized P (400 mg/ d Utrogestan; Besins International) and micronized 17bestradiol (2 mg/d Provames; Aventis) from the day of ovarian puncture to the day of pregnancy test. If a pregnancy occurred, P administration was extended until evidence of fetal heart activity at ultrasound examination.
MATERIALS AND METHODS
Hormonal and Ultrasound Monitoring
Subjects
Both serum samples and vaginal ultrasound examinations were performed on days 1 (S1), 6 (S6), and 8 (S8) of ovarian stimulation and on the day of or the day preceding hCG
From December 2006 to July 2010, 472 women undergoing IVF/ICSI cycles were enrolled in this prospective randomized 1360
Protocols
VOL. 97 NO. 6 / JUNE 2012
Fertility and Sterility® administration (DhCG or DhCG1, respectively). Serum E2, LH, and P concentrations were measured at each visit, whereas serum FSH concentrations were determined only at S1. Hormonal measurements were carried out in each center with the use of commercially available chemoluminescence immunoassays with automated Elecsys immunoanalyzer (ECLIA; Roche Diagnostic; or Advia Centaur; Bayer Diagnostics). The sensitivity of the assay was 5 pg/mL for E2, 0.03 ng/ mL for P, 0.3 IU/L for FSH, and 0.07 IU/L for LH. Intra- and interassay coefficients of variation were, respectively, 5% and 10% for E2, 3% and 5% for P, 1.4% and 2.4% for FSH, and 2.3% and 2.6% for LH. Ultrasound assessments were performed with a 6-Mhz vaginal transducer. The number of follicles sized 2–5 mm, 6–9 mm, and >10 mm in both ovaries, as well as the endometrium thickness, were recorded. Follicle diameter was calculated as the mean diameter measured in two dimensions.
Sample Size Estimate The primary end point was the number of retrieved oocytes. Based on our previous study, the mean SD of the number of retrieved oocytes is 6.5 and a difference of two oocytes was expected between pretreated patients and patients without pretreatment. The calculated sample size with a ¼ b ¼ .05 was 225 subjects per group. Ten more patients per group were planned to be included to take cancellations into account. The major secondary end points were the consumption of gonadotropins, the duration of treatment, and implantation and pregnancy rates.
Statistical Analysis Statistical analysis was performed using Statview (Abacus Concepts). Nominal or continuous variables were analyzed with c2, analysis of variance (ANOVA), or ANOVA for repeated measures as required. Analyses were adjusted for center if necessary. A P value of < .05 was considered to be statistically significant.
RESULTS Baseline Characteristics A total of 472 patients were randomized, 238 allocated to estrogen pretreatment and 234 to no pretreatment. The flow chart of the patients during the study is given in Supplemental Figure 1 (available online at www.fertstert.org). Patients from the two groups were similar for age, BMI, indications of assisted reproduction procedure, rank of attempt, cycle length, basal hormonal levels of FSH, LH, or E2, and AFC (Table 1).
Hormonal and Ultrasound Assessments Before Ovarian Stimulation Sixteen randomized patients did not start IVF procedure: discontinuation for reasons unrelated to protocol (3 patients in each group) and spontaneous pregnancies (1 in estrogenpretreated patients and 9 in patients without pretreatment). Therefore, 233 pretreated patients and 220 nonpretreated VOL. 97 NO. 6 / JUNE 2012
TABLE 1 Patient characteristics.
Age (y) BMI (kg/m2) Primary infertility Duration (y) Indications (%) Male Tubal Endometriosis Unexplained Attempt rank Cycle length (d) Day 3 FSH (IU/L) LH (IU/L) E2 (pg/mL) AFC
Estrogen (n [ 238)
No pretreatment (n [ 234)
P value
31.1 3.6 22.3 3.2 74% 3.3 1.7
31.2 3.7 22.3 3.0 75% 3.4 2.0
NS NS NS NS
67 16 9 23 1.2 0.5 29.3 2.6
57 19 8 15 1.2 0.5 29.3 2.2
NS NS NS NS NS NS
6.8 5.1 40 16.3
6.6 4.9 40 16.1
NS NS NS NS
Note: Results are expressed as mean SD. AFC ¼ antral follicle count. C edrin-Durnerin. Estradiol treatment and antagonist protocol. Fertil Steril 2012.
patients were evaluated before starting ovarian stimulation. After a mean duration of 11 days of estrogen pretreatment, serum FSH and LH levels were significantly lower compared with the control group (Table 2) and, as expected, serum E2 levels were significantly higher because of the short free interval after discontinuation of estrogen pretreatment. Serum P levels were not significantly different between groups. At ultrasound examination, AFC was similar between the two groups, but endometrium was slightly thicker in estrogenpretreated patients.
Ovarian Stimulation Parameters Three patients did not start ovarian stimulation, one in the pretreated group (no period) and two in nonpretreated patients (1 cyst, 1 protocol change). In relation to lower endogenous gonadotrophin levels at the start of ovarian stimulation, the serum E2 level and the number of recruited and large follicles were significantly lower on day 6 of ovarian stimulation in patients pretreated with estrogen (Table 3). At this time, no difference was observed between groups in LH
TABLE 2 Hormonal measurements and ultrasound assessment before the start of ovarian stimulation.
Treatment duration (d) FSH (IU/L) LH (IU/L) E2 (pg/mL) P (ng/mL) AFC Endometrium (mm)
Estrogen (n [ 233)
No pretreatment (n [ 220)
P value
10.7 3.1 4.5 2.2 4.8 3.1 187 179 0.6 0.6 16.3 5.9 5.1 2
– 7 2.1 5.7 2.4 42 20 0.7 0.5 16.1 6.1 4.3 2
.0001 .018 < .0001 NS NS .015
Note: Results are expressed as mean SD. AFC ¼ antral follicle count. C edrin-Durnerin. Estradiol treatment and antagonist protocol. Fertil Steril 2012.
1361
ORIGINAL ARTICLE: ASSISTED REPRODUCTION
TABLE 3 Ovarian stimulation parameters. Parameter
Estrogen (n [ 233)
No pretreatment (n [ 220)
P value
158 23
156 24
NS
530 330 3.6 4 6.3 0.6 0.3 4.8 3.3 21 (9) 7.3 2 16 (6.9) 4.5 1.4 1,557 408 162 31 10.8 1.4
673 358 3.8 4.2 6.7 0.7 0.4 5.1 3.4 68 (30.9) 7.6 1.6 16 (7.3) 3.9 1.2 1,389 347 158 30 10.0 1.5
.002 NS NS NS .005 < .0001 NS NS < .0001 < .0001 NS < .0001
1,890 906 2 1.6 0.9 0.5 6.0 3.5 3.5 1.9 4.7 2.8 10 2
1,544 740 1.3 1.2 0.9 0.4 5.7 4.1 3.2 2.1 4.2 2.8 10 2
< .0001 < .0001 NS NS NS NS NS
FSH starting dose (IU/d) At stimulation day 6 E2 (pg/mL) LH (IU/L) Patient with LH >10 IU/L (%) P (ng/mL) Foll. 11–14 mm (n) Patient with foll. R15 mm, n (%) Endometrium (mm) Cancellations, n (%) Antagonist treatment (d) Total FSH dose (IU) Daily FSH dose (IU) hCG administration (d) At day of hCG administration E2 (pg/mL) LH (IU/L) P (ng/mL) Foll. 11–14 mm (n) Foll. 15–16 mm (n) Foll. >17 mm (n) Endometrium (mm) Note: Results are expressed as mean SD. Foll. ¼ follicles.
Cedrin-Durnerin. Estradiol treatment and antagonist protocol. Fertil Steril 2012.
levels or the percentage of patients with LH >10 IU/L. The cancellation rate was similar (poor response: 10 and 9; high response: 2 and 3; other reasons: 4 and 4; in pretreated and nonpretreated patients, respectively). Despite similar starting doses, the total amount of FSH used during stimulation was significantly higher in estrogen-pretreated patients than in patients without pretreatment. Furthermore, the duration of ovarian stimulation was significantly longer, by almost 1 day, in estrogen-pretreated patients. As a consequence of both higher FSH amount and longer duration of stimulation, serum E2 and LH levels were significantly higher on the day of hCG administration in estrogen-pretreated patients.
Biological Parameters and Cycle Outcome No significant difference was observed in the mean number of retrieved or mature oocytes, in the fertilization rate, in the number of obtained embryos, and the percentage of goodquality embryos (Table 4). The mean number and quality of transfered embryos were similar between the two groups. No significant difference was observed in pregnancy rate regarding positive pregnancy tests or ultrasound pregnancies. The implantation rates also were not significantly different. Additionally, no difference was observed in miscarriage, multiple pregnancy, and delivery rates. In each group, one supplemental delivery (not included in Table 4) was obtained resulting from spontaneous pregnancy in one patient canceled for spontaneous ovulation and in one patient with fertilization failure.
DISCUSSION Our results did not confirm the potential beneficial effects of estrogen pretreatment in GnRH antagonist cycles on the 1362
number of retrieved oocytes but showed that this pretreatment, which is convenient for scheduling IVF, does not negatively affect cycle outcome. These data challenge the concept of the synchronization of the follicular cohort by the use of a steroid pretreatment. Similar lack of difference was seen with OCP pretreatment. Nevertheless, we can extrapolate from our previous study (11) that follicle size within the recruitable cohort should have been smaller in estrogen-pretreated patients than in patients without pretreatment. Indeed, in contrast to the design of our previous study, ovarian stimulation was started on the first day of estrogen discontinuation after a short free interval similar to that used in the study by Fanchin et al. (10). At that time, the FSH rebound has not yet occurred, as shown in our previous study (11) and therefore FSH levels are significantly lower than in nonpretreated patients. This observation leads us to suggest that the duration of the wash-out period after E2 pretreatment should be extended by 1 day to allow recovery of baseline FSH levels before starting ovarian stimulation. Indeed, the duration of the wash-out period is an important determining factor of FSH consumption, as outlined for OCP pretreatment, where a 5-day wash-out seems to be a better compromise than a 3-day wash-out between homogenization of the cohort and FSH consumption. Despite the use of similar starting doses of exogenous FSH, we observed on day 6 of ovarian stimulation a significantly lower number of medium-size follicles and a lower proportion of cycles with follicles >15 mm in patients pretreated by estrogen, suggesting a delay at the initiation of follicular growth. Furthermore, the duration of ovarian stimulation was nearly one day longer and total FSH requirement was significantly increased by 150 IU in estrogen-pretreated patients. VOL. 97 NO. 6 / JUNE 2012
Fertility and Sterility®
TABLE 4 Biologic parameters and cycle outcome. Parameter Oocytes retrieval (n) ICSI/IVF (n) Oocytes M2 oocytes Fertilization rate (%) Embryos Good-quality embryos (%) Transfers (n) D2/D3/D5 Replaced embryos Good-quality embryos (%) Positive b-hCG (n) US pregnancy (n) IR (yolk sacs/embryos) IR (fetal heart/embryos) Miscarriages or losses (n) Singleton (n) Twin (n) Total delivery (n) Delivery rate/cycle Delivery rate/retrieval Delivery rate/transfer
Estrogen (n [ 233)
No pretreatment (n [ 220)
P value
217 144/73 10.9 5.7 8.0 4.6 71 5.5 3.7 59 197 111/86/0 1.7 0.6 84.9 84 77 89/337 (26.4%) 84/337 (24.9%) 15 52 10 62 26.6% 28.6% 31.5%
204 147/57 10.2 5.6 7.6 4.7 68.1 4.8 3.2 65 192 99/92/1 1.7 0.6 88.8 84 78 95/323 (29.4%) 88/323 (27.2%) 12 52 14 66 30% 32.3% 34.3%
NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS
Note: Results are expressed as mean SD. D2/D3/D5 ¼ days 2, 3, and 5; ICSI ¼ intracytoplasmic sperm injection; IR ¼ implantation rate; IVF ¼ in vitro fertilization; M2 ¼ metaphase II. Cedrin-Durnerin. Estradiol treatment and antagonist protocol. Fertil Steril 2012.
The increased homogeneity of the follicular cohort induced by steroid pretreatment is supposed to improve the synchrony of follicular growth during stimulation and is expected to result in an increase in both oocyte yield and gamete quality. But we did not observe any significant difference in the number of medium, intermediate, and large follicles on the day of hCG administration, nor in the number of retrieved or mature oocytes. Our data contrast with those reported by Fanchin et al., who observed on the day of hCG an increased number of large-size follicles and mature oocytes, but difference in FSH requirement between groups was not assessed. In studies that showed a higher oocyte yield after OCP pretreatment, a high concomitant increase in total amount of gonadotropin was reported (12, 13), which biased the analysis of the specific effect of pretreatment. Finally, compelling evidence for an effect of pretreatment on the follicular cohort only came from the analysis of the kinetics of follicular growth during stimulation. The initiation of follicular growth was delayed, but a higher number of large-size follicles on the day of hCG administration was reported for both OCP (13) and E2 (10) pretreatments compared with no pretreatment. This pattern of follicular growth closely mimics the kinetics observed following down regulation in long GnRH agonist protocols (13–15). However, this change in follicular growth does not seem to result in better oocyte quality after OCP pretreatment, as reflected by the number of good-quality embryos obtained (13). In contrast, promising results were obtained after estrogen pretreatment with an increased number of good-quality embryos (10). Nevertheless, this was not confirmed by the present study performed in an adequately sized population. Finally, it seems that both OCP and E2 pretreatments, when administered to a selected population of young normal responders, are unable to VOL. 97 NO. 6 / JUNE 2012
increase quantity and quality of yielded oocytes, although the follicular cohort has been homogenized. This is not surprising, because follicular cohorts of normal size are rarely disorganized, in contrast to what can be observed in patients with decreased ovarian reserve. In the latter population, it is likely that the benefit of steroid pretreatment in terms of oocyte yield could be more easily demonstrated. Another important issue is the potential impact of steroid pretreatments on cycle outcome. Some concerns regarding the negative effect of OCP on the likelihood of pregnancy have been raised by randomized studies (13, 16, 17). Although those conclusions were not confirmed in retrospective analyses of large number of patients (18, 19), this major drawback has been reemphasized by successive metaanalyses of randomized studies (7, 8). Two hypotheses have been advanced to explain the underlying mechanism for this deleterious effect: a decrease in LH exposure or an effect on the endometrium. First, a decrease in serum LH levels induced by OCP pretreatment has been reported to impair ovarian response and implantation in a subset of patients (20). However, the impact of low LH levels on success rate in GnRH antagonist protocols is still a quite controversial issue, with some studies showing an improved outcome (21) and others an impaired outcome (22) or no consequence (23). Similarly, a slight but significant decrease in serum LH levels was observed in the present study after discontinuation of estrogen pretreatment. It was shown that these residual LH levels increased with the length of exposure to estrogen pretreatment after the occurrence of menses but were not significantly correlated with cycle outcome (24). On the other hand, an increased incidence of premature LH rise before the introduction of GnRH antagonist could impair outcome of cycles programmed by 1363
ORIGINAL ARTICLE: ASSISTED REPRODUCTION estrogen pre-treatment (15). Nevertheless, our results do not corroborate any significant difference in the incidence of premature LH surge in estrogen-pretreated patients compared with patients without pretreatment. Second, OCP was reported to negatively affect implantation by lowering endometrial thickness (16), or by altering E2 and P endometrial receptors (17). Estrogen pretreatment could offer an interesting alternative, because the endometrium is thicker at the initiation of ovarian stimulation. However, the trend to higher pregnancy rate previously reported with E2 pretreatment (10) has not been confirmed by our results. Therefore, in the absence of any deleterious effect on cycle outcome, estrogen pretreatment should be used in clinical practice for programming antagonist cycles to favor oocyte retrievals during working days. In conclusion, our data confirm that estrogen pretreatment is associated with requirement of higher FSH doses and longer duration of ovarian stimulation, whereas we did not observe any significant increase in the number of retrieved oocytes and good-quality embryos. In addition, estrogen pretreatment did not affect cycle outcome. The impact of increased FSH requirement and slightly longer duration of treatment in terms of cost and discomfort for patients is balanced by the improvement in scheduling allowed by estrogen pretreatment.
9.
10.
11.
12.
13.
14.
15.
16.
Acknowledgments: The authors thank Dr. C. Perves for her help in organizing investigator meetings.
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
Devroey P, Aboulghar M, Garcia-Velasco J, Griesinger G, Humaidan P, Kolibianakis E, et al. Improving the patient's experience of IVF/ICSI: a proposal for an ovarian stimulation protocol with GnRH antagonist co-treatment. Hum Reprod 2009;24:764–74. Al-Inany HG, Youssef MAFM, Aboulghar M, Broekmans F, Sterrenburg M, Smit J, et al. Gonadotrophin-releasing hormone antagonists for assisted reproductive technology. Cochrane Database Syst Rev 2011;(5):CD001750. Griesinger G, Felderbaum R, Diedrich K. GnRH antagonists in ovarian stimulation: a treatment regimen of clinicians'second choice? Data from the german national IVF registry. Hum Reprod 2005;20:2373–5. Al-Inany HG, Abbou Setta AM, Aboulghar M. Gonadotrophin-releasing hormone antagonists for assisted conception: a Cochrane review. Reprod Biomed Online 2007;14:640–9. Kolibianakis EM, Collins J, Tarlatzis BC, Devroey P, Diedrich K, Griesinger G. Among patients treated for IVF with gonadotrophins and GnRH analogues, is the probability of live birth dependent on the type of analogue used? A systematic review and meta-analysis. Hum Reprod Update 2007;12: 651–71. Barmat LI, Chantilis SJ, Hurst BS, Dickey RP. A randomized prospective trial comparing gonadotropin-releasing hormone (GnRH) antagonist/recombinant follicle-stimulating hormone (rFSH) versus GnRH-agonist/rFSH in women pretreated with oral contraceptives before in vitro fertilization. Fertil Steril 2005;83:321–30. Griesinger G, Venetis CA, Marx T, Diedrich K, Tarlatzis BC, Kolibianakis EM. Oral contraceptive pill pre-treatment in ovarian stimulation with GnRH antagonists for IVF: a systematic review and meta-analysis. Fertil Steril 2008; 90:1055–63. Smulders B, van Oirschot SM, Farquhar C, Rombauts L, Kremer JAM. Oral contraceptive pill, progestogen or estrogen pre-treatment for ovarian stimulation protocols for women undergoing assisted reproductive techniques [review]. Cochrane Database Syst Rev 2010;(1):CD006109.
1364
17.
18.
19.
20.
21.
22.
23.
24.
Cedrin-Durnerin I, Bulwa S, Herve F, Martin-Pont B, Uzan M, Hugues JN. The hormonal flare-up following GnRH agonist administration is influenced by a progestogen pre-treatment. Hum Reprod 1996;11:1859–63. Fanchin R, Cunha-Filho JS, Schonaeur LM, Kadoch IJ, Cohen Bacri P, Frydman R. Coordination of early antral follicles by luteal estradiol administration provides a basis for alternative controlled ovarian hyperstimulation regimens. Fertil Steril 2003;79:316–21. Cedrin-Durnerin I, Bst€andig B, Parneix I, Bied-Damon V, Avril C, Decanter C, et al. Effects of oral contraceptive, synthetic progestogen or natural estrogen pre-treatments on the hormonal profile and the antral follicle cohort before GnRH antagonist protocol. Hum Reprod 2007;22:109–16. Huirne JA, Hugues JN, Pirard C, Fischl F, Sage JC, Pouly JL, et al. Cetrorelix in an oral contraceptive-pretreated stimulation cycle compared with buserelin in IVF/ICSI patients treated with r-hFSH: a randomized, multicentre, phase IIIb study. Hum Reprod 2006;21:1408–15. Rombauts L, Healy D, Norman RJ, Orgalutran Scheduling Study Group. A comparative randomized trial to assess the impact of oral contraceptive pre-treatment on follicular growth and hormone profiles in GnRH antagonist-treated patients. Hum Reprod 2006;21:95–103. Cunha Filho JS, Terres LF, Holanda F, Freitas F, Glitz C, Genro VK, et al. Luteal phase administration in ovarian stimulation cycles with antagonist is comparable to the GnRH agonist (long) protocol. J Assist Reprod Genet 2007;24: 326–30. Ye H, Huang GN, Zeng PH, Pei L. IVF/ICSI outcomes between cycles with luteal estradiol (E2) pre-treatment before GnRH antagonist protocol and standard long agonist protocol: a prospective and randomized study. J Assist Reprod Genet 2009;26:105–11. Kolibianakis EM, Papanikolaou EG, Camus M, Tournaye H, van Steirteghem AC, Devroey P. Effect of oral contraceptive pill pre-treatment on ongoing pregnancy rates in patients stimulated with GnRH antagonists and recombinant FSH for IVF. A randomized controlled trial. Hum Reprod 2006;21:352–7. Nyboe Andersen A, Witjes H, Gordon K, Mannaerts B, Xpect Investigators. Predictive factors of ovarian response and clinical outcome after IVF/ICSI following a rFSH/GnRH antagonist protocol with or without oral contraceptive pre-treatment. Hum Reprod 2011;12:3413–23. Pinkas H, Sapir O, Avrech OM, Ben-Haroush A, Ashkenzi J, Fisch B, et al. The effect of oral contraceptive pill for cycle scheduling prior to GnRH-antagonist protocol on IVF cycle parameters and pregnancy outcome. J Assist Reprod Genet 2008;25:29–33. Bellver J, Albert C, Labarta E, Pellicer A. Early pregnancy loss in women stimulated with gonadotropin-releasing hormone antagonist protocols according to oral contraceptive pill pre-treatment. Fertil Steril 2007;87: 1098–101. Meldrum DR, Scott RT, Levy MJ, Alper MM, Noyes N. Oral contraceptive pretreatment in women undergoing controlled ovarian stimulation in ganirelix acetate cycles may, for a subset of patients be associated with low serum luteinizing hormone levels, reduced ovarian response to gonadotropins, and early pregnancy loss. Fertil Steril 2009;91:1963–5. Kolibianakis EM, Zikopoulous K, Schiettecatte J, Smitz J, Tournaye H, Camus M, et al. Profound LH suppression after GnRH antagonist administration is associated with a significantly higher ongoing pregnancy rate in IVF. Hum Reprod 2004;19:2490–6. Huirne JAF, van Loenen ACD, Schats R, McDonnell J, Hompes PGA, Schoemaker J, et al. Dose-finding study of daily GnRH antagonist for the prevention of premature LH surges in IVF/ICSI patients: optimal change in LH and progesterone for clinical pregnancy. Hum Reprod 2005;20:359–67. Doody KJ, Devroey P, Leader A, Witges H, Mannaerts BM. No association between endogenous LH and pregnancy in a GnRH antagonist protocol: part1, corifollitropin alpha. RBM Online 2011;23:449–56. Guivarc'h-Lev^eque A, Homer L, Arvis P, Broux PL, Moy L, Priou G, et al. Programming in vitro fertilization retrievals during working days after gonadotropin releasing hormone antagonist protocol with estrogen pretreatment: does the length of exposure to estradiol impact on controlled ovarian hyperstimulation outcomes? Fertil Steril 2011;96:872–6.
VOL. 97 NO. 6 / JUNE 2012
Fertility and Sterility®
SUPPLEMENTAL FIGURE 1 Enrolment Assessed for eligibility (n=not available) Randomized (n=472) Allocation Allocated to estrogen (n=238) Received estrogen (n=234) Did not receive estrogen (n=4) Spontaneous pregnancy (n=1) Discontinued (n=3)
Allocated to no pre-treatment (n=234) Received no pre-treatment (n=222) Did not receive no prettt (n=12) Spontaneous pregnancy (n=9) Discontinued (n=3)
Stimulation Start stimulation (n=233) No stimulation (n=1) No periods (n=1)
Start stimulation (n=220) No stimulation (n=2) Ovarian cyst (n=1) Protocole exchange (n=1)
Oocyte retrieval Retrieval (n=217) No retrieval (n=16) Poor response (n=10) High response (n=2) Other reason (n=4) 1 preg.
Retrieval (n=204) No retrieval (n=16) Poor response (n=9) High response (n=3) Other reason (n=4)
Transfer Transfer (n=197) No transfer (n=20) No mature oocyte (n=4) Fertilization failure (n=9) OHSS risk (n=2) Other (n=5)
Transfer (n=192) No transfer (n=12) No mature oocyte (n=2) Fertilization failure (n=5) 1 preg. OHSS risk (n=1) Other (n=4)
Study flow chart. Cedrin-Durnerin. Estradiol treatment and antagonist protocol. Fertil Steril 2012.
VOL. 97 NO. 6 / JUNE 2012
1364.e1