Hum. Reprod. Advance Access originally published online on July 29, 2008
Human Reproduction 2008 23(11):2421-2426; doi:10.1093/humrep/den289
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Human chorionic gonadotropin levels in early IVF/ICSI pregnancies are higher in singletons after single embryo transfer compared with singletons after double embryo transfer
1 Department of Obstetrics and Gynaecology, Ghent University, De Pintelaan 185 B -9000, Gent, Belgium 2 Department of Applied Mathematics and Computer Science, Ghent University, Krijgslaan 281 S9, B-9000 Gent, Belgium 3 Department of Public Health, Ghent University, De Pintelaan 185, B-9000 Gent, Belgium
4 Correspondence address. E-mail: ilse.delbaere{at}ugent.be
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
BACKGROUND: The aim of this study is to assess differences in human chorionic gonadotropin (hCG) values between singleton IVF/ICSI pregnancies after elective single (eSET) versus double embryo transfer (DET).
METHODS: This is a unicentre, retrospective, non-randomized study of women with eSET or DET in years 2000–2006, giving birth to a singleton child with a birthweight of at least 500 g (N = 790). These women had between 1 and 6 hCG assessments in our centre. Log hCG values from Days 15 to 35 after oocyte pick-up were compared between eSET and DET patients in a multivariate analysis.
RESULTS: We compared log hCG values in 456 eSET patients with log hCG values in 334 DET patients. Log hCG values were significantly higher after eSET than after DET from Day 15 until Day 35 after oocyte pick-up (P < 0.001).
CONCLUSION: The lower hCG values in early pregnancy are possibly due to a delay in implantation when two embryos are transferred and one of them does not implant, but the exact mechanism is unknown.
Key words: hCG/implantation/SET/DET
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
In a recent paper, we observed higher birthweights in singletons after single embryo transfer (SET) compared with singletons after double embryo transfer (DET) (De Sutter et al., 2006
). It was hypothesized that this difference was due to a higher incidence of ‘vanishing twins’ in the DET group. Moreover, the difference in birthweight (120 g) between both groups corresponds to the findings of Pinborg et al. (2005) between ‘real singletons’ and ‘singletons who started as a twin’. In order to confirm or refute this hypothesis, it is of interest to assess early implantation events after both SET and DET.
Human chorionic gonadotropin (hCG) is a glycoprotein, expressed by both the cytotrophoblast and syncytiotrophoblast of the human blastocyst. Since hCG is the first observable blastocyst signal in maternal blood, its detection in serum is used as a marker of pregnancy. hCG is present in the blood from the time of implantation (6–7 days after fertilization), rises exponentially in the first trimester with a peak at 9–12 weeks gestational age, and decreases from that point onwards (Jaffe et al., 1969; Jameson et al., 1993; Handschuh et al., 2007).
The ‘pregnancy hormone’ hCG probably has a function in the differentiation of human endometrial cells into decidua. Furthermore, it prevents regression of the corpus luteum, and supports early pregnancy, since ovarian progesterone secretion is maintained (Handschuh et al., 2007). hCG receptors have also been found in the uterine vasculature and in other tissue, indicating that this hormone has a function at other levels as well (Stenman et al., 2006). Recently, it has been found that hCG administration has an effect on endometrial thickness and as such, may enhance implantation (Tesarik, 2003).
In general, multiple pregnancies show higher mean values of hCG and spontaneous abortions have lower mean values than (ongoing) singleton pregnancies (Liu et al., 1988). The higher values in multiple pregnancies are probably more likely due to a greater number of trophoblastic cells than to increased trophoblastic activity (Liu et al., 1988).
When hCG rises less steeply than in normal twin pregnancies, this is indicative for a ‘vanishing twin’ pregnancy. If this is the case, the slower rate manifests from the beginning of pregnancy (Kelly et al., 1991).
Earlier studies assessing hCG values in relation to outcome, reported on prediction of ongoing pregnancy and ectopic pregnancy with cut-off hCG levels at a given time. These studies found lower hCG values at Day 12–15 after embryo transfer in non-viable pregnancies (Bjercke, 1999; Homan, 2000; Takata et al., 2005; Stone et al., 2006; Lambers et al., 2006).
Since median hCG values are higher in twin pregnancies, we expect higher values in early ongoing twin pregnancies after DET compared with pregnancies after SET. However, what happens at the time of implantation when only one embryo evolves after DET versus SET is not well documented. Given our previous finding of lower birthweights in singletons after DET versus SET, we wanted to assess to what extent the hypothesis of ‘vanishing twins’ in early implantation can explain that finding.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Study population
The Ghent University Hospital Infertility Centre collects data of its activities, since 1992. For the current analysis, we used the same criteria of inclusion as for our previous paper. In the same fashion, we aimed to incorporate a homogenous population by including fresh, first, second or third cycles from in-house patients in this retrospective non-randomized study. Patients undergoing SET or DET in years 2000–2006 and giving birth to a singleton baby of at least 500 g of birthweight were included. From a total of 8728 in-house cycles resulting in 1683 (19.3%) ongoing pregnancies, 1472 were singletons. After exclusion of 34 vanishing twins and 648 women who did not have an hCG-assessment in our centre, 790 women were included in this study. The major part of these underwent SET (N = 456; 57.7%), in the remaining part (N = 334; 42.3%) two embryo's were transferred (DET). We excluded all women with only one fertilized oocyte available, so all SET cycles were in fact elective SET (eSET) cycles. DET cycles which resulted in a twin pregnancy were excluded from this study as well; however, we included mean log hCG values for this group in Fig. 1 for explicative purposes. In total, there were 1966 hCG measurements in our population, 1087 measurements in the SET group and 858 measurements in the DET group. All hCG measurements from 15 days until 35 days after pick-up were included. The number of hCG assessments per pregnancy ranges from one to six, with a mean of two measurements per patient included. All hCG measurements occurred in the same laboratory.
|
Clinical and laboratory protocols
IVF and ICSI were performed according to accepted standard protocols of pituitary suppression, ovarian stimulation, oocyte retrieval, gamete handling, embryo culture techniques, embryo transfer and luteal supplementation, as described in previous studies from our group (Laverge et al., 2001
; De Sutter et al., 2003). All embryo transfers occurred at Day 2 or 3 after oocyte retrieval, this procedure was independent of the number of embryos transferred. Patients younger than 36 years, who entered the program before 1 July 2003, were given the choice between the transfer of one (SET) or two (DET) embryos. Before that date, SET was mainly performed when the transferred embryo was of good quality, which is defined as an embryo with four or five blastomeres on Day 2, seven or more blastomeres on Day 3 and <20% fragmentation and total absence of multinucleated blastomeres at any stage of cleavage (Van Royen et al., 1999). If no high-quality embryo was available, two embryos were transferred. After 1 July 2003, legal prescriptions were followed, meaning that SET was always applied at the first IVF/ICSI attempt in patients younger than 36 years of age. In the second and third cycle, young patients were encouraged to accept SET, but DET was also an option if only embryos of moderate or poor quality were available. Patients aged 36–39 years of age had the option to choose DET for the first and second attempt, and only exceptionally three embryos were transferred from the third cycle on. In patients older than 39 years, no legal limit to the number of transferred embryos was imposed, but most patients received a maximum of two embryos in their first two cycles.
The assessment of hCG for the confirmation of incipient pregnancy is generally planned at Day 16 after pick-up in our centre, but some women do the test earlier. In order to exclude bias from interference with luteal phase agents in patients who received hCG in order to support the luteal phase, we included measurements from Day 15 after pick-up only. Day 35 was established as upper limit for the analyses, since we are interested in the effect of eSET versus DET on early implantation. At that time, patients are in their 7th week of pregnancy and all included women had an ultrasound by then. All included women had an early hCG assessment in order to confirm pregnancy; in a large number of patients (57%), a second hCG assessment occurred later in pregnancy. Since this is a retrospective study, there was no planned interval between two hCG measurements and the number of embryos transferred was not an indication to perform several assessments. Immunometric assay (Roche Diagnostics Gmbh, 68 298 Mannheim, Germany) was used to determine hCG quantitatively in serum. Auto-analyser Modular E170 (T0470) and Cobas 6000 (T0080) were used as equipment. The detection limit is 
0.1 mIU/ml and the measurement range goes up to 10 000 mIU/ml. There is an internal and external quality control of the results. Internal control occurs on a daily basis; which implies that results are tested for conformity with the Westgard rules.
Data collection
From the large number of variables available in our database, we analysed the following for the present study: number of embryos transferred, hCG value, number of foetal sacs visible at the first ultrasound, number of live-born children, number of available oocytes, number of fertilized oocytes, date of hCG assessment, maternal age at time of pick-up, type of applied technique (IVF or ICSI), medication used for support of the luteal phase, type of pathology (female, male, combined or unknown), sex of the child and occurrence of hypertension during pregnancy. Although our transfer policy depends on number of blastomeres, fragmentation and number of multinucleated cells, our database only holds information on fragmentation of the embryo. So we were not able to adjust for embryo quality in a proper manner. Support for the luteal phase occurred with hCG (Pregnyl®, Organon, Oss the Netherlands or oral micronized progesterone), Utrogestan® (Besins, France), both hCG and Utrogestan® or other medication. Although all these pregnancies resulted in a viable pregnancy, we considered it useful to include several hCG assessments in order to enable illustration of the progress.
Statistical analysis
Profile plots with local regression smoother were used to describe the evolution in log hCG over time for eSET and DET patients separately. In a univariate analysis, independent sample t-tests were used to compare continuous patient characteristics between eSET and DET pregnancies (after confirming normality of the data in each group) and Pearson Chi-square tests to compare categorical patient characteristics between these groups. In the multivariate analysis, linear mixed models were used to adjust the comparison of eSET and DET pregnancies for the year in which the procedure occurred, maternal age, parity, moment of hCG assessment, number of oocytes available for fertilization, applied technique (IVF or ICSI), medication used for the luteal phase, underlying pathology of infertility and sex of the child. Specifically, we used linear mixed models with random intercepts and slopes, independent residual errors, and we allowed for nonlinear (fourth order) time effects. All analyses were obtained using SAS 9.1.3 (SAS Institute Inc., Cary, NC, USA), except graphics were obtained using R 2.4 (the R Foundation for statistical computing, ISBN 3–900051–07–0).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Patient characteristics
Patient characteristics are described in Table I. Mean maternal age is higher (P < 0.001) in DET patients. Although no difference in underlying infertility type was found, there is a higher use of IVF in the DET group and ICSI in the eSET group. For the support of the luteal phase, exogeneous hCG or ‘other agent than hCG or Utrogestan®’ were more frequently applied in the DET group, whereas Utrogestan® was more popular in the eSET group. The reason for these differences is not clear, but they were accounted for in the multivariate analysis. On average, hCG assessment occurred on Day 17 or 18. There were on average 2.1 measurements in eSET patients and 1.8 measurements in DET patients (P < 0.001). Because of the large amount of data and the small difference in the timing and number of measurements, we found a significant difference between eSET and DET patients for both of these variables, although this difference is clinically not relevant. The number of aspirated ooctyes ranged from 2 to 45, with a mean of 13 in both groups. This resulted in 8–9 fertilized oocytes (range 2–34). The DET procedure resulted in two foetal sacs in 20% of the cases, but these twin pregnancies were excluded from this study.
|
Univariate analysis
Fig. 1 depicts that overall mean log hCG values are higher in eSET patients compared with DET patients with a singleton as outcome (P < 0.001). Mean log hCG values for both singletons after eSET and DET are lower than in twin pregnancies. Figures were included in order to illustrate the progress of mean log hCG values; however, due to variability, the marked percentages are not very precise. In order to assess exact differences between log hCG values in singleton pregnancy after eSET and DET, we performed more powerful analyses. In eSET patients, log hCG values start off at a higher level and remain significantly increased relative to DET patients during the first 25 days. At Day 15 after pick-up, median log hCG (more precisely, geometric mean hCG) in eSET patients is 37% (P < 0.001) higher compared with DET patients. Twenty days after pick-up, median log hCG levels in eSET patients are 31% (P < 0.001) higher. After 25 and 30 days, median log hCG levels are, respectively, 24% (P < 0.01) and 19% (P < 0.05) higher in eSET patients. Consequently, log hCG levels in eSET patients rise slower than those in DET patients. Since the indication to apply SET/DET was established by means of legal regulations in Belgium during the study, we performed sub-analyses on our data with embryo transfer before 1 July 2003 on the one hand (N = 271 patients) and after this date on the other hand (N = 433). In both sub analyses, equal trends as described above were observed, in that log hCG values of DET patients started at a lower level but rose more rapidly afterwards. In the sub analysis with data on embryo transfers before 1 July 2003, the difference was not significant (P = 0.13).
Multivariate analysis
While significant differences were observed in mean maternal age, parity, type of infertility treatment and supportive medication for the luteal phase between eSET and DET patients, these characteristics were not significantly associated with the hCG evolution in the multivariate analysis. Although both groups had their first hCG assessment on average on Day 17, we found a significant difference (P < 0.05) in a moment of hCG assessment. The difference is, however, small and not problematic for our analysis because linear mixed models compare hCG measurements between eSET and DET patients on the same day of measurement. Only the year in which the procedure occurred was significantly associated with the evolution in hCG. This is probably due to the earlier described legal prescriptions to transfer only one good-quality embryo when criteria are met. As such, before 2003, DET was more common than SET; whereas after 2003 SET was applied more. Notwithstanding these changes, after correction for year of transfer, our results remained qualitatively unchanged. On Day 15 after pick-up, median hCG-values in eSET patients were 30% higher when compared with DET patients (P < 0.01). On Days 20 and 25, median hCG values were, respectively, 24% (P < 0.01) and 18% (P < 0.05) higher in eSET patients. On Day 30, the difference in median hCG values between eSET and DET patients was no longer significant (P = 0.098).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
At time of implantation and during the first week of pregnancy, median hCG values are higher after eSET than after DET resulting in a singleton. Since median hCG values are higher in twin pregnancies (Liu et al., 1988
), as can also be derived from Fig. 1, our analyses suggest that the hypothesis that the lower mean birthweight in singletons after DET is due to a higher incidence of ‘vanishing twins’ in this group may be refuted.
Bjercke (1999) found that the large majority of embryos with a hCG value of 55 IU/l on Day 12 (log hCG 1.7) had a vital pregnancy. Poikkeus (2002) decided that when there is a hCG level above 200 IU/l (or log hCG = 2.3) at Day 12 after transfer (which corresponds with Day 15 after pick-up in our results), miscarriages are rare. In our study, median eSET and DET values were largely above these threshold values, but earlier studies point out that higher hCG values in early pregnancy are preferred. Moreover, since previous studies reported higher mean hCG values in spontaneous pregnancies compared with IVF pregnancies (Tulchinsky, 1996; Ertzeid 2000; Poikkeus, 2002), our findings are once again indicative that singletons after SET compare favourably with spontaneously conceived singletons.
Our results may be indicative that implantation after DET is hampered or delayed in the presence of a second embryo that will not implant. In SET, there may be an all or nothing effect when an embryo with intermediate quality is transferred: when it is capable to implant, the implantation process succeeds in a normal way. The finding that initial log hCG values are lower in DET, but rise more rapidly afterwards, may be indicative for a delayed implantation due to the competition between the two transferred embryos or a negative influence of the non-implanting on the implanting embryo.
Apparently, the presence of a second embryo that does not implant may be interfering with or disturbing the implantation process. This finding connects with the collaborative model of Matorras et al. (2005), which suggest a higher implantation potential of a second embryo after implantation of a first embryo (Matorras et al., 2005). Our results indicate that if an embryo does not implant, the implantation of the other is impaired.
We cannot exclude that cohort-intrinsic factors of patients where only one embryos implant after DET, are responsible for our results. Differences in log hCG values between eSET and DET were no longer significant in our sub-analysis on data of embryo transfer before the IVF reimbursement law. However, the fact that the trend of lower log hCG levels in DET patients remained equal, suggest that this may be due to lack of power. Considering the indications to transfer one or two embryos after 1 July 2003; patient characteristics are more divergent between the eSET and DET population in that period. By definition, patients who received two embryos during transfer after 2003 are older have embryos of lower quality or have had more unsuccessful IVF/ICSI cycles than eSET patients. However, adjustment for maternal age and other differences between SET and DET patients did not modify our results. Because there is only 1 year of difference in mean maternal age between our SET and DET group and since SET is obligatory applied in the first cycle of young women since 2003 (regardless of embryo quality), it is possible that SET patients after 2003 are similar to our DET patients included, but that there are more first cycles in the SET population and more second and third cycles in the DET population. In spite of that, some intrinsic differences between both populations may not have been accounted for in our analysis. It would be particularly important to adjust for embryo quality adequately in future studies on this subject. Furthermore, with the exclusion of twin pregnancies, we cannot completely eliminate some kind of selection bias in our results, in that embryos with maximal implantation potential (twin pregnancies) after DET were excluded. However, the choice not to include twin pregnancies was driven by our intention to assess the ‘vanishing twin’ hypothesis as was described earlier.
It is possible that one embryo deceases at or shortly after double embryo transfer. A deceased embryo will not implant and may be responsible for biochemical reactions in utero which may cause the above described delay for the implantation process of the other embryo.
In conclusion: we found no evidence that lower birthweight in singletons after DET is related to a higher amount of ‘vanishing twins’ during early implantation in this population. Implantation is possibly retarded after DET, since hCG values are lower at onset. Possibly one of both transferred embryos dies during or shortly after transfer, which may induce processes in utero hampering the implantation process of the other embryo. However, studies with more homogeneous study groups and known embryo quality are needed to investigate the matter further.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Petra De Sutter is holder of a fundamental clinical research mandate by the Flemish Foundation for Scientific Research (FWO—Vlaanderen).
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Bjercke S. Human chorionic gonadotrophin concentrations in early pregnancy after in-vitro fertilization. Hum Reprod (1999) 14:1642–1646.
De Sutter P, Van der Elst J, Coetsier T, Dhont M. Single embryo transfer and multiple pregnancy rate reduction in IVF/ICSI: a 5-year appraisal. Reprod Biomed Online (2003) 6:464–469.[Medline]
De Sutter P, Delbaere I, Gerris J, Verstraelen H, Goetgeluk S, Van der Elst J, Temmerman M, Dhont M. Birthweight of singletons after assisted reproduction is higher after single- than after double-embryo transfer. Hum Reprod (2006) 21:2633–2637.
Ertzeid G. Human chorionic gonadotropin levels in successful implantations after assisted reproduction techniques. Gynecol Endocrinol (2000) 14:258–263.[Web of Science][Medline]
Handschuh K, Guibourdenche J, Tsatsaris V, Guesnon M, Laurendeau I, Evain-Brion D, Fournier T. Human chorionic gonadotropin expression in human trophoblasts from early placenta: comparative study between villous and extravillous trophoblastic cells. Placenta (2007) 28:175–184.[CrossRef][Medline]
Homan G. Human chorionic gonadotropin as a predictor of outcome in assisted reproductive technology pregnancies. Fertil Steril (2000) 73:270–274.[CrossRef][Web of Science][Medline]
Jaffe R, Lee P, Midgley JA. Serum gonadotropins before, at the inception of, and following human pregnancy. J Clin Endocrinol Metab (1969) 29:1281–1283.
Jameson J, Hollenberg A. Regulation of chorionic gonadotropin gene expression. Endocr Rev (1993) 14:203–221.
Kelly MP, Molo MW, Maclin VM, Binor Z, Rawlins RG, Radwanska E. Human chorionic gonadotropin rise in normal and vanishing twin pregnancies. Fertil Steril (1991) 56:221–224.[Web of Science][Medline]
Lambers MJ, van Weering HGI, van ‘t Grunewold MS, Lambalk CB, Homburg R, Schats R, Hompes PGA. Optimizing hCG cut - off values: a single determination on day 14 or 15 is sufficient for a reliable prediction of pregnancy outcome. Eur J Obstet Gynecol Reprod Biol (2006) 127:94–98.[CrossRef][Web of Science][Medline]
Laverge H, De Sutter P, Van der Elst J, Dhont M. A prospective, randomized study comparing day 2 and day 3 embryo transfer in human IVF. Hum Reprod (2001) 16:476–480.
Liu HC, Kreiner D, Muasher SJ, Jones G, Jones H Jr, Rosenwaks Z. Beta-human chorionic gonadotropin as a monitor of pregnancy outcome in in vitro fertilization-embryo transfer patients. Fertil Steril (1988) 50:89–94.[Web of Science][Medline]
Matorras R, Matorras F, Mendoza R, Rodriguez M, Remohi J, Rodriguez-Escudero FJ, Simon C. The implantation of every embryo facilitates the chances of the remaining embryos to implant in an IVF programme: a mathematical model to predict pregnancy and multiple pregnancy rates. Hum Reprod (2005) 20:2923–2931.
Pinborg A, Lidegaard O, Cour Freiesleben N, Andersen AN. Consequences of vanishing twins in IVF/ICSI pregnancies. Hum Reprod (2005) 20:2821–2829.
Poikkeus P. Serum HCG 12 days after embryo transfer in predicting pregnancy outcome. Hum Reprod (2002) 17:1901–1905.
Stenman UH, Tiitinen A, Alfthan H, Valmu L. The classification, functions and clinical use of different isoforms of hCG. Hum Reprod Update (2006) 12:769–784.
Stone BA, Vargyas JM, Ringler GE, March CM, Marrs RP. The rate at which serum total [beta]-subunit human chorionic gonadotropin increases after embryo transfer is a predictor of the viability of pregnancy and an identifier of determinants of pregnancy. Fertil Steril (2006) 86:1626–1633.[CrossRef][Web of Science][Medline]
Takata K, Suganuma N, Wada H, Takeuchi K, Kitamura K, Kanayama N. Human chorionic gonadotropin in cervical fluid as a predictor of miscarriage. Fertil Steril (2005) 84:687–691.[CrossRef][Web of Science][Medline]
Tesarik J. Luteinizing hormone affects uterine receptivity independently of ovarian function. Reprod Biomed Online (2003) 7:59–64.[Medline]
Tulchinsky D. Delayed embryo implantation following in vitro fertilization and embryo transfer (IVFET). J Assist Reprod Genet (1996) 13:536–539.[CrossRef][Web of Science][Medline]
Van Royen E, Mangelschots K, De Neubourg D, Valkenburg M, Van de Meerssche M, Ryckaert G, Eestermans W, Gerris J. Characterization of a top quality embryo, a step towards single-embryo transfer. Hum Reprod (1999) 14:2345–2349.
Submitted on December 18, 2007; resubmitted on June 10, 2008; accepted on June 12, 2008.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||