Human Reproduction Update Advance Access originally published online on April 3, 2007
Human Reproduction Update 2007 13(4):343-355; doi:10.1093/humupd/dmm007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Is progesterone elevation on the day of human chorionic gonadotrophin administration associated with the probability of pregnancy in in vitro fertilization? A systematic review and meta-analysis
1 Unit for Human Reproduction, 1st Department of Obstetrics and Gynaecology, Papageoriou General Hospital, Aristotle University of Thessaloniki, Nea Efkarpia, Peripheral Road, Thessaloniki, Greece 2 Centre for Reproductive Medicine, Dutch-Speaking Brussels Free University, and Laarbeeklaan 101, 1090 Brussels, Belgium
3 Correspondence address: Unit for Human Reproduction, 1st Department of Obstetrics and Gynaecology, Papageorgiou General Hospital, Nea Efkarpia, Peripheral Road, Thessaloniki 54603, Greece. Tel: +30 2310 414089; Fax: +30 2310 414089; E-mail: stratis.kolibianakis{at}irg.gr
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and methodsResultsDiscussionSupplementary materialAcknowledgementsReferences |
|---|
The role of progesterone elevation on in vitro fertilization (IVF) outcome has remained a debatable issue for several years. The aim of this systematic review and meta-analysis was to evaluate whether progesterone elevation on the day of human chorionic gonadotrophin (hCG) administration is associated with the probability of pregnancy. Eligible studies were considered those in which patients did not participate more than once. A literature search in MEDLINE, EMBASE and CENTRAL identified 12 eligible studies, 10 of which were retrospective. The majority (n = 10) of these studies did not detect a statistically significant association between progesterone elevation and the probability of pregnancy. Meta-analysis was performed only for the studies (n = 5) that provided data on clinical pregnancy per patient reaching hCG administration for final oocyte maturation. No statistically significant association between progesterone elevation and the probability of clinical pregnancy was detected (Odds ratio: 0.75, 95% confidence interval 0.53–1.06; P = 0.10). This finding persisted in the sensitivity analyses performed, which excluded the studies that did not report clearly that measurement of progesterone did not affect patients' management and those that did not report definition of clinical pregnancy. In addition, subgroup analyses were conducted on the basis of type of gonadotrophin-releasing hormone GnRH analogue used and on the value of serum threshold used to classify patients in those with or without progesterone elevation. These analyses, however, did not materially change the results obtained. In conclusion, the best available evidence does not support an association between progesterone elevation on the day of hCG administration and the probability of clinical pregnancy in women undergoing ovarian stimulation with GnRH analogues and gonadotrophins for IVF.
Key words: hCG / ovarian stimulation / pregnancy / progesterone
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary materialAcknowledgementsReferences |
|---|
The introduction of gonadotrophin-releasing hormone (GnRH) analogues for pituitary suppression in in vitro fertilization (IVF) significantly decreased the incidence of premature luteinizing hormone (LH) surge (Smitz et al., 1992
). Despite pituitary down-regulation, however, several researchers have described a phenomenon reported as premature luteinization (Hofmann et al., 1993; Legro et al., 1993; Ubaldi et al., 1996a; Bosch et al., 2003). This refers to a rise in serum progesterone levels on the day of human chorionic gonadotrophin (hCG) administration for final oocyte maturation above a threshold level, which is usually arbitrarily defined. Its incidence varies, ranging between 5 and 30% in IVF patients (Edelstein et al., 1990; Schoolcraft et al., 1991; Silverberg et al., 1991; Fanchin et al., 1993; Givens et al., 1994; Ubaldi et al., 1995).
At present, there is no consensus on whether progesterone elevation on the day of hCG administration is associated with the achievement of pregnancy. Several studies have denied the presence of such an association (Edelstein et al., 1990; Silverberg et al., 1991; Antoine et al., 1992; Check et al., 1994; Givens et al., 1994; Bustillo et al., 1995; Levy et al., 1995; Ubaldi et al., 1995; Abuzeid and Sasy, 1996; Huang et al., 1996; Miller et al., 1996; Moffit et al., 1997; Doldi et al., 1999; Lindheim et al., 1999; Urman et al., 1999; Martinez et al., 2004), whereas others have confirmed the presence of a negative association (the probability of pregnancy decreases significantly when serum progesterone on the day of hCG administration for final oocyte maturation rises above a threshold level) (Schoolcraft et al., 1991; Check et al., 1993a; Fanchin et al., 1993, 1997a; Harada et al., 1995; Shulman et al., 1996; Bosch et al., 2003).
It should be noted that if a negative association between progesterone elevation on the day of hCG administration and the probability of pregnancy exists, it might be worth examining the possibility of cryopreserving the resulting embryos and their transfer in a subsequent frozen-thawed cycle (Silverberg et al., 1991; Legro et al., 1993; Silverberg et al., 1994) or alternatively, administering hCG at an earlier time in the follicular phase, prior to progesterone elevation (Harada et al., 1996). On the contrary, absence of an association indicates that assessment of serum progesterone on the day of hCG administration might be redundant.
The purpose of this systematic review and meta-analysis was to address the following question: among patients undergoing ovarian stimulation for IVF using GnRH analogues and gonadotrophins, is progesterone elevation on the day of hCG administration associated with the probability of pregnancy?
|
Materials and methods |
|---|
|
TOPAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary materialAcknowledgementsReferences |
|---|
Search strategy
For the purpose of this systematic review, a literature search in MEDLINE, EMBASE and CENTRAL electronic databases was performed by two of the authors (C.A.V. and E.M.K. from 10 October 2005 to 31 December 2005 to identify studies which answered the following question: among patients undergoing ovarian stimulation for IVF using GnRH analogues and gonadotrophins, is progesterone elevation on the day of hCG administration associated with the probability of pregnancy? A computerized literature search was performed using the terms ‘progesterone’ as medical subject heading under ‘ovulation induction’ or ‘hormonal therapy’ or ‘infertility therapy’ and limited to ‘human’ and ‘female’. No language limitations were applied. Additionally, references of retrieved articles were hand-searched.
In order for the studies to be eligible for this systematic review, the following inclusion criteria were established prior to the literature search: (i) GnRH analogue should have been used for down-regulation and gonadotrophins for ovarian stimulation, (ii) patients should have been classified as showing or not showing progestone elevation on the day of hCG administration for final oocyte maturation, (iii) patients should not have entered the study more than once (number of patients equal to number of cycles performed), (iv) the study should provide data on pregnancy outcome in patients showing or not showing progesterone elevation on the day of hCG administration and (v) the study should have been published in a peer reviewed journal.
In case the pregnancy outcome reported in the studies evaluated was not using as a denominator patients reaching hCG administration for final oocyte maturation, but patients reaching oocyte retrieval or patients in whom embryo transfer (ET) was performed, the authors were contacted and asked to provide the missing information.
All studies identified, which addressed the research question, were initially considered for the present systematic review, regardless of the direction of study (retrospective or prospective), their sample size or the threshold of serum progesterone on the day of hCG administration used to classify the patients as showing or not showing progesterone elevation. Studies that reported that progesterone levels during ovarian stimulation were used to modify patients' management were excluded.
Literature search was conducted for the time interval between 1984, when the first study describing the use of agonists in IVF appeared in the literature (Porter et al., 1984), and 2005. This electronic search resulted in the retrieval of 1114 publications. Subsequently, the titles of these manuscripts were examined to exclude irrelevant studies, resulting in 104 potentially eligible publications. The abstracts of these studies were examined and, eventually, 45 manuscripts that could provide data to answer the research question were identified (Fig. 1). The full text of these studies was examined thoroughly, resulting in the exclusion of 33 publications (Supplementary material, Table I). Specifically, studies were excluded because: (i) they did not properly address the research question (n = 6), (ii) they did not use GnRH analogues for pituitary suppression (n = 2), (iii) they included patients more than once (n = 18), (iv) they were letters to journals or reviews (n = 6) and (v) measurement of serum progesterone during ovarian stimulation affected patients' management (n = 1). The references of all the studies in which the full text was retrieved were hand-searched. However, no additional studies that could provide data to answer the research question were found. Eventually, 12 studies associating progesterone elevation on the day of hCG administration with pregnancy achievement were identified (Table 1).
|
|
Data extraction
The following data were recorded from each of the 12 eligible studies: demographic (type of study, citation data, country, study period, number of patients included, number of cycles performed, selection of cycles and indication for treatment), procedural (type of down-regulation and protocol of ovarian stimulation, type of gonadotrophin administered, dose and criteria for hCG administration, time of oocyte retrieval, type of fertilization, day of ET, number of embryos transferred, type of luteal support administered, serum progesterone threshold level used for classification of patients in those showing and those not showing progesterone elevation and type of assay used for the measurement of progesterone on the day of hCG) and outcome data [pregnancy achievement, duration and total dose of follicle-stimulating hormone (FSH) required for ovarian stimulation, serum estradiol (E2) levels on the day of hCG administration, number of cumulus–oocyte complexes (COCs) retrieved and fertilization rates]. Any disagreement was resolved unanimously by discussion.
The main outcome measures chosen for meta-analysis were clinical pregnancy (defined as the detection of fetal heart by ultrasound at 6–8 weeks of gestation), ongoing pregnancy (defined as the confirmation of viable pregnancy beyond 12 weeks of gestation) and live birth/delivery per patient reaching hCG administration for final oocyte maturation. In case the studies did not report data regarding one or more of the above outcome measures, the authors were contacted and asked to provide the missing information. Secondary outcome measures included duration and total dose of FSH required for ovarian stimulation, serum E2 levels on the day of hCG administration, number of COCs retrieved and fertilization rates.
The dichotomous data results for each of the studies eligible for meta-analysis were expressed as an odds ratio (OR) with 95% confidence intervals (CI). These results were combined for meta-analysis using the Mantel/Haenszel model, when using the fixed effects method, and the DerSimonian and Laird model, when using the random effects method.
When the outcome of interest was of a continuous nature, the differences were pooled across the studies, which provided information on this outcome parameter, resulting in a weighted mean difference (WMD) with 95% CI. The inverse variance method and the DerSimonian and Laird method were used, respectively, when the fixed or random effects method was applied.
All results were combined for meta-analysis with Revman Software (The Cochrane Collaboration, 2000). Study-to-study variation was assessed by using the 
2-statistic (the hypothesis tested was that the studies are all drawn from the same population, i.e. from a population with the same effect size). Moreover, due to the fact that the 2 test is considered to have low power to detect inconsistency across studies, when the studies have a small sample size or are limited in number, a P-value of 0.10 was used to determine statistical significance (Dickersin and Berlin, 1992). A fixed effects model was used, where no heterogeneity was present, whereas in the presence of significant heterogeneity, a random effects model was applied. A funnel plot analysis and Egger's test were performed to detect the presence of publication bias.
Subgroup analyses were performed depending, on (i) the serum progesterone level used as a threshold by each study to classify patients as showing or not showing progesterone elevation on the day of hCG administration and (ii) the type of GnRH analogue used for down-regulation.
Sensitivity analyses were carried out to check the stability of the results obtained by excluding (i) the studies that did not report clearly that the measurement of progesterone during ovarian stimulation did not affect patients' management and (ii) the studies that did not provide the definition of the pregnancy outcome examined.
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary materialAcknowledgementsReferences |
|---|
Twelve studies fulfilled the inclusion criteria for the systematic review (see Table 1). A total of 2733 patients were reviewed (GnRH antagonists: n = 109, GnRH agonists: n = 2624).
Characteristics of the eligible studies are listed in Tables 1 and 2. All included studies were published between 1990 and 2004. The majority of the studies were published in Fertility Sterility (n = 6) and Human Reproduction (n = 2). Ten studies were retrospective in design, one study was prospective and in one study it was unclear whether a retrospective or prospective design was followed. The size of the studies ranged from 24 to 911 patients, and the median number of patients included was 125.
|
For ovarian stimulation, a combination of urinary and recombinant gonadotrophins was used in two studies, in nine studies this was achieved with urinary gonadotrophins, whereas in one study ovarian stimulation was performed with recombinant gonadotrophins. Criteria for triggering final oocyte maturation varied markedly across studies and were based on follicular data (n = 2), on a combination of follicular data and E2 levels (n = 8), whereas in two studies these criteria were not reported. In the majority of the studies (n = 10), the medication used for triggering final oocyte maturation was urinary hCG (10 000 IU), whereas in two studies this was not reported. Oocyte retrieval was performed 34–37 h after hCG administration in 10 studies, whereas 2 studies did not provide any details about the timing of oocyte retrieval (Supplementary material, Table II).
To inhibit premature LH surge, the long luteal agonist protocol was used in eight studies, in two studies this was performed using both the long luteal and the short agonist protocol in the same study, whereas in two studies a fixed day-6 antagonist protocol was applied. In the agonist group of studies, nafarelin, triptorelin, leuprolide acetate and buserelin were employed in various protocols, whereas cetrorelix was the analogue used in the antagonist group. Details about the GnRH analogue protocols used in the studies analysed are presented in Supplementary material, Table II.
For the measurement of serum progesterone on the day of hCG administration, all the studies used commercially available immunoassay kits. The inter-assay coefficient of variance (CV) ranged between 3.9 and 34%, whereas the intra-assay CV ranged between 2.9 and 11.9%. In one study, the inter- or intra-assay CV was not reported (Table 2).
Fertilization methods included IVF (n = 7), intracytoplasmic sperm injection (ICSI) (n = 2) and IVF/ICSI (n = 3). ETs were performed in the majority of studies (n = 8) on day 2 or 3 after oocyte retrieval, in one study on day 1, whereas in three studies the day of ET was not reported. Luteal support varied between studies as well as within a study (Miller et al., 1996). Four studies did not provide details about the type of luteal support used (Supplementary material, Table II).
The threshold of serum progesterone on the day of hCG administration, used to classify patients as showing or not showing progesterone elevation varied among studies. For this purpose, eight studies used 0.9 ng/mL, three studies used 1.0 ng/mL and one study used 1.2 ng/mL. In the majority of the studies (n = 9) analysed, this threshold was selected arbitrarily, whereas in three studies, receiver operator characteristics (ROC) curve analysis was used (Table 2).
Two studies (Check et al., 1993a; Bosch et al., 2003) detected a negative association between elevation of progesterone on the day of hCG administration and the pregnancy outcomes examined. On the contrary, Urman et al. (1999) reported a significantly higher probability of clinical, but not ongoing, pregnancy in patients with progesterone elevation. The remaining studies (n = 9) did not find any association between the elevation of progesterone on the day of hCG administration and the pregnancy outcomes analysed (Table 3).
|
Meta-analysis
Clinical pregnancy rate
The probability of clinical pregnancy did not differ significantly between patients with and those without progesterone elevation on the day of hCG administration (OR: 0.75, 95% CI: 0.53–1.06; P = 0.10; heterogeneity: P = 0.12; fixed effects model) (Fig. 2). Clinical pregnancy rate was lower, but not significantly so, in the group with progesterone elevation [rate difference (RD): –10%, (95% CI: –22%–+2%; P = 0.11; heterogeneity: P = 0.04; random effects model] (Fig. 3). A funnel plot of the included studies is shown in Supplementary material, Figure 1. No publication bias was detected in the studies analysed (Egger's test: P = 0.60). The exclusion of the study that did not provide definition for clinical pregnancy (Edelstein et al., 1990) led to similar results (OR: 0.59, 95% CI: 0.24–1.43; P = 0.24; heterogeneity: P = 0.08, random effects model) (Supplementary material, Figure 2). Finally, these results did not change substantially, by excluding the studies that did not report clearly that the measurement of progesterone during ovarian stimulation did not affect patients' management (OR: 0.86, 95% CI: 0.59–1.25; P = 0.44; heterogeneity: P = 0.46; fixed effects model) (Supplementary material, Figure 3).
|
|
Subgroup analyses regarding the probability of clinical pregnancy were conducted on the basis of serum progesterone threshold, used to classify patients in those showing and those not showing progesterone elevation and in addition, on the type of analogue used for down–regulation. When only the studies that used 0.9 ng/mL as a threshold for the classification of patients in those without progesterone elevation were analysed (n = 4), the probability of clinical pregnancy between patients with and those without progesterone elevation on the day of hCG administration remained not significantly different (OR: 0.88, 95% CI: 0.61–1.28; P = 0.51; heterogeneity: P = 0.56; fixed effects model) (Fig. 4). The subgroup analysis based on the type of GnRH analogue used for down-regulation did not materially change the original results. In the agonist group of studies (n = 3), the probability of clinical pregnancy between patients with and those without progesterone elevation on the day of hCG administration was not significantly different (OR: 0.86, 95% CI: 0.59–1.25; P = 0.44; heterogeneity: P = 0.46; fixed effects model), which was also the case when the two antagonist studies were pooled (OR: 0.57, 95% CI: 0.09–3.56; P = 0.55; heterogeneity: P = 0.10; random effects model) (Fig. 5).
|
|
It should be noted that both female age (Supplementary material, Figure 4) and the number of embryos transferred (Supplementary material, Figure 5) were not significantly different between the groups that did and did not exhibit progesterone elevation on the day of hCG administration (female age: WMD: 0.06 years, 95% CI: –1.25–+1.38, P = 0.92; heterogeneity: P = 0.02; random effects model; number of embryos transferred: WMD: 0.07 embryos 95% CI: –0.06–+0.19, P = 0.29; heterogeneity: P = 0.68; fixed effects model).
Ongoing pregnancy rate/live birth rate
Pooling of data for these outcome measures in the studies eligible for meta-analysis was not considered appropriate, due to the fact that these were either not uniformly defined or not reported (Table 3). Ongoing pregnancy rate per patient reaching hCG administration was present in two of the studies included in the current systematic review (Edelstein et al., 1990; Bosch et al., 2003). In the study by Bosch et al. (2003), ongoing pregnancy was defined as the confirmation of pregnancy beyond 20 weeks of gestation, whereas in the study by Edelstein et al. (1990), ongoing pregnancy was not defined. Live birth rate per patient reaching hCG administration was reported only in one study (Martinez et al., 2004).
FSH requirement
The number of FSH ampoules required was not significantly different between patients with and those without progesterone elevation (WMD: 0.92 ampoules, 95% CI: –1.74–+3.59; P = 0.50; heterogeneity: P = 0.06; random effects model) (Supplementary material, Figure 6). Four studies offered data for this outcome measure.
|
Duration of FSH stimulation
The duration of stimulation did not differ significantly between patients with and those without progesterone elevation (WMD: 0.08 days, 95% CI: –0.57–+0.73; P = 0.81; heterogeneity: P = 0.003; random effects model) (Supplementary material, Figure 7). Three studies offered data for this outcome measure.
E2 levels on the day of hCG administration
The E2 levels (pg/mL) on the day of hCG administration were significantly higher in the group of patients that exhibited progesterone elevation on the day of hCG compared with those who did not (WMD: 413.06 pg/mL, 95% CI: 240.14–585.99; P < 0.00001; heterogeneity: P = 0.21; fixed effects model) (Fig. 6). Three studies offered data for this outcome measure.
COCs retrieved
No statistically significant difference in the number of COCs retrieved was detected between the patients with and those without progesterone elevation on the day of hCG administration (WMD: 1.87 COCs, 95% CI: –0.07–+3.87; P = 0.06; heterogeneity: P = 0.10; random effects model) (Supplementary material, Figure 8). Four studies offered data for this outcome measure. However, a subgroup analysis depending on the type of GnRH analogue used showed that in the group of agonist studies significantly more oocytes were retrieved in the progesterone elevation group compared with the group without progesterone elevation (WMD: 2.96 COCs, 95% CI: +1.74–+4.18; P < 0.00001; heterogeneity: P = 0.13; fixed effects model). This difference was not present when the two antagonist studies were pooled (WMD: 0.00 COCs, 95% CI: –2.98–+2.99; P = 1.00; heterogeneity: P = 0.44; fixed effects model) (Supplementary material, Figure 9).
Fertilization rates
Fertilization rates were not significantly different between patients with and those without progesterone elevation on the day of hCG administration (WMD: –1.71%, 95% CI: –5.32–+1.90; P = 0.35; heterogeneity: P = 0.78; fixed effects model) (Supplementary material, Figure 10). Two studies offered data for this outcome measure.
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary materialAcknowledgementsReferences |
|---|
The current systematic review suggests that progesterone elevation, on the day of hCG administration for final oocyte maturation, does not appear to be associated with the probability of pregnancy, in women undergoing ovarian stimulation with GnRH analogues and gonadotrophins for IVF (Table 3).
In the present review, an effort was made to establish firm inclusion criteria, so that reliable conclusions could be drawn. Accordingly, only studies that did not violate the independence assumption were considered eligible. The statistical tests used in the studies analysed, to evaluate the effect of progesterone elevation on the day of hCG administration on the probability of pregnancy, assume that all the observations are made on subjects that are independent from each other (Cochran, 1974; Kruskal, 1988). The inclusion of patients more than once violates the assumption of independence and has been shown to inflate P values (Zimmerman, 1993; Levine and Rockhill, 2006).
Meta-analysis was performed only for the studies that provided data on clinical pregnancy per patient reaching hCG administration for final oocyte maturation (Fig. 2). Studies that reported pregnancy rate per patient reaching oocyte retrieval or per patient having ET performed were not considered for pooling of data. This was due to the fact that the knowledge of the association of progesterone elevation on the day of hCG administration and the probability of clinical pregnancy is important, prior to hCG administration. If an adverse effect of progesterone elevation on the probability of clinical pregnancy was supported by the results of the current meta-analysis, the clinician would still be able to administer earlier hCG, before progesterone elevation occurs (Harada et al., 1996), or to cryopreserve embryos and cancel a fresh transfer (Silverberg et al., 1991, 1994; Legro et al., 1993). The current meta-analysis, however, suggests that an association between progesterone elevation and the probability of clinical pregnancy per patient reaching hCG administration is not present.
The lack of association between progesterone elevation on the day of hCG and clinical pregnancy remained constant in the subgroup analysis performed, based on the type of GnRH analogue used for pituitary suppression (agonist–antagonist) (Fig. 5). In the studies where LH surge was inhibited by GnRH agonists, exactly the same threshold level of progesterone (0.9 ng/mL) was used to classify patients in those with or without progesterone elevation. Subsequently, in this case, both statistical and clinical heterogeneity is likely to be limited rendering the results obtained more solid. In addition, no association was detected between progesterone elevation and the achievement of clinical pregnancy when the two studies (Ubaldi et al., 1996b; Bosch et al., 2003) in which pituitary suppression was performed by GnRH antagonists were pooled. Significant statistical heterogeneity was present (P = 0.10) and although, this might be attributed to the use of a different threshold of progesterone (0.9 ng/mL versus 1.2 ng/mL, respectively), a solid conclusion regarding the source of this heterogeneity could not be drawn, due to the limited number of eligible studies (n = 2).
This lack of association was also confirmed in the sensitivity analyses performed by excluding the studies that did not report the definition of clinical pregnancy (Supplementary material, Figure 2) or those that did not report clearly whether the measurement of serum progesterone level influenced the way ovarian stimulation was performed (Supplementary material, Figure 3).
It should be noted that the available studies, which examined the research question, have often chosen arbitrarily the threshold of serum progesterone used to classify patients in those with or without progesterone elevation (Table 2). A subgroup analysis, including only studies that used exactly the same progesterone threshold, did not, however, change the results obtained (Fig. 4).
A more appropriate method to analyse the possible association of serum progesterone on the day of hCG with the probability of pregnancy might be the use of ROC curve analysis, which was performed in three of the studies included in the current systematic review (Silverberg et al., 1991; Urman et al., 1999; Bosch et al., 2003) (Table 2) with contradictory, however, results. This method is able to identify an optimal serum progesterone threshold on the basis of which patients can be classified into pregnant or not pregnant with a certain probability. This threshold characterizes the specific population analysed, the method of progesterone assessment and the protocol of treatment used.
It should be noted that the proportion of patients with progesterone elevation varied widely even among studies (Silverberg et al., 1991; Martinez et al., 2004) in which the same serum progesterone threshold (0.9 ng/mL) and the same type of GnRH analogue (agonist) were used (12.4% versus 52.3%, respectively). This marked variation in the incidence of progesterone elevation has been previously described in the literature (Edelstein et al., 1990; Schoolcraft et al., 1991; Silverberg et al., 1991; Fanchin et al., 1993; Givens et al., 1994; Ubaldi et al., 1995), and although, it might be assumed that it is due to discrepancies in population characteristics and/or treatment protocols among the studies, its explanation is not clear. The effect of this phenomenon on the association of progesterone elevation and the achievement of pregnancy was not possible to be explored in the current meta-analysis due to the limited number of eligible studies.
Considering secondary outcome measures, it appears that no difference is present between patients with and without progesterone elevation regarding the total dose of FSH administered (Supplementary material, Figure 6), the mean duration of stimulation (Supplementary material, Figure 7) and fertilization rates (Supplementary material, Figure 10).
The lack of association between progesterone elevation and fertilization rates, might be indicative of the absence of a detrimental effect of elevated progesterone on the day of hCG administration on oocyte quality, which has also been suggested previously (Hofmann et al., 1993; Legro et al., 1993; Check et al., 1994; Fanchin et al., 1996; Shulman et al., 1996; Moffitt et al., 1997; Bosch et al., 2003; Martinez et al., 2004).
Many researchers in the past have adopted the term ‘premature luteinization’ for patients with progesterone elevation on the day of hCG administration for final oocyte maturation (Hofmann et al., 1996; Legro et al., 1993; Ubaldi et al., 1996a, b; Bosch et al., 2003). This suggests that the excessive amount of progesterone is produced by granulosa cells that have started the process of luteinization.
However, the fact that a significantly higher mean number of COCs (Supplementary material, Figure 9), accompanied by a higher mean E2 level on the day of hCG administration (Figure 6), were present in the group of patients with elevated progesterone, when only the agonist studies were pooled, suggests an alternative explanation. It is likely, at least regarding the patients treated with GnRH agonists, that the elevated progesterone might be attributed to an excess number of follicles, each one producing a normal, for the late follicular phase, amount of progesterone. In this way, excess of proliferating granulosa cells leads to an increased progesterone production, independently of LH exposure.
In theory, if the follicles of patients that exhibit progesterone elevation had started the process of luteinization, then it might be expected that the resulting oocytes from those patients would be of lower quality. However, several studies have failed to detect a detrimental effect of progesterone elevation on oocyte and embryo quality (Hofmann et al., 1993; Legro et al., 1993; Check et al., 1994; Fanchin et al., 1996; Shulman et al., 1996; Moffitt et al., 1997; Martinez et al., 2004), which is in line with the results of the present study.
Thus, at least for the studies using GnRH agonists to inhibit LH surge, the use of the term ‘premature luteinization’ in the presence of normal LH levels might not be appropriate. On the other hand, this might not be the case for the two antagonist studies, since it has been suggested that premature LH surge occurs more frequently in GnRH antagonist when compared with GnRH agonist cycles (Kolibianakis et al., 2006). However, the small number of patients analysed in the two eligible studies does not allow for solid conclusions to be drawn.
Despite the sustained interest in the role of progesterone elevation on IVF outcome, reflected on the numerous relevant studies published until today, this systematic review reveals the lack of well-designed prospective studies that could answer the research question asked. In conclusion, the best available evidence does not support an association between progesterone elevation on the day of hCG administration and the probability of clinical pregnancy in women undergoing ovarian stimulation with GnRH analogues and gonadotrophins for IVF.
|
Supplementary material |
|---|
|
TOPAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary materialAcknowledgementsReferences |
|---|
Supplementary material is available at Human Reproduction Update online.
|
Acknowledgements |
|---|
|
TOPAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary materialAcknowledgementsReferences |
|---|
We would like to express our gratitude to the following persons for their support in this work: M.I. Abuzeid (Michigan, USA); J.M. Antoine (Paris, France); E. Bosch (Valencia, Spain); M. Bustillo (Florida, USA); J.H. Check (Pennsylvania, USA); J. Goldberg (Ohio, USA); S.R. Lindheim (Wisconsin, USA); F. Martinez (Barcelona, Spain); D.V. Moffitt (Arizona, USA); A. Pellicer (Valencia, Spain); M. Plachot (Sevres, France); B. Schoolcraft (Colorado, USA); K. Silverberg (Texas, USA); D. Smotrich (California, USA); F.M. Ubaldi (Rome, Italy); B. Urman (Istanbul, Turkey) and Y. Yaron (Tel Aviv, Israel).
|
References |
|---|
|
TOPAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary materialAcknowledgementsReferences |
|---|
Abuzeid MI, Sasy MA. Elevated progesterone levels in the late follicular phase do not predict success of in vitro fertilization-embryo transfer. Fertil Steril (1996) 65:981–5.[Web of Science][Medline]
Antoine JM, Firmin C, Salat-Baroux J, et al. Prognostic value of preovulatory elevations of plasma progesterone during in vitro fertilization using LHRH agonists in a long protocol. J Gynecol Obstet Biol Reprod (Paris) (1992) 21:629–34.[Medline]
Bodis J, Torok A, Garai J, et al. Preovulatory rise in progesterone (multiple letters) [2]. Hum Reprod (1998) 13:2657–8.
Bosch E, Valencia I, Escudero E, et al. Premature luteinization during gonadotropin-releasing hormone antagonist cycles and its relationship with in vitro fertilization outcome. Fertil Steril (2003) 80:1444–9.[CrossRef][Web of Science][Medline]
Bustillo M, Stern JJ, Coulam CB. Serum progesterone at the time of human chorionic gonadotrophin does not predict pregnancy in in-vitro fertilization and embryo transfer. Hum Reprod (1995) 10:2862–2867.
Check JH, Lurie D, Askari HA, et al. The range of subtle rise in serum progesterone levels following controlled ovarian hyperstimulation associated with lower in vitro fertilization pregnancy rates is determined by the source of manufacturer. Eur J Obstet Gynecol Reprod Biol (1993a) 52:205–209.[CrossRef][Web of Science][Medline]
Check JH, Lurie D, Hoover L, et al. Association of higher pregnancy rates with low serum progesterone levels (by radioimmunoassay) at the time of human chorionic gonadotropin not corroborated when using a nonisotopic immunoassay. J Assist Reprod Genet (1993b) 10:384–7.[CrossRef][Web of Science][Medline]
Check JH. Predictive value of serum progesterone levels for pregnancy outcome? Fertil Steril (1994) 62:1090–1.[Web of Science][Medline]
Check JH, Hourani C, Choe JK, et al. Pregnancy rates in donors versus recipients according to the serum progesterone level at the time of human chorionic gonadotropin in a shared oocyte program. Fertil Steril (1994) 61:262–4.[Web of Science][Medline]
Cochran WG. Some consequences when the assumptions in the analysis of variance are not satisfied. Biometrics (1974) 3:22–28.[CrossRef]
Dickersin K, Berlin JA. Meta-analysis: state-of-the-science. Epidemiol Rev (1992) 14:154–76.[Web of Science][Medline]
Dirnfeld M, Goldman S, Gonen Y, et al. A modest increase in serum progesterone levels on the day of human chorionic gonadotropin (hCG) administration may influence pregnancy rate and pregnancy loss in in vitro fertilization–embryo transfer (IVF–ET) patients. J Assist Reprod Genet (1993) 10:126–9.[CrossRef][Web of Science][Medline]
Doldi N, Marsiglio E, Destefani A, et al. Elevated serum progesterone on the day of HCG administration in IVF is associated with a higher pregnancy rate in polycystic ovary syndrome. Hum Reprod (1999) 14:601–5.
Dumesic DA. Periovulatory serum progesterone levels as a predictor of pregnancy outcome during ovarian hyperstimulation for assisted reproductive technology. Fertil Steril (1994) 62:911–2.[Web of Science][Medline]
Edelstein MC, Seltman HJ, Cox BJ, et al. Progesterone levels on the day of human chorionic gonadotropin administration in cycles with gonadotropin-releasing hormone agonist suppression are not predictive of pregnancy outcome. Fertil Steril (1990) 54:853–7.[Web of Science][Medline]
Emperaire JC, Ruffie A, Ruedas E, et al. The value of plasma progesterone for monitoring the IVF cycle. J Gynecol Obst Bio R (1989) 18:103–8.
Fanchin R. Understanding the genesis and consequences of premature progesterone elevation. Assist Reprod (1999) 9:121–8.
Fanchin R, de Ziegler D, Taieb J, et al. Premature elevation of plasma progesterone alters pregnancy rates of in vitro fertilization and embryo transfer. Fertil Steril (1993) 59:1090–4.[Web of Science][Medline]
Fanchin R, Righini C, Olivennes F, et al. Premature progesterone elevation does not alter oocyte quality in in vitro fertilization. Fertil Steril (1996) 65:1178–83.[Web of Science][Medline]
Fanchin R, Hourvitz A, Olivennes F, et al. Premature progesterone elevation spares blastulation but not pregnancy rates in in vitro fertilization with coculture. Fertil Steril (1997a) 68:648–52.[CrossRef][Web of Science][Medline]
Fanchin R, Righini C, Olivennes F, et al. Consequences of premature progesterone elevation on the outcome of in vitro fertilization: insights into a controversy. Fertil Steril (1997b) 68:799–805.[CrossRef][Web of Science][Medline]
Farina A, Ogasawara M, Kajiura S, et al. Predictive value of serum progesterone levels [3] (multiple letters). Fertil Steril (1998) 69:977–8.[Web of Science][Medline]
Givens CR, Schriock ED, Dandekar PV, et al. Elevated serum progesterone levels on the day of human chorionic gonadotropin administration do not predict outcome in assisted reproduction cycles. Fertil Steril (1994) 62:1011–7.[Web of Science][Medline]
Harada T, Yoshida S, Katagiri C, 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.
Harada T, Toda T, Katagiri C, et al. Altering the timing of human chorionic gonadotropin injection according to serum progesterone (P) concentrations improves embryo quality in cycles with subtle P rise. Fertil Steril (1996) 65:594–7.[Web of Science][Medline]
Hassiakos D, Toner JP, Muasher SJ, et al. Implantation and pregnancy rates in relation to oestradiol and progesterone profiles in cycles with and without the use of gonadotrophin-releasing hormone agonist suppression. Hum Reprod (1990) 5:1004–8.
Hofmann GE, Bentzien F, Bergh PA, et al. Premature luteinization in controlled ovarian hyperstimulation has no adverse effect on oocyte and embryo quality. Fertil Steril (1993) 60:675–9.[Web of Science][Medline]
Hofmann GE, Khoury J, Johnson CA, et al. Premature luteinization during controlled ovarian hyperstimulation for in vitro fertilization-embryo transfer has no impact on pregnancy outcome. Fertil Steril (1996) 66:980–6. 1996.[Web of Science][Medline]
Huang JC, Jackson KV, Hornstein MD, et al. The effect of elevated serum progesterone during ovulation induction in in vitro fertilization-embryo transfer. J Assist Reprod Genet (1996) 13:617–24.[CrossRef][Web of Science][Medline]
Jacobson A, Weckstein LN, Galen DI. Predictive value of progesterone and estradiol levels for pregnancy. Fertil Steril (1994) 61:192–3.[Web of Science][Medline]
Kagawa T, Yamano S, Nishida S, et al. Relationship among serum levels of luteinizing hormone, estradiol, and progesterone during follicle stimulation and results of in vitro fertilization and embryo transfer (IVF–ET). J Assist Reprod Genet (1992) 9:106–12.[CrossRef][Web of Science][Medline]
Kolibianakis EM, Collins J, Tarlatzis BC. 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 (2006) 12:651–71.
Kruskal W. Miracles and statistics: The casual assumption of independence. J Am Stat Assoc (1988) 83:929–937.[CrossRef][Web of Science]
Legro RS, Ary BA, Paulson RJ, et al. Premature luteinization as detected by elevated serum progesterone is associated with a higher pregnancy rate in donor oocyte in-vitro fertilization. Hum Reprod (1993) 8:1506–11.
Levine DW, Rockhill B. I Remember You: Independence and the Binomial Model. Teach Stat (2006) 28:44–8.[CrossRef]
Levy MJ, Smotrich DB, Widra EA, et al. The predictive value of serum progesterone and 17-OH progesterone levels on in vitro fertilization outcome. J Assist Reprod Genet (1995) 12:161–6.[CrossRef][Web of Science][Medline]
Lindheim SR, Cohen MA, Chang PL, et al. Serum progesterone before and after human chorionic gonadotropin injection depends on the estradiol response to ovarian hyperstimulation during in vitro fertilization—embryo transfer cycles. J Assist Reprod Genet (1999) 16:242–6.[CrossRef][Web of Science][Medline]
Martinez F, Coroleu B, Clua E, et al. Serum progesterone concentrations on the day of HCG administration cannot predict pregnancy in assisted reproduction cycles. Reprod Biomed Online (2004) 8:183–90.[Web of Science][Medline]
McFaul PB, Traub AI, Thompson W. Premature luteinization and ovulation induction using human menopausal gonadotrophin or pure follicle stimulating hormone in patients with polycystic ovary syndrome. Acta Eur Fertil (1989) 20:157–61.[Medline]
Miller KF, Behnke EJ, Arciaga RL, et al. The significance of elevated progesterone at the time of administration of human chorionic gonadotropin may be related to luteal support. J Assist Reprod Genet (1996) 13:698–701.[CrossRef][Web of Science][Medline]
Mio Y, Sekijima A, Iwabe T, et al. Subtle rise in serum progesterone during the follicular phase as a predictor of the outcome of in vitro fertilization. Fertil Steril (1992) 58:159–66.[Web of Science][Medline]
Moffitt DV, Queenan JT Jr, Shaw R. Progesterone levels on the day of human chorionic gonadotropin do not predict pregnancy outcome from the transfer of fresh or cryopreserved embryos from the same cohort. Fertil Steril (1997) 67:296–301.[CrossRef][Web of Science][Medline]
Porter RN, Smith W, Craft IL, et al. Induction of ovulation for in-vitro fertilisation using buserelin and gonadotropins. Lancet (1984) 2:1284–5.[Web of Science][Medline]
Prien SD, Canez MS, Messer RH. Increases in progesterone after human chorionic gonadotropin administration may predict cycle outcome in patients undergoing in vitro fertilization–embryo transfer. Fertil Steril (1994) 62:1066–8.[Web of Science][Medline]
Randall GW, Gantt PA, Gantt D, et al. Elevated serum progesterone values at the time of ovulation induction in luteal leuprolide acetate-down-regulated GIFT cycles are associated with decreased clinical pregnancy rates. J Assist Reprod Genet (1996) 13:459–63. 1996.[CrossRef][Web of Science][Medline]
Schoolcraft W, Sinton E, Schlenker T, et al. Lower pregnancy rate with premature luteinization during pituitary suppression with leuprolide acetate. Fertil Steril (1991) 55:563–6.[Web of Science][Medline]
Sekijima A. Relation between the outcome of IVF–ET and subtle rise in serum progesterone during the follicular phase stimulated with a combination of gonadotropin-releasing hormone analog and gonadotropin. Nippon Sanka Fujinka Gakkai Zasshi (1993) 45:249–55.[Medline]
Shulman A, Ghetler Y, Beyth Y, et al. 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.[CrossRef][Web of Science][Medline]
Silverberg KM, Burns WN, Olive DL, et al. 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, et al. 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.[Web of Science][Medline]
Smitz J, Ron-El R, Tarlatzis BC. The use of gonadotrophin releasing hormone agonists for in vitro fertilization and other assisted procreation techniques: experience from three centres. Hum Reprod (1992) 7:49–66.
Ubaldi F, Smitz J, Wisanto A, et al. Oocyte and embryo quality as well as pregnancy rate in intracytoplasmic sperm injection are not affected by high follicular phase serum progesterone. Hum Reprod (1995) 10:3091–6.
Ubaldi F, Camus M, Smitz J, et al. Premature luteinization in in vitro fertilization cycles using gonadotropin-releasing hormone agonist (GnRH-a) and recombinant follicle-stimulating hormone (FSH) and GnRH-a and urinary FSH. Fertil Steril (1996a) 66:275–80.[Web of Science][Medline]
Ubaldi F, Albano C, Peukert M, et al. Subtle progesterone rise after the administration of the gonadotrophin-releasing hormone antagonist cetrorelix in intracytoplasmic sperm injection cycles. Hum Reprod (1996b) 11:1405–7.
Urman B, Alatas C, Aksoy S, et al. Elevated serum progesterone level on the day of human chorionic gonadotropin administration does not adversely affect implantation rates after intracytoplasmic sperm injection and embryo transfer. Fertil Steril (1999) 72:975–9.[CrossRef][Web of Science][Medline]
Yovel I, Yaron Y, Amit A, et al. High progesterone levels adversely affect embryo quality and pregnancy rates in in vitro fertilization and oocyte donation programs. Fertil Steril (1995) 64:128–31.[Web of Science][Medline]
Zimmerman DW. Effect of nonindependence of sample observations on some parametric and nonparametric statistical tests. Commun Stat B-Simul (1993) 22:779–89.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. Kyrou, B. Popovic-Todorovic, H.M. Fatemi, C. Bourgain, P. Haentjens, L. Van Landuyt, and P. Devroey Does the estradiol level on the day of human chorionic gonadotrophin administration have an impact on pregnancy rates in patients treated with rec-FSH/GnRH antagonist? Hum. Reprod., November 1, 2009; 24(11): 2902 - 2909. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Venetis, E. M. Kolibianakis, and B. C. Tarlatzis Progesterone elevation and probability of pregnancy after IVF: facts and fiction Hum. Reprod. Update, September 1, 2008; 14(5): 538 - 538. [Full Text] [PDF]
|
|
|
|
|
|
R. Fleming Progesterone elevation on the day of hCG: methodological issues Hum. Reprod. Update, June 2, 2008; (2008) dmn019v1. [Full Text] [PDF]
|
|
|
|
|
|
D. de Ziegler, G. Bijaoui, and C. Chapron Pre-hCG elevation of plasma progesterone: good, bad or otherwise Hum. Reprod. Update, June 2, 2008; (2008) dmn020v1. [Full Text] [PDF]
|
|
|
|
|
|
C.A. Venetis, E.M. Kolibianakis, and B.C. Tarlatzis Reply: Comment on: Is progesterone elevation on the day of human chorionic gonadotrophin administration associated with the probability of pregnancy in in vitro fertilization? A systematic review and meta-analysis. By Venetis et al (2007) Hum. Reprod. Update, March 1, 2008; 14(2): 195 - 196. [Full Text] [PDF]
|
|
|
|
|
|
E. Bosch Comment on: Is progesterone elevation on the day of human chorionic gonadotrophin administration associated with the probability of pregnancy in in vitro fertilization? A systematic review and meta-analysis. By Venetis et al (2007) Hum. Reprod. Update, March 1, 2008; 14(2): 194 - 195. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||