Hum. Reprod. Advance Access published online on December 3, 2007
Human Reproduction, doi:10.1093/humrep/dem305
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Urinary hMG versus recombinant FSH for controlled ovarian hyperstimulation following an agonist long down-regulation protocol in IVF or ICSI treatment: a systematic review and meta-analysis
1 Assisted Conception Unit, 4th Floor, Thomas Guy House, Guys Hospital, Thomas Street, London SE1 9RT, UK 2 Department of Public Health and Epidemiology, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK 3 Birmingham Women’s Hospital, Metchley Park Road, Edgbaston Birmingham, UK 4 Center for Reproductive Medicine, Obstetrics and Gynaecology, H4-205, Academic Medical Center, University of Amsterdam, PO Box 22660 1100 DD, Amsterdam, The Netherlands 5 Department for Clinical Epidemiology and Biostatistics, J1b-226 Academic Medical Center, University of Amsterdam, PO Box 22660 1100 DD, Amsterdam, The Netherlands
6 Correspondence address. Tel: +44-207-188-0496; Fax: +44-207-188-0490; E-mail: arricoomar{at}blueyonder.co.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
BACKGROUND: Since the most recent Cochrane review on hMG versus rFSH for controlled ovarian hyperstimulation following a long down-regulation protocol, several new trials have emerged.
METHODS: We conducted a systematic review and meta-analysis of randomized trials comparing the effectiveness of hMG versus rFSH following a long down-regulation protocol in IVF–ICSI cycles, on the primary outcome of live birth per woman randomized, as well as several other secondary outcomes. Searches were conducted in MEDLINE, EMBASE, Science Direct, Cochrane Library and databases of abstracts (last search January 2007).
RESULTS: Seven randomized trials, consisting of a total of 2159 randomized women, were identified. A meta-analysis of these trials showed a significant increase in live birth rate with hMG when compared with rFSH (relative risk, RR = 1.18, 95% CI: 1.02–1.38, P = 0.03). The heterogeneity test was non-significant (P = 0.97), suggesting that there was no statistical inconsistency between the seven studies. The pooled risk difference (RD) for the outcome of live birth rate was 4% (95% CI: 1–7%) for these study populations. There was an increase in clinical pregnancy rates with hMG when compared with rFSH (RR = 1.17, 95% CI 1.03–1.34). No significant differences were noted for gonadotrophin use, spontaneous abortion, multiple pregnancy, cancellation and ovarian hyperstimulation syndrome rates.
CONCLUSIONS: For the populations in the randomized trials, hMG was associated with a pooled 4% increase in live birth rate when compared with rFSH in IVF–ICSI treatment following a long down-regulation protocol.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Various preparations containing FSH can be used for controlled ovarian hyperstimulation in IVF or ICSI. The two commonly used forms are urinary hMG (which contains FSH and LH activity) and recombinant FSH (which contains just FSH) (van Wely et al., 2003
). Several studies have evaluated the effectiveness of these two alternatives in IVF or ICSI following a long down-regulation protocol, which is currently the most widely used protocol. In the most recent Cochrane review on this issue, we identified four randomized trials that reported the clinically relevant outcomes of ongoing pregnancies or live births (van Wely et al., 2003). A meta-analysis of these four trials showed a statistically non-significant increase in ongoing pregnancies or live births with hMG compared with rFSH (OR = 1.27, 95% CI: 0.98–1.64), resulting in our call for ‘large randomized trials to estimate the difference between hMG and rFSH more precisely' (van Wely et al., 2003). Since the publication of our Cochrane review, several new randomized trials have become available. In this systematic review, we evaluated the effectiveness of hMG versus rFSH in women undergoing IVF or ICSI following a long GnRHa protocol with live birth rate as the primary outcome.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Identification of literature
The following electronic databases were searched: MEDLINE (1985 to January 2007), EMBASE (1985 to January 2007), Science Direct (1985 to January 2007), Cochrane Central Register of Controlled Trials (CENTRAL) and Web of Science (1985 to January 2007). A search strategy was carried out based on the following terms: gonadotrophins, hMG, menotropins, FSH, recombinant FSH and IVF. In addition, references from all identified articles were checked, and a hand search of the abstracts from the annual meetings of the American Society for Reproductive Medicine and the European Society for Human Reproduction and Embryology was performed (1991–2006). If necessary, additional information was sought from the authors. The search was not restricted by language. The searches were conducted independently by M.van W. and A.C.
Study selection and data extraction
Studies were selected if the target population was infertile couples with any cause other than polycystic ovarian syndrome (PCOS); the therapeutic interventions were hMG or rFSH for ovarian stimulation in IVF or ICSI treatment, following a long down-regulation protocol and the studies were of randomized design. Only trials using hMG that typically contained FSH and LH activity at a 1:1 ratio were included. Trials in which LH or clomiphene citrate were concurrently administered were excluded. The primary outcome measure of interest was live birth rate per randomized woman, as this is the most clinically relevant outcome for the infertile couples. We also reported on secondary outcome measures such as clinical pregnancy rates, the amount of gonadotrophin use, spontaneous abortion, multiple pregnancy, cancellation and ovarian hyperstimulation syndrome (OHSS) rates.
Studies were selected in a two-stage process. First, the titles and abstracts from the electronic searches were scrutinized by two reviewers independently (A.C. and M. van W.) and full manuscripts of all citations that were likely to meet the predefined selection criteria were obtained. Secondly, final inclusion or exclusion decisions were made on examination of the full manuscripts. In cases of duplicate publication, the most recent and complete versions were selected. Any disagreements about inclusion were resolved by consensus or arbitration by a third reviewer.
The selected studies were assessed for methodological quality by using the components of study design that are related to internal validity (Juni et al., 2001). Information on the adequacy of randomization, concealment and blinding was extracted. From each study, outcome data were extracted in 2 x 2 tables. Data extraction was performed in duplicate by M. van W. and A.C.
Statistical analysis
Relative risks and risk differences (RDs) from individual studies were meta-analysed using fixed effects model (Mantel and Haenszel, 1959) if significant heterogeneity could be excluded. We also performed meta-analysis using random effects models as a sensitivity analysis to examine the robustness of our findings to alternative methods of pooling (DerSimonian and Laird, 1986). Heterogeneity of treatment effects was evaluated graphically using forest plot (Lewis and Clarke, 2001) and statistically using the Breslow and Day chi-square test. Exploration of the causes of heterogeneity, if present, was planned using variation in features of the population, intervention and study quality. Meta-regression analysis was employed for assessing differential effects. To assess for publication and related biases, we performed a funnel plot analysis, using Egger’s test to evaluate asymmetry (Egger et al., 1997).
All statistical analyses were performed using Stata 8.0 (TX, USA) and RevMan 4.2 (Cochrane Collaboration, Oxford, UK).
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
The search strategy yielded 693 publications (689 from electronic searchers; one from hand search of Human Reproduction and Fertility and Sterility; and three from reference lists of relevant publications). Six hundred and fifty publications were excluded as it was clear from the title that they did not fulfil the selection criteria. From the remaining 43 articles, 20 were excluded on the basis of the abstract. For the remaining 23 articles, we obtained the full manuscripts, and following scrutiny of these, we identified 15 potentially relevant studies. From these 15 trials, eight were excluded for the following reasons: two studies were excluded as pregnancy outcome was not studied (Duijkers et al., 1996
; Teissier et al., 1999); one trial was excluded as GnRH analogue down-regulation was not used (Jansen et al., 1998) and another trial was excluded because a short protocol was used (Strehler et al., 2001); one trial was excluded because the number of women in each treatment group could not be determined, even after our attempts to get additional information from the authors (Kornilov et al., 1999); another was excluded as the hMG contained 25 IU LH instead of 75 IU LH (Drakakis et al., 2002) and two further trials were excluded as they used pseudo-randomization processes (Serhal et al., 2000; Duijkers, 1995). Therefore, the total number of trials included in the review is seven. The quality and the main characteristics of the seven trials are presented in Table I. Of the seven trials, six had adequate concealment of allocation. The studies were generally small and underpowered for the clinically relevant outcome of live births, with sample sizes varying from 40 to 781. Both IVF and ICSI were performed in 2/7, IVF alone in 2/7 and ICSI alone in 3/7 of the trials. Although the exact inclusion criteria varied between the trials, most had broadly similar inclusion criteria that included couples with tubal, male factor or unexplained infertility. Three studies excluded women over the age of 37, another three excluded women over the age of 39 and one excluded women over the age of 40 years. Four studies excluded women with a BMI > 27 kg/m2. In all trials, women with PCOS were excluded. The hMG preparations were Menogon, Pergonal, Humegon and Menopur at variable doses, although the commonly used dosage was 225 IU (Table I). The rFSH preparations were Gonal-F (Follitropin alpha) and Puregon (Follitropin beta), again at variable doses, but commonly used dosage being 225 IU.
|
Pooling of results from the seven studies showed a statistically significant 18% relative increase in live birth rate with hMG compared with rFSH (RR = 1.18, 95% CI: 1.02–1.38, P = 0.03, Fig. 1). The pooled RD was 4% (95% CI: 1–7%) for the review study populations. This finding remained unaltered regardless of the statistical method for pooling, with fixed (Mantel and Haenszel, 1959
) or random (DerSimonian and Laird, 1986) effect models. The heterogeneity test was non-significant (P = 0.97), indicating that there was no statistical inconsistency in the studies on the finding of benefit towards hMG compared with rFSH. We performed a sensitivity analysis in which we added the only quasi-randomized study (Duijkers, 1995) that reported on live birth outcome to the meta-analysis: the finding of benefit for hMG when compared to rFSH remained unaltered (RR = 1.19, 95% CI 1.02–1.38). To explore for possible differential effects of degree of purification, we first separated the studies that used highly purified hMG (hp-hMG) (Diedrich et al., 2002; Kilani et al., 2003; Andersen et al., 2006) from studies that used standard hMG (Gordon et al., 2001; Ng et al., 2001; Westergaard et al., 2001; Balasch et al., 2003). We then performed a test of interaction to assess for differential effects. The P-value for interaction showed that differential effects were unlikely (P = 0.77). However, given the small number of studies, this analysis should be interpreted with caution. Although funnel plot analysis was again limited due to the small number of studies, it indicated that publication and related biases were unlikely (Egger’s test, P = 0.84).
|
Meta-analysis for the secondary outcome of clinical pregnancy rates showed a statistically significant 17% relative increase in clinical pregnancy rates with hMG compared with rFSH (RR = 1.17, 95% CI: 1.03–1.34, Table II); the test for statistical heterogeneity again showed a high level lack of heterogeneity (P = 0.99). There were no statistically significant differences between hMG and rFSH for the other secondary outcomes (Table II).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Our systematic review of randomized trials, representing the current best evidence on the comparison hMG versus rFSH for controlled ovarian hyperstimulation following an agonist long down-regulation protocol in IVF or ICSI treatment, showed a significant 4% increase in live birth rate with hMG for the study populations. Although this difference is statistically significant, the clinical significance of this difference is a matter of judgement for couples undergoing IVF or ICSI treatment, and the clinicians who look after them.
None of the seven trials individually showed a statistically significant benefit towards hMG, although five of the seven trials including the three largest trials, showed a trend favouring hMG (Fig. 1). The lack of statistical significance of the individual trials has resulted in several authors claiming a negative result. Should the statistically non-significant difference in pregnancy rates between hMG and rFSH in individual trials be interpreted as a 'negative' finding?—i.e. has a clinically important difference in pregnancy rates between hMG and rFSH been ruled out? The answer to this question depends on examining the precision [defined as the statistical (un)certainty] around of the results, and examining the findings of the totality of the existing literature (Guyatt and Rennie, 2002). Sample size calculations show that for a study to have 80% power (thus accepting a still high level of a 'false negative' result of 20%) to detect a difference of 5% (22% versus 27%, Andersen et al., 2006) in ongoing pregnancies (or live births), an individual trial will need to randomize over 2400 women. Each of the seven trials in this review had far fewer participants than this. Thus, the interpretation of individual studies comparing hMG with rFSH is likely to suffer with the possibility of a false negative finding (type II error). One approach to minimize the risk of a false negative result is to improve the power by pooling (meta-analysing) all the randomized trials on this subject (Guyatt and Rennie, 2002), which is what we have done in this review.
Although one aims to include in a meta-analysis trials that are comparable in patient characteristics and methods, there will always be some clinical heterogeneity between the trials included in a meta-analysis. However, a degree of clinical heterogeneity is not necessarily a drawback. It can be an advantage, as it increases the generalizability of a meta-analysis over that of single studies. Yet, on the other hand, between-trial heterogeneity may generate misleading results by ignoring meaningful heterogeneity among studies. In our meta-analysis, there was clinical heterogeneity between the included trials in type of hMG (highly purified or not), the exact protocol used for down-regulation, proportion of IVF versus ICSI and in baseline characteristics such as the type of infertility and age. Although we were able to examine the differential effect of the type of hMG through a study-level analysis using interaction tests, the small number of studies would make multiple interaction testing less meaningful, and therefore, we avoided this. We hope to examine the effect of various characteristics on the outcome through a more robust approach using individual patient data meta-analysis (Simmonds et al., 2005).
A systematic review performed to evaluate whether endogenous LH levels predicted the likelihood of ongoing pregnancy beyond 12 weeks in women undergoing ovarian stimulation for IVF found that low endogenous LH levels were not associated with a significantly decreased probability of ongoing pregnancy, and high endogenous LH levels during down-regulation were, in fact, associated with a decreased probability of ongoing pregnancy (Kolibianakis et al., 2006). This indirect evidence, therefore, does not support the use of hMG which differs from rFSH in containing LH activity. However, a recent Cochrane review that compared the effectiveness of a combination of recombinant LH (rLH) and recombinant FSH with recombinant FSH alone in controlled ovarian hyperstimulation protocols in IVF or ICSI did suggest that poor responders or older women with a higher risk of abortions, may benefit from co-treatment with rLH due to a decrease in early pregnancy-losses (Mochtar et al., 2007).
One important limitation of the existing trials of hMG versus rFSH is that they have not reported on the outcomes for the frozen-thawed cycles in the two groups, and it is the cumulative live birth rate that is the most relevant outcome measure. However, the evidence on the number of spare embryos available (Diedrich et al., 2002) and the data from a retrospective study (Seelig et al., 2002) on the outcome of frozen embryo cycles did not indicate superiority of either hMG or rFSH in frozen-thawed cycles.
Despite the clinical effectiveness and the reduced cost of hMG, a few more issues deserve addressing. First, clinicians and couples may still prefer to use rFSH for reasons such as elimination of the risk of viral or prion infections, although these infections have never been demonstrated, and therefore remain only as a theoretical risk. Secondly, rFSH represents a potentially unlimited source of FSH. Thirdly, further research into rFSH holds the promise of longer-acting forms (with the result of reduced numbers of injections) or even orally active versions of FSH. Finally, if the clinical superiority of hMG is due to the LH it contains, then it may be possible to add recombinant LH or even hCG to rFSH to achieve the same results—However, these possibilities need to be confirmed in trials before such a step is considered for clinical practice.
|
Conclusion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
The randomized trials summarized in this meta-analysis showed a significant increase of 4% in live birth rate with the use of hMG when compared with rFSH following a long down-regulation protocol in IVF–ICSI treatment cycles. Although the observed difference was a 4% increase in live birth rate with hMG when compared with rFSH, the lower limit of the confidence interval was 1%, and thus the finding of this meta-analysis should not be over-interpreted.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
We would like to acknowledge Professor Balasch for providing data on live births. Furthermore, we wish to acknowledge Dr Nyboe Andersen and Dr Arce for elaborating on their data before the MERIT Trial was published.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Andersen AN, Devroey P, Arce JC, for the MERIT Group. Clinical outcome following stimulation with highly purified hMG or recombinant FSH in patients undergoing IVF: a randomized assessor-blind controlled trial. Hum Reprod (2006) 21:3217–3227.
Balasch J, Penarrubia J, Fabregues F, Vidal E, Casamitjana R, Manau D, Carmona F, Creus M, Vanrell JA. Ovarian responses to recombinant FSH or HMG in normogonadotrophic women following pituitary desensitization by a depot GnRH agonist for assisted reproduction. Reprod Biomed Online (2003) 7:35–42.[Medline]
DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials (1986) 7:177–188.[CrossRef][Web of Science][Medline]
Diedrich, et al, European Israeli Study Group on Highly Purified Menotropin versus Recombinant Follicle-Stimulating Hormone. Efficacy and safety of highly purified menotropin versus recombinant follicle-stimulating hormone in in vitro fertilization/intracytoplasmic sperm injection cycles: a randomized, comparative trial. Fertil Steril (2002) 78:520–528.[CrossRef][Web of Science][Medline]
Drakakis P, Loutradis D, Kallianidis K, Milingos S, Dionyssiou-Asteriou A, Michalas S. The clinical efficacy of recombinant FSH (r-FSH) as compared to highly purified urinary gonadotrophin (hMG-FD) and the use of a low starting dose of r-FSH in IVF or ICSI. Ital J Gynaecol Obstet (2002) 14:64–68.
Duijkers IJ. Chapter 7. A comparison of recombinant human follicle-stimulating hormone (Puregon) and human menopausal gonadotrophin in in vitro fertilitzation cycles. In: Gonadotrophins, Some Basic and Clinical Aspects (1995) Thesis, University of Nijmegen. 115–129.
Duijkers IJ, Willemsen WN, Hollanders HM, Hamilton CJ, Thomas CM, Vemer HM. Follicular fluid hormone concentrations after ovarian stimulation using gonadotrophin preparations with different FSH/LH ratios. II. Comparison of hMG and recombinant FSH. Int J Fertil Womens Med (1996) 42:431–435.
Egger M, Davey SG, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ (1997) 315:629–634.
Gordon UD, Harrison RF, Fawzy M, Hennelly B, Gordon AC. A randomized prospective assessor-blind evaluation of luteinizing hormone dosage and in vitro fertilization outcome. Fertil Steril (2001) 75:324–331.[CrossRef][Web of Science][Medline]
Guyatt G, Rennie D. Users’ Guides to the Medical Literature: A Manual for Evidence-Based Clinical Practice (2002) Chicago: American Medical Association.
Jansen CA, van Os HC, Out HJ, Coelingh Bennink HJ. A prospective randomized clinical trial comparing recombinant follicle stimulating hormone (Puregon) and human menopausal gonadotrophins (Humegon) in non-down-regulated in-vitro fertilization patients. Hum Reprod (1998) 13:2995–2999.
Juni P, Altman DG, Egger M. Systematic reviews in health care: assessing the quality of controlled clinical trials. BMJ (2001) 323:42–46.
Kilani Z, Dakkak A, Ghunaim S, Cognigni GE, Tabarelli C, Parmegiani L, Filicori M. A prospective, randomized, controlled trial comparing highly purified hMG with recombinant FSH in women undergoing ICSI: ovarian response and clinical outcomes. Hum Reprod (2003) 18:1194–1199.
Kolibianakis EM, Collins J, Tarlatzis B, Papanikolaou E, Devroey P. Are endogenous LH levels during ovarian stimulation for IVF using GnRH analogues associated with the probability of ongoing pregnancy? A systematic review. Hum Reprod Update (2006) 12:3–12.
Kornilov NV, Shlykova SA, Loginova JA, Tomas C, Ashorn RG. Comparison of four different gonadotropins for ovarian stimulation in IVF treatment. (1999) 11th World Congress of In Vitro Fertilization & Human Reproductive Genetics: Monduzzi, Bologna, Italy. 379–383.
Lewis S, Clarke M. Forest plots: trying to see the wood and the trees. BMJ (2001) 322:1479–1480.
Mantel N, Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst (1959) 22:719–748.[Web of Science][Medline]
Mochtar MMH, van Wely M, van der Veen F. Recombinant luteinizing hormone (rLH) for controlled ovarian hyperstimulation in assisted reproductive cycles. Cochrane Database Syst Rev (2007) 2007:1.
Ng EH, Lau EY, Yeung WS, Ho PC. HMG is as good as recombinant human FSH in terms of oocyte and embryo quality: a prospective randomized trial. Hum Reprod (2001) 16:319–325.
Seelig AS, Al Hasani S, Katalinic A, Schöpper B, Sturm R, Diedrich K, Ludwig M. Comparison of cryopreservation outcome with gonadotropin-releasing hormone agonists or antagonists in the collecting cycle. Fertil Steril (2002) 77:472–475.[CrossRef][Web of Science][Medline]
Serhal P, Phopong P, Ranieri DM. Comparison between human menopausal gonadotrophin and recombinant FSH for ovarian stimulation in patients undergoing in-vitro fertilization. Hum Reprod (2000) 15:143.
Simmonds MC, Higgins JP, Stewart LA, Tierney JF, Clarke MJ, Thompson SG. Meta-analysis of individual patient data from randomized trials: a review of methods used in practice. Clin Trials (2005) 2:209–217.
Strehler E, Abt M, El Danasouri I, De Santo M, Sterzik K. Impact of recombinant follicle-stimulating hormone and human menopausal gonadotropins on in vitro fertilization outcome. Fertil Steril (2001) 75:332–336.[CrossRef][Web of Science][Medline]
Teissier MP, Chable H, Paulhac S, Aubard Y. Recombinant human follicle stimulating hormone versus human menopausal gonadotrophin induction: effects in mature follicle endocrinology. Hum Reprod (1999) 14:2236–2241.
van Wely M, Westergaard LG, Bossuyt PM, Van der Veen F. Human menopausal gonadotropin versus recombinant follicle stimulation hormone for ovarian stimulation in assisted reproductive cycles. Cochrane Database Syst Rev (2003) CD003973.
Westergaard LG, Erb K, Laursen SB, Rex S, Rasmussen PE. Human menopausal gonadotropin versus recombinant follicle-stimulating hormone in normogonadotropic women down-regulated with a gonadotropin-releasing hormone agonist who were undergoing in vitro fertilization and intracytoplasmic sperm injection: a prospective randomized study. Fertil Steril (2001) 76:543–549.[CrossRef][Web of Science][Medline]
Submitted on July 11, 2006; resubmitted on July 12, 2007; accepted on August 28, 2007.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  C. Hayden GnRH analogues: applications in assisted reproductive techniques Eur. J. Endocrinol., December 1, 2008; 159(suppl_1): S17 - S25. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Bosch, C. Vidal, E. Labarta, C. Simon, J. Remohi, and A. Pellicer Highly purified hMG versus recombinant FSH in ovarian hyperstimulation with GnRH antagonists--a randomized study Hum. Reprod., October 1, 2008; 23(10): 2346 - 2351. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Pinborg, A. Loft, and A. N. Andersen Acupuncture with in vitro fertilisation BMJ, March 8, 2008; 336(7643): 517 - 518. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||