Hum. Reprod. Advance Access originally published online on February 24, 2006
Human Reproduction 2006 21(5):1305-1308; doi:10.1093/humrep/dei492
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Maternal serum inhibin levels in twin and singleton pregnancies conceived by assisted reproduction
1 Department of Obstetrics and Gynecology, Assaf Harofe Medical Center, Zerifin and Sackler Faculty of Medicine, Tel Aviv University, Israel and 2 Reproductive Epidemiology, School of Medicine, University of Leeds, UK.
3 To whom correspondence should be addressed at: Department of Obstetrics and Gynecology, Assaf Harofe Medical Center, Zerifin 70300, Israel. E-mail: intposgr{at}post.tau.ac.il
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
BACKGROUND: To investigate whether second trimester serum inhibin levels differ in pregnancies conceived by assisted reproduction technology (ART). METHODS: In Israel, serum samples from twin pregnancies were obtained for inhibin testing from women either referred for routine ultrasound monitoring, follow up after multi-fetal reduction or amniocentesis, largely for advanced maternal age. In the UK, inhibin had been tested prospectively in singleton and twin pregnancies of women having routine Down’s syndrome (DS) screening. Results were available from 207 ART pregnancies: 170 singletons and 37 twins. This includes 15 twins from Israel, known to have been reduced from triplets to twins. Comparison was made with 4384 spontaneous pregnancies: 4334 singletons and 50 twins. Results were expressed in multiples of the gestation-specific median (MoM) for normal spontaneous pregnancies. RESULTS: In ART singletons, the median maternal inhibin level was higher (1.11 MoM) than in spontaneous singletons (0.99 MoM, P < 0.001, two-tail Wilcoxon Rank Sum Test). In twins, there was no material difference between ART and spontaneous pregnancies with medians of 1.98 and 2.18 MoM, respectively (P = 0.62). There was no effect of multi-fetal reduction, with medians of 1.76 and 1.81 MoM in reduced and non-reduced twins, respectively (P = 0.46). CONCLUSION: It appears that serum inhibin levels are increased on average in ART singletons but not in ART twin pregnancies. More data will be needed before deciding whether risk calculation parameters need to be altered when using inhibin for DS screening in pregnancy.
Key words: antenatal screening/assisted reproduction technology/Down’s syndrome/inhibin/twins
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
In obstetrics, the most common application of maternal serum inhibin A measurement is in Down’s syndrome (DS) screening. In the second trimester, inhibin levels are increased about two-fold, on average, in DS pregnancies, and it is therefore beneficial to add this analyte to screening protocols currently based on
-fetoprotein (AFP) and HCG or free 
-HCG, with or without unconjugated estriol (uE3). Statistical modelling predicts that the addition of inhibin to a three marker combination will increase the detection rate by 6–8%, depending on the cut-off adopted (Cuckle et al., 2005
).
The interpretation of DS screening tests in pregnancies conceived by assisted reproduction technology (ART) is more difficult than in spontaneous pregnancies, as the distributions of marker levels may be altered. There is clear evidence that first and second trimester serum HCG and free -HCG levels are increased on average in ART (Maymon and Shulman, 2001), by 12% in the combined results from 16 series (Cuckle and Arbuzova, 2004). However, there is only one published series of maternal serum inhibin levels in ART where the median in 39 singleton pregnancies conceived by IVF was not significantly different from 195 controls (Wald et al., 1999).
A further complication with ART is that a large proportion of pregnancies are twins, and in countries where the transfer of more than two embryos is common, many of these have been reduced from a high-order multiple gestation. It is well established that all the serum markers used in DS screening are increased about two-fold, on average, in twins, although inhibin has only been investigated in one series (Watt et al., 1996). The interpretation of screening tests in twins is similar to singletons, in that a DS risk is estimated from the maternal age and overlapping marker distributions. For DS twin pregnancies, the average level although increased is not doubled and can be calculated from the averages in DS and unaffected singletons (Cuckle, 1998).
Because there is little published data on inhibin levels in ART singletons and none in ART twins, DS screening algorithms assume that the distributions of levels are the same as in spontaneous pregnancies. It is therefore of clinical and scientific interest to determine whether there is any evidence that inhibin levels are altered in ART pregnancies and in particular when there has been fetal reduction. The current study aims to address this issue.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
In Assaf Harofe Medical Center, Israel, serum samples from twin pregnancies were obtained for inhibin testing from women referred for routine ultrasound monitoring, follow-up after fetal reduction (which was conducted at 12–13 weeks gestation) or amniocentesis (mainly for advanced maternal age). Cases where the indication for amniocentesis was abnormal, maternal serum marker levels were not included. The gestational range was 15–22 weeks gestation. In Leeds, UK, singleton and twin pregnancies were from a complete series of women tested for inhibin as part of routine antenatal screening for DS, without exclusions. The screening protocol for those presenting at 13–19 weeks gestation included inhibin in addition to AFP, free
-HCG and uE3, with or without nuchal translucency (NT). In Israel, detailed information on ART was extracted from the obstetric history taken at the time of referral, and in Leeds, it was from the DS screening request form. Information on fetal reduction was only available for the Israel data.
Following counselling and subject to patient’s consent, blood samples were obtained for research purposes from all women attending the Assaf Harofe Medical Center. Serum was stored at –70°C. Aliquots were retrieved from storage and tested for inhibin at the same laboratory as the routine screening samples from Leeds and in the same analytical batches as routine screening samples. Inhibin A was measured, without knowledge of which were cases and controls, using enzyme immunoassay (Oxford Bio-Innovation Ltd, Oxford, UK).
All results were expressed in multiples of the gestation-specific median (MoM) for non-DS spontaneous pregnancies of the same gestation and corrected for maternal weight. The normal gestation-specific medians were based on routine screening results in Leeds and calculated from a regression equation. The weight correction equation was also based on a regression equation derived from the Leeds screening results. Gestational age was estimated from ultrasound biometry except for pregnancies in Leeds resulting from embryo transfer where the date of implantation was used to estimate the duration of pregnancy.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Inhibin results were available from a total of 4591 pregnancies: 207 ART, 4384 spontaneous, 87 twin and 4504 singleton (Table I). In Assaf Harofe Medical Center, the 21 ART twin pregnancies comprised 17 conceived by IVF and four had controlled ovarian stimulation alone. In Leeds, UK, the 170 ART singleton pregnancies comprised 111 conceived by IVF, including seven from donor oocytes and two by gamete intra-Fallopian transfer (GIFT), 39 after hormone therapy alone, and in 20, the treatment was unspecified. The 16 Leeds ART twins comprised 15 conceived by IVF, including one oocyte donation and one GIFT and one by controlled ovarian stimulation. The 15 ART twins who had been reduced from triplets included one where the procedure was carried out because of anencephaly. No other pregnancies were associated with congenital malformations or chromosomal disorders.
|
The median maternal serum inhibin level was higher in ART singletons (1.11 MoM) than in spontaneous singletons (0.99 MoM) a highly statistically significant difference (P < 0.001, two-tail Wilcoxon Rank Sum Test). The difference was significant for the IVF cases separately (P < 0.005) and the medians were similar in the IVF, hormone and ‘other’ subgroups: 1.10, 1.14 and 1.11 MoM, respectively. The median inhibin level in ART singletons was unaltered after excluding the 20 Leeds cases with unspecified treatment.
The median inhibin levels in ART and spontaneous twins were 1.98 and 2.18 MoM, respectively, which is not statistically significant (P = 0.62, two-tail Wilcoxon Rank Sum Test). The twin medians in Leeds were higher than for the Israel twins, but there was no effect of ART in either series: 2.25 and 2.23 MoM in Leeds, 1.76 and 1.77 MoM in Israel. Among the Israel ART twins, there was no effect of multi-fetal reduction, with medians of 1.76 and 1.81 MoM in reduced and non-reduced twins, respectively (P = 0.46).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
We found that maternal serum inhibin levels were increased on average in singleton ART pregnancies. The only other study included 39 pregnancies and reported a median of 0.89 MoM (Wald et al., 1999
). The 95% confidence intervals (CI) of this study and Wald et al. (1999) overlap: 1.00–1.23 MoM and 0.74–1.08 MoM, respectively. In these circumstances, the best estimate of the average inhibin level is obtained by taking the geometric average of the two results weighed by the number of women studied. This yields a value of 1.07 MoM with 95% CI 0.97–1.17 MoM.
A finding of increased inhibin in ART is consistent with the observation of increased maternal serum HCG species in such pregnancies. Both are placental markers, and levels are correlated within unaffected and within DS pregnancies. Various patho-physiologic explanations have been put forward for the raised HCG in ART including extra-placental production linked with the greater number of corpus luteum and undiagnosed multiple early implantations (Marrs et al., 1979; Ribbert et al., 1996). Exogenous hormones administered during the process of ART has also been proposed, but it is unlikely that most of these proteins will remain in the maternal circulation long enough to influence mid-trimester screening tests (Marrs et al., 1979; Ribbert et al., 1996). Moreover, although inhibin and HCG are correlated, there is no reason to believe that the mechanism by which ART or infertility itself influences hormone levels is the same for both the markers.
Maternal serum inhibin levels in twins did not differ between ART and spontaneous pregnancies. Although ART might be expected to alter inhibin levels to a similar extent in twins and singleton pregnancies, there could be other confounding factors. For example, ART twins are predominantly dizygotic while 5–15% of spontaneous pregnancies are monozygotic depending on age. And it may be relevant that although HCG levels are increased in ART singletons, there is no evidence that this is the case in twins. Räty et al. (2000) found the median second trimester level in 29 IVF twins was 2.20 MoM compared with 1.83 MoM in 39 spontaneous twins, which did not reach statistical significance (P = 0.08). On the other hand, Orlandi et al. (2002) studied first trimester levels in 30 ART twins and observed a median of 1.71 MoM compared with 2.01 MoM in 150 controls. The weighted geometric averages for the two studies are 1.94 MoM for ART and 1.97 MoM for spontaneous twins.
Currently the overall median inhibin level in twins (spontaneous and ART) was 2.12 MoM and the 10–90th centile range was 1.18–3.93 MoM. This is consistent with the only other twin series where the median in 200 pregnancies was 1.99 MoM and the 10–90th centile range 1.10–3.68 MoM (Watt et al., 1996). Taking the weighted geometric average of both yields a value of 2.03 MoM.
The overall median for twins was higher in the UK than in Israel: 2.23 MoM compared with 1.76 MoM. This could be a chance effect, although the difference reached statistical significance (P = 0.05). Differences in the method of gestational estimation for twins or ethnicity could have contributed to this as could different rates of smoking (Renier et al., 1998; Ferriman et al., 1999; Rudnicka et al., 2002). However, we did not collect this information in both series; hence, these possibilities could not be investigated directly.
Fetal reduction had no material effect on inhibin levels in the present study. Similarly, second trimester uE3 and HCG levels do not appear to be altered (Groutz et al., 1996; Shulman et al., 1996; Shulman and Phillips, 1997; Rotmensch et al., 1999), but AFP levels can be greatly elevated (Grau et al., 1990; Lynch and Berkowitz, 1993; Abbas et al., 1994; Groutz et al., 1996; Shulman et al., 1996; Shulman and Phillips, 1997; Rotmensch et al., 1999). Although information of reduction is not recorded on the Leeds DS screening request form, this will not have materially distorted any comparisons, as reduction is rare in the UK. Triplet pregnancies following successful infertility therapy are a major concern in many countries where the transfer of three or more embryos is still common (Nyboe Andersen et al., 2004). This adverse overcome can be avoided by the use of fetal reduction to twins (Maymon et al., 1995), although care is needed in the interpretation of screening tests which include AFP, and there is an increasing trend to transfer only one or two embryos.
There are two practical conclusions of our study. Firstly, when DS screening tests that include inhibin are performed in singleton ART pregnancies, test interpretation should take account of the increased average level as is already done by some centers for HCG and free -HCG. Secondly, there is now sufficient data on inhibin levels in twins, even when there has been fetal reduction from triplets, to confidently calculate DS risks for protocols which include this analyte. Serum markers alone are of limited value in screening twins for DS, because the unaffected co-twin masks the abnormal maternal serum distribution seen in singleton analyses (Cuckle, 1998). Furthermore, while this screening may identify a high-risk pregnancy, unlike NT, this will not pinpoint which is the high-risk fetus (Spencer and Nicolaides, 2003). Despite these limitations, the addition of first or second trimester serum markers to NT screening will increase the detection rate for DS twins (Cuckle and Arbuzova, 2004).
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
The authors are grateful to Oxford Bio-Innovation Ltd, Oxford, UK, for their support.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Abbas A, Johnson M, Bersinger N and Nicolaides K (1994) Maternal alpha-fetoprotein levels in multiple pregnancies. Br J Obstet Gynaecol 101, 156–158.[Web of Science][Medline]
Cuckle H (1998) Down’s syndrome screening in twins. J Med Screen 5,3–4.
Cuckle H and Arbuzova S (2004) Multianalyte maternal serum screening for chromosomal defects. In Milunsky A (ed.) Genetic Disorders and the Fetus: Diagnosis, Prevention and Treatment, 5th edn. Johns Hopkins University Press, Baltimore, pp. 795–835.
Cuckle H, Benn P and Wright D (2005) Down syndrome screening in the first and/or second trimester: model predicted performance using meta-analysis parameters. Semin Perinatol, 29,252–257.[CrossRef][Web of Science][Medline]
Ferriman EL, Sehmi IK, Jones R and Cuckle HS (1999) The effect of smoking in pregnancy on maternal serum inhibin A levels. Prenat Diagn 19,372–374.[CrossRef][Web of Science][Medline]
Grau P, Robinson L, Tabsh K and Crandall BF (1990) Elevated maternal serum alpha-fetoprotein and amniotic fluid alpha-fetoprotein after multifetal pregnancy reduction. Obstet Gynecol 76,1042–1045.[Web of Science][Medline]
Groutz A, Amit A, Yaron Y, Yovel I, Wolman I, Legum C and Lessing JB (1996) Second-trimester maternal serum alpha-fetoprotein, human chorionic gonadotropin and unconjugated oestriol after early transvaginal multifetal pregnancy reduction. Prenat Diagn 16,723–727.[CrossRef][Web of Science][Medline]
Lynch L and Berkowitz RL (1993) Maternal serum alpha-fetoprotein and coagulation profiles after multifetal pregnancy reduction. Am J Obstet Gynecol 169,987–990.[Web of Science][Medline]
Marrs RP, Kletzky OA, Howard WF and Mishel DR (1979) Disappearance of human chorionic gonadotropin and resumption of ovulation following abortion. Am J Obstet Gynecol 135,731–736.[Web of Science][Medline]
Maymon R, Herman A, Shulman A, Halperin R, Arieli S, Bukovsky I and Weinraub Z (1995) First trimester embryo reduction: a medical solution to an introgenic problem. Hum Reprod 10,668–673.
Maymon R and Shulman A (2001) Comparison of triple serum screening and pregnancy outcome in oocyte donation versus IVF pregnancies. Hum Reprod 16,691–695.
Nyboe Andersen A, Gianaroli L and Nygren KG (2004) European IVF-monitoring programme; European Society of Human Reproduction and Embryology 2004 Assisted reproductive technology in Europe, 2000. Results generated from European registers by ESHRE. Hum Reprod 19,490–503.
Orlandi F, Rossi C, Allegra A, Krantz D, Hallahan T, Orlandi E and Macri J (2002) First trimester screening with free beta-hCG, PAPP-A and nuchal translucency in pregnancies conceived with assisted reproduction. Prenat Diagn 22,718–721.[CrossRef][Web of Science][Medline]
Räty R, Virtanen A, Koskinen P, Laitinen P, Forsstrom J, Salonen R, Morsky P and Ekblad U (2000) Maternal midtrimester serum AFP and free beta-hCG levels in in vitro fertilization twin pregnancies. Prenat Diagn 20,221–223.[CrossRef][Web of Science][Medline]
Renier MA, Vereecken A, Van Herck E, Straetmans D, Ramaekers P and Buytaert P (1998) Second trimester maternal dimeric inhibin-A in the multiple-marker screening test for Down’s syndrome. Hum Reprod 13,744–748.
Ribbert LS, Kornman LH, DeWolf BT, Simons AH, Jansen CA, Beekhuis JR and Mantingh A (1996) Maternal serum screening for fetal Down syndrome in IVF pregnancies. Prenat Diagn 16,35–38.[CrossRef][Web of Science][Medline]
Rotmensch S, Celentano C, Shalev J, Vishne TH, Lipitz S, Ben-Rafael Z and Glezerman M (1999) Midtrimester maternal serum screening after multifetal pregnancy reduction in pregnancies conceived by in vitro fertilization. J Assist Reprod Genet 16,8–12.[CrossRef][Web of Science][Medline]
Rudnicka AR, Wald NJ, Huttly W and Hackshaw AK (2002) Influence of maternal smoking on the birth prevalence of Down syndrome and on second trimester screening performance. Prenat Diagn 22,893–897.[CrossRef][Web of Science][Medline]
Shulman LP and Phillips OP (1997) Second-trimester maternal serum alpha-fetoprotein, human chorionic gonadotropin and unconjugated oestriol after early transvaginal multifetal pregnancy reduction. Prenat Diagn 17,488–490.[CrossRef][Web of Science][Medline]
Shulman LP, Phillips OP and Cervetti TA (1996) Maternal serum analyte levels after first-trimester multifetal pregnancy reduction. Am J Obstet Gynecol 174,1072–1074.[CrossRef][Web of Science][Medline]
Spencer K and Nicolaides KH (2003) Screening for trisomy 21 in twins using first trimester ultrasound and maternal serum biochemistry in a one-stop clinic: a review of three years experience. Br J Obstet Gynaecol 110, 276–280.
Wald N, White N, Morris JK, Huttly WJ and Canick JA (1999) Serum markers for Down’s syndrome in women who have had in vitro fertilisation: implications for antenatal screening. Br J Obstet Gynaecol 106,1304–1306.[Web of Science][Medline]
Watt HC, Wald NJ and George L (1996) Maternal serum inhibin-A levels in twin pregnancies: implications for screening for Down’s syndrome. Prenat Diagn 16,927–929.[CrossRef][Web of Science][Medline]
Submitted on September 19, 2005; resubmitted on December 13, 2005; accepted on December 19, 2005.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||