Circulating levels of inhibin A, activin A and follistatin in missed and recurrent miscarriages

S. Muttukrishna1,4, E. Jauniaux1, N. Greenwold1, H. McGarrigle1, S. Jivraj2, S. Carter2, S. Elgaddal2, N. Groome3 and L. Regan2

1 Department of Obstetrics and Gynaecology, Royal Free-UCL Medical School, 86–96 Chenies Mews, London WC1E 6HX, 2 Academic Department of Obstetrics and Gynaecology, St. Mary’s Hospital, Imperial College Medical School, London and 3 School of Biological and Molecular Sciences, Oxford Brookes University, Gipsy Lane, Headington, Oxford, UK


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: The aim of this study was to investigate the changes in circulating levels and the clinical use of inhibin A, activin A and follistatin as endocrine markers of early pregnancy loss. METHODS: Blood samples were collected from women presenting with a sporadic missed miscarriage (n = 10), and controls having pregnancy termination at 8–12 weeks (n = 15) and from women with a history of unexplained recurrent miscarriages (n = 12) at 6–12 weeks gestation. All samples were assayed for inhibin A, inhibin B, activin A, follistatin, hCG, estradiol and progesterone. RESULTS: Serum inhibin A, hCG, estradiol and progesterone levels were significantly (~2–3 fold) decreased in sporadic miscarriages compared with controls. In the recurrent miscarriage group, time dependent changes in plasma inhibin A and hCG levels were significantly (P < 0.05) altered in the group that had a subsequent miscarriage compared with those who had a live birth. At 6–7 weeks gestation, plasma inhibin A (~4 fold, P < 0.01), hCG (~4 fold, P < 0.01) and estradiol (~2 fold, P < 0.001) levels were significantly lower in women who went on to have another miscarriage than those with a live birth. Inhibin B levels were near the detection limit of the assay. CONCLUSIONS: Our findings suggest that inhibin A is a specific marker of early pregnancy loss before the onset of the clinical symptoms of recurrent miscarriage. There is a high degree of association between levels of inhibin A and hCG in cases of miscarriage, indicating that these two proteins could be used in combination to predict future pregnancy outcome.

Key words: activin A/early pregnancy/follistatin/inhibin A/miscarriage


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Acknowledgements
 References
 
High concentrations of serum inhibin A, activin A and follistatin are found in the circulation of women with normal pregnancies (Muttukrishna et al., 1995Go, 1996Go; Fowler et al., 1998Go). Inhibin A is from the feto–placental unit (Birdsall et al., 1997Go; Muttukrishna et al., 1997Go) and the corpus luteum (Treetampinich et al., 2000Go). There is evidence for the secretion of these proteins by placental explants (Keelan et al., 1998Go; Riley et al., 2000Go) and trophoblasts in culture (Mohan et al., 2001Go). Both placental cytotrophoblast and the syncytiotrophoblast synthesise inhibin A, activin A and follistatin. Inhibin is a heterodimeric glycoprotein hormone with an {alpha} and a ß subunit, activin is a homo-dimer of two ß subunits and follistatin is a single chain glycoprotein with a high affinity binding for activin. There is limited information about the biological role of these proteins throughout pregnancy. Some data (Canniggia et al., 1997) suggest that in early pregnancy, activin A may promote invasion of extra-villous trophoblast. We have also shown that inhibin A and activin A concentrations are raised at 15–18 weeks gestation in women who develop pre-eclampsia later (Muttukrishna et al., 2000Go). In contrast, it has been reported in IVF pregnancies that inhibin A levels are low in missed abortions compared with ongoing IVF pregnancies (Lockwood et al., 1997Go; Treetampinich et al., 2000Go). These findings suggest that these proteins may play a role in placentation.

Miscarriage is the most common complication of pregnancy. Early pregnancy loss occurs in 10–20% of clinically recognized gestations (Alberman, 1992Go). Of these, a proportion will present to medical practitioners as ‘missed miscarriages’.

As distinct to recurrent miscarriage, cases of sporadic/missed miscarriage have no defined past history of miscarriage. In missed miscarriage, also referred to as missed abortion in the literature, failure of pregnancy is identified before expulsion of the fetal and placental tissue. The diagnosis is made on ultrasound by either failure to identify fetal heart action or by the presence of an empty gestational sac where early embryonic death has been followed by resorption of the fetal tissue, thus giving rise to the more recent description of missed miscarriage as early pregnancy failure (EPF). These empty gestation sacs were in the past described as ‘anembryonic pregnancies’, this being rather misleading in its inference that there was total non-development of the fetus. Isolated cases of EPF presenting at random to the early pregnancy unit with a history of bleeding per vaginam or with a complaint of diminishing pregnancy symptoms will be referred to in this study as sporadic miscarriage.

Recurrent miscarriage is defined as three or more consecutive pregnancy losses and affects 1% of couples trying to conceive (Regan, 1997Go). Sporadic miscarriages are essentially linked with chromosomal abnormalities of the conceptus (Hassold, 1986Go) whereas recurrent miscarriages have been associated mainly with karyotypic abnormalities in either partner, or uterine anomalies such as fibroids and congenital malformation of the mesonephric systems. More recently primary antiphospholipid syndrome, thrombophilic states including activated protein C resistance and Factor V Leiden have been found in association with recurrent miscarriage (Regan, 1997Go). The predictive value of human chorionic gonadotrophin (hCG), estradiol and progesterone has been investigated extensively in predicting sporadic miscarriage (Nygren et al., 1973Go; Confino et al., 1986Go; Whittaker et al., 1989Go; Johnson et al., 1993Go). Inhibin A has been shown to be low in non-viable IVF (Lockwood et al., 1997Go) and spontaneous pregnancies with symptoms of miscarriage (Phipps et al., 2000Go) compared with ongoing pregnancies. However, there is only one study of the circulatory levels of immuno reactive (ir) inhibin [dimeric inhibins (A and B) + monomeric {alpha} subunit] in recurrent miscarriage (France et al., 1996Go). These authors found that ir-inhibin concentrations are not significantly altered in women before miscarriage but the assay used in their study was not specific for biologically active dimeric inhibin A or inhibin B, because monomeric {alpha} subunit is produced by several tissues including the ovaries. The aim of the present study was to further investigate circulating levels of inhibin A, inhibin B, activin A and follistatin in women with sporadic or recurrent miscarriages. We have evaluated whether circulating levels of inhibin A, activin A and follistatin are endocrine markers of early pregnancy loss in patients with a history of recurrent miscarriage and are linked with other serum markers such as estradiol, progesterone and hCG.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Acknowledgements
 References
 
Study 1
Serum samples were obtained from women presenting with a sporadic missed miscarriage (n = 10) at the time of surgical uterine evacuation between 8 and 12 gestational weeks. Serum samples from women with normal pregnancy (n = 15) undergoing termination of pregnancy at 8–12 weeks gestation for social reasons, were used as controls.

Study 2
Longitudinal samples were collected weekly or fortnightly between 6 and 12 weeks gestation in 12 women referred to the recurrent miscarriage clinic at St Mary’s Hospital. In addition, at 6–7 weeks gestation, samples were collected from 18 women attending the clinic. None of these women had any infection. All women were investigated according to the clinic protocol. The mean maternal age of the group with live births (33.8 ± 6.5 years; mean ± SD; range 23–41) and recurrent miscarriages (36.8 ± 5.1 years; mean ± SD; range 29–44) were not significantly different. Each group (live/recurrent) included five nulliparous and four parous (one/two) women. Patients on any medication other than folic acid were excluded. Women and their partners were tested for peripheral blood karyotype abnormalities. Women were also tested for antiphospholipid antibodies (IgG and IgM anticardiolipin antibodies and lupus anticoagulant), early follicular gonadotrophin and testosterone levels, and had pelvic ultrasound scans to determine ovarian and uterine morphology. In addition, blood was taken for activated protein C resistance ratio and restriction fragment length polymorphism analysis for factor V Leiden. Unexplained recurrent miscarriage was diagnosed when no cause for recurrent miscarriage was found. Women were subsequently followed-up in the early pregnancy clinic for weekly or fortnightly ultrasound scans and consultations in their next pregnancy. For patients with unexplained recurrent miscarriage, apart from folic acid (Department of Health guidelines) no other medication was administered. Pregnancy outcome, including a subsequent miscarriage or a live birth, was determined prospectively in each patient by way of a return questionnaire.

In all groups, gestational age was determined from the first day of the last menstrual period and confirmed by ultrasound measurement of the fetal crown–rump length in ongoing pregnancies. Cytogenetic analysis was obtained from placental tissue in all cases of miscarriages and in controls.

Blood samples were collected in tubes containing EDTA from women with a history of unexplained recurrent miscarriages and centrifuged to separate blood cells and platelets from plasma. Serum and plasma was then stored at –80°C and analysed later for inhibin A, inhibin B, activin A, follistatin, progesterone, estradiol and hCG. Written consent was obtained from each woman after receiving complete information about the procedure. Approval for this study was obtained from the local research ethics committees.

Hormone assays
Inhibin A was measured using a two-site ELISA (enzyme linked immunosorbent assay) that has been previously validated for human serum (Muttukrishna et al., 1994Go). The minimum detection limit of this assay for human recombinant inhibin A (National Institute for Biological Standards, Potters Bar, Herts, UK) was 2 pg/ml. Intra- and interassay variations were 4.5 and 5.1% respectively.

Inhibin B was measured using an enzyme immunoassay as described in detail elsewhere (Groome et al., 1996Go). An in-house standard preparation (partially purified human follicular fluid) was standardized against human recombinant inhibin B (Genentech, San Francisco, CA, USA) and was used as the assay standard. The minimum detection limit of the assay for human recombinant inhibin B was 15 pg/ml. The mean intra and interassay coefficients of variation (CV) were 6.2 and 7.2% respectively.

Activin A was measured using a two-site ELISA specific for ‘total’ activin A (follistatin bound + unbound activin A) as previously described (Muttukrishna et al., 1996Go). The detection limit of this assay for human recombinant activin A (Genentech) was 50 pg/ml. Intra- and interassay variations were 8.5 and 9.8% respectively.

Follistatin was measured using a two-site ELISA as previously described (Evans et al., 1998Go). The sensitivity of this assay was 20 pg/ml. Intra- and interassay variations were 6.8 and 9.15% respectively.

hCG was measured by an immuno-radiometric assay (IRMA) using a commercial kit purchased from EURO DPC (Llanberis, UK). Intra- and interassay CV were <10%. The minimum detection limit of the assay was 0.03 mIU/ml.

Estradiol and progesterone were measured by radioimmunoassay (following ether extraction and using tritium labelled tracers) as previously described (Darne et al., 1989Go). Intra- and interassay CV for both assays were <10%. The minimum detection limit of the estradiol RIA was <80 pmol/l and progesterone was 1 nmol/l.

Statistical analysis
Data were log transformed to obtain a normal distribution. In the missed miscarriage study, an unpaired Student’s t-test was used to analyse differences between the groups. In the recurrent miscarriage group, analysis of variance was carried out to study the time course changes in the longitudinal study and paired sample t-tests were used to study the difference between the women with subsequent miscarriages and women with a normal outcome. Pearson’s correlation analysis was carried out to study the relationship between the different hormones measured. The SPSS statistical package was used for the analysis (SPSS Inc., Chicago, Illinois, USA). P was considered significant at <0.05.


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Acknowledgements
 References
 
Serum inhibin A concentrations were significantly (P < 0.03) lower in the women with sporadic missed miscarriage (71.1 ± 12.1 pg/ml) than in controls (193 ± 48 pg/ml). Serum activin A (510.8 ± 53 pg/ml versus 544.7 ± 81 pg/ml) and follistatin levels (2572 ± 955 pg/ml versus 2111 ± 545 pg/ml) were similar in the miscarriage and control groups whereas progesterone (36.1 ± 4.96 nmol/l versus 75.1 ± 7.2 nmol/l; P < 0.01), estradiol (0.98 ± 0.16 nmol/l versus 3.3 ± 0.52 nmol/l; P < 0.01) and hCG levels (23.84 ± 4.3 versus 95.5 ± 11.55; P < 0.001) were significantly lower in the miscarriage than in the control group (Figure 1Go). Inhibin B levels were either undetectable or near the detection limit of the assay (15–20 pg/ml) with no variations between the groups.



View larger version (26K):
[in this window]
[in a new window]
 
Figure 1. Individual concentrations of (a) inhibin A, (b) activin A (c) follistatin, (d) progesterone, (e) estradiol and (f) hCG in the maternal serum of control pregnant women undergoing pregnancy termination (n = 15) and patients with missed abortions (n = 10) at 8–12 weeks gestation. The line straight across is the median. Student’s t-test results: * = P < 0.05, ** = P < 0.01, *** = P < 0.001, NS = not significant.

 
In the recurrent miscarriage group, samples taken at 6–8, 8–10 and 10–12 weeks in women with recurrent miscarriages were assayed individually and grouped for analysis (Figure 2Go). Analysis of variance showed that all measured parameters except inhibin B varied significantly with increasing gestational age in women with a subsequent miscarriage compared with a live birth. Overall, levels of inhibin A (P < 0.05) and hCG (P < 0.05) were significantly different in both subgroups whereas no difference was found for activin A, follistatin, progesterone and estradiol. At 6–7 weeks gestation, levels of inhibin A (139.5 ± 39 versus 34 ± 12.9 pg/ml; P < 0.01), estradiol (2 ± 0.26 versus 0.9 ± 0.15 nmol/l; P < 0.001) and hCG (46 ± 9.3 versus 9.7 ± 2.1 IU/ml; P < 0.01) were significantly higher in women who had a live birth compared with women who subsequently miscarried. Plasma activin A (345.2 ± 64 versus 246.2 ± 64.3 pg/ml), follistatin (1622 ± 400 versus 1325 ± 108 pg/ml) and progesterone (60.1 ± 9 versus 42 ± 8.5 nmol/l) levels were not significantly different in both subgroups (Figure 3Go).



View larger version (24K):
[in this window]
[in a new window]
 
Figure 2. Time dependent changes in the mean concentrations of (a) inhibin A, (b) activin A (c) follistatin, (d) progesterone, (e) estradiol and (f) hCG in the maternal serum of recurrent miscarriage patients who subsequently had live births (n = 6) or recurrent miscarriage(n = 6). Analysis of variance results showing the significance in difference in levels between the two groups are given.

 


View larger version (23K):
[in this window]
[in a new window]
 
Figure 3. Individual concentrations of (a) inhibin A, (b) activin A (c) follistatin, (d) progesterone, (e) estradiol and (f) hCG in the maternal plasma of recurrent miscarriage patients who subsequently had live births (n = 9) or recurrent miscarriages (n = 9). Samples were taken at 6–7 weeks gestation. The line straight across is the median. Paired Student’s t-test results; *** = P < 0.001, ** = P < 0.01, NS = not significant.

 
In normal controls, serum inhibin A concentrations were positively correlated with hCG (r = 0.548, P = 0.05). Progesterone levels were positively correlated with estradiol (r = 0.558, P < 0.05). In sporadic miscarriages, inhibin A level was positively correlated with progesterone (r = 0.756, P = 0.03) and hCG (r = 0.789, P = 0.02).

In the recurrent miscarriage group, the data regarding women who had another miscarriage were analysed separately from those with a live birth. In the women with a live birth, plasma inhibin A showed a positive correlation with activin A (r = 0.563, P < 0.01), estradiol (r = 0.61, P < 0.01) and hCG (r = 0.66, P = 0.001). Plasma activin A had a negative correlation with follistatin (r = –0.41, P = 0.05) and a positive correlation with hCG (r = 0.557, P < 0.01). Plasma estradiol had a positive correlation with hCG (r = 0.473, P < 0.05).

In women with a subsequent miscarriage, plasma inhibin A was positively correlated with activin A (r = 0.511, P < 0.05), progesterone (r = 0.55, P < 0.05), estradiol (r = 0.45, P < 0.05) and hCG (r = 0.989, P < 0.001). Plasma activin A was also positively correlated with progesterone (r = 0.55, P < 0.05), estradiol (r = 0.58, P < 0.01) and hCG (r = 0.511, P < 0.05). Plasma progesterone levels were positively correlated with estradiol (r = 0.714, P < 0.001) and hCG (r = 0.49, P < 0.05).


    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Acknowledgements
 References
 
Our data indicate that inhibin A, like hCG, estradiol and progesterone, could be a specific marker of early pregnancy loss from as early as 6–7 weeks gestation. This is the first study to show altered levels of inhibin A before the onset of the clinical symptoms of a subsequent miscarriage in recurrent miscarriage patients. The feto–placental unit secretes inhibin A and activin A in early pregnancy (Birdsall et al., 1997Go; Muttukrishna et al., 1997Go). Our recent study of these proteins in the different compartments of the first trimester gestational sac has shown that placental extracts contain large quantities of these proteins, suggesting that placental inhibin A, activin A and follistatin may act locally in an autocrine/paracrine manner (Muttukrishna et al., 2002Go).

Several studies have shown that estradiol, progesterone and hCG can predict pregnancy loss (Check et al., 1992Go; Ogasawara et al., 1997Go; Ong et al., 2000Go; Phipps et al., 2000Go; Zayed et al., 2001Go). Phipps et al.(2000) have shown that inhibin A levels are low in those patients with clinical symptoms of spontaneous miscarriage. After 6–7 weeks gestation the trophoblast is considered to be the principal source of estrogens and progesterone (Csapo and Pulkkinen, 1978Go). Thus, a pregnancy failure after this time can mainly be linked to placental dysfunction either secondary to a fetal abnormality or as the result of primary trophoblast dysfunction. In the present study, women with miscarriages all presented after 8 weeks gestation and four out of 10 of those presenting with a sporadic miscarriage had a fetal chromosomal abnormality, suggesting placental dysfunction as the main cause of miscarriage. Anatomical studies of the placental bed in women with pre-eclampsia and miscarriage have shown a similar defect in the trophoblastic infiltration of the placenta–decidual interface (Hustin et al., 1990Go). In both types of pregnancy complications, this leads to an incomplete transformation of the utero-placental arteries and abnormal villous development. In miscarriage, this defect of placentation is often very severe and associated with the premature entry of a large amount of maternal blood inside the placenta, which progressively dislocates the entire gestational sac (Jauniaux et al., 1994Go). Serum inhibin A was ~3 fold less in women presenting with a sporadic miscarriage compared with controls, however, activin A and follistatin levels were not significantly altered at 8–12 weeks gestation in cases of miscarriage. In early pregnancy, inhibin A levels gradually rise up to 12 weeks and then decline to a nadir and remain low in the second trimester (Muttukrishna et al., 1995Go). Activin A levels are similar in the first and second trimester (Muttukrishna et al., 1996Go) whereas follistatin levels progressively rise throughout pregnancy (Fowler et al., 1998Go). Although inhibin A and activin A are in circulation in pregnancy, a precise biological role for these proteins in pregnancy is, as yet, unclear. Studies in other species suggest that activin A is important for early embryonic development (Asashima et al., 1990Go; Green and Smith, 1990Go; Ariizumi et al., 1991Go; Jenkin et al., 1995Go).

Serum progesterone (~2 fold), estradiol (~3.5 fold) and hCG (~4 fold) levels were also significantly lower in miscarriage patients compared with controls. Serum progesterone and hCG were significantly correlated with levels of inhibin A suggesting similar source(s) and mechanisms regulating these hormones. Activin A and follistatin have several sources such as the placenta, peripheral mononuclear cells (Muttukrishna et al., 2001Go), vascular endothelial cells (Phillips et al., 2001Go) and bone marrow (Mather et al., 1997Go). The lack of difference in the levels of activin A and follistatin in miscarriage patients could be due to other sources for these hormones, possibly masking the reflection of any changes in production by the feto–placental unit.

In women with recurrent miscarriages all samples were collected before the onset of the clinical symptoms. In this group, time dependent changes in inhibin A and hCG levels were significantly lower in women who had another miscarriage compared with those who had live births. At 6–7 weeks, women who had a subsequent miscarriage had significantly lower levels of inhibin A (~4 fold), estradiol (~2 fold) and hCG (~4 fold) compared with those whose pregnancy ended with a live birth. These data indicate that inhibin A and hCG levels are consistently lower in all cases of miscarriage and are positively and significantly correlated with each other. Serum hCG levels rise until 8–9 weeks in normal early pregnancy and then decrease and remain low during the second and third trimesters. Whilst inhibin A levels rise in early pregnancy and decrease in the second trimester, inhibin A increases again in the third trimester of pregnancy.

It has been previously reported that lower levels of inhibin A are found in the maternal circulation in IVF missed abortions (Lockwood et al., 1997Go) and spontaneous non-viable pregnancies with clinical symptoms of miscarriage (Phipps et al., 2000Go). Our missed miscarriage data confirm these previous studies with lower maternal circulatory levels of inhibin A in miscarriage patients compared with the controls. Consistent with our current observation in sporadic missed miscarriages, activin A levels were not different in IVF missed abortions (Lockwood et al., 1997Go). These data suggest that inhibin A is a more specific marker of placental function than activin A or follistatin, as these are also produced by several other tissues. Our previous study investigating the clearance of inhibin A, activin A and hCG after pregnancy termination suggested that inhibin A concentrations clear from circulation within hours (Muttukrishna et al., 1997Go), unlike hCG which has a longer half life. This indicates that any variation in inhibin A production will quickly be reflected in the circulation, rendering it a sensitive marker of early placental dysfunction.

A previous study has reported ir-inhibin in recurrent miscarriage patients and found there was no correlation between hCG and ir-inhibin in these patients (France et al., 1996Go). In our study, we report a strong positive correlation between inhibin A and hCG (r = 0.989, P < 0.001) in the recurrent miscarriage group. The discrepancy between the two studies could be due to the specificity of the inhibin assay. We have specifically measured dimeric inhibin A, which is the bioactive dimeric inhibin form in the circulation of pregnant women, whilst in another study (France et al., 1996Go) a radioimmunoassay was used for inhibin which crossreacted with dimeric inhibins (A and B) and monomeric {alpha} subunit. It is clear from previous studies that inhibin A is the predominant molecular form of inhibin in the circulation during pregnancy (Muttukrishna, 2001Go).

In conclusion, this pilot study shows that inhibin A in combination with hCG could be useful in monitoring unexplained recurrent miscarriage patients to predict a subsequent miscarriage as early as 6 weeks gestation. However, larger studies with different groups of recurrent miscarriage patients are needed to confirm this observation and the potential clinical application. Future studies should focus on the role of inhibins in early pregnancy in order to inform therapeutic interventions.


    Acknowledgements
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Acknowledgements
 References
 
We wish to thank the National Institute for Biological Standards and Controls (Potters Bar, Hertfordshire, UK) for the recombinant inhibin A standard used in the assays and the Wellcome trust (Grant 059743) for their financial support.


    Notes
 
4 To whom correspondence should be addressed. E-mail: s.muttukrishna{at}ucl.ac.uk Back


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Acknowledgements
 References
 
Alberman, E. (1992) Spontaneous abortion: epidemiology. In Stabile, S., Grudzinskas, G. and Chard, T. (eds) Spontaneous abortion: diagnosis and treatment. Springer-Verlag, London, UK. pp. 9–30.

Asashima, M., Nakano, H., Shimada, K., Kinoshita, K., Shibai, H. and Ueno, N. (1990) Mesodermal induction in early amphibian embryos by activin A erythroid differentiation factor. Wilhelm Roux’s Arch. Dev. Biol., 198, 330–335.[ISI]

Ariizumi, T., Moriya, N., Uchiyama, H. and Asashima, M. (1991) Concentration dependent inducing activity of activin A. Wilhelm Roux’s Arch. Dev. Biol., 200, 230–233.[ISI]

Birdsall, M., Ledger, W., Groome, N., Abdalla, H. and Muttukrishna, S. (1997) Inhibin A and activin A in the first trimester of human pregnancy. J. Clin. Endocrinol. Metab., 82, 1557–1560.[Abstract/Free Full Text]

Caniggia, I., Lye, S.J. and Cross, J.C. (1997) Activin is a local regulator of human cytotrophoblasts cell differentiation. Endocrinology, 138, 3976–3986.[Abstract/Free Full Text]

Check, J.H., Lurie, D., Davies, E. and Vetter, B. (1992) Comparison of first trimester serum estradiol levels in aborters versus nonaborters during maintenance of normal progesterone levels. Gynecol. Obstet. Invest., 34, 206–210.[ISI][Medline]

Confino, E., Demir, R.H., Friberg, J. and Gleicher, N (1986) The predictive value of hCG-ß subunit levels in pregnancies achieved by in vitro fertilization and embryo transfer: an international collaborative study. Fertil. Steril., 45, 526–531.[ISI][Medline]

Csapo, A.I. and Pulkkinen, M. (1978) Indispensability of the human corpus luteum in the maintenance of early pregnancy. Lutectomy evidence. Obs. Gyn. Surv., 33, 69–81.

Darne, F.J., McGarrigle, H.H. and Lachelin G.C.L. (1989) Diurnal variation of plasma and saliva oestrogen, progesterone, cortisol and plasma dehydroepiandrosterone sulphate in late pregnancy. Eur. J. Obs. Gyn. Reprod. Biol., 32, 57–66.[ISI]

Evans, L.W., Muttukrishna, S. and Groome, N. (1998) Development, validation and application of an ultra sensitive two-site enzyme immunoassay for human follistatin. J. Endocrin., 156, 275–282.[Abstract/Free Full Text]

Fowler, P.A, Evans, L.W., Groome, N.P., Templeton, A. and Knight, P.G. (1998) A longitudinal study of maternal serum inhibin-A, inhibin-B, activin-A, activin-AB, pro-alpha C and follistatin during pregnancy. Hum. Reprod., 13, 3530–3536.[Abstract]

France, J.T., Keelan, J., Song, L., Liddell, H., Zanderigo, A. and Knox, B. (1996) Serum concentrations of human chorionic gonadotrophin and immunoreactive inhibin in early pregnancy and recurrent miscarriage: a longitudinal study. Aust. New Zea. J. Obs. Gyn., 36, 325–330.

Grandone, E., Margaglione, M., Colaizzo, D., d’Addedda, M., Cappucci,G., Vecchione, G., Scianname, N., Pavone, G. and Di Minno, G. (1997) Factor V Leiden is associated with repeated and recurrent unexplained fetal losses. Thromb. Haemost., 77, 822–824.[ISI][Medline]

Green, J.B. and Smith, J.C., (1990) Graded changes in dose of a Xenopus activin A homologue elicit stepwise transitions in embryonic cell fate. Nature, 347, 391–394.[ISI][Medline]

Groome, N.P., Illingworth, P.J., O’Brien, M., Pai, R., Rodger, F.E., Mather, J. and McNeilly, A.S. (1996) Measurement of dimeric inhibin-B throughout the human menstrual cycle. J. Clin. Endocrinol. Metab., 81, 1401–1405.[Abstract]

Hassold, T. (1986) Chromosome abnormalities in human reproductive wastage. Trends Genet., 2, 105–110.[ISI]

Hustin, J., Jauniaux, E. and Schaaps, J.P. (1990) Histological study of the materno–embryonic interface in spontaneous abortion. Placenta, 11, 477–486.[ISI][Medline]

Jauniaux, E., Zaidi, J., Jurkovic, D., Campbell, S. and Hustin, J. (1994) Comparison of color Doppler features and pathologic findings in complicated early pregnancy. Hum. Reprod., 9, 2432–2437.[Abstract]

Jenkin, G., McFarlane, J.R,. and de Kretser, D. (1995) Implication of inhibin and related proteins in fetal development. Reprod. Fert. Dev., 7, 323–331.[ISI][Medline]

Johnson, M.R., Riddle, A.F., Grudzinskas, J.G., Sharma, V., Collins, W.P. and Nicolaides, K.H. (1993) The role of trophoblast dysfunction in the aetiology of miscarriage. Brit. J. Obs. Gyn., 100, 353–359.[ISI]

Keelan, J. L., Groome, N. and Mitchell, M. (1998) Regulation of activin A production by human amnion, decidua and placenta in vitro by pro-inflammatory cytokines. Placenta, 19, 429–434.[ISI][Medline]

Lockwood, G.M., Ledger, W.L., Barlow, D.H., Groome, N.P. and Muttukrishna, S. (1997) Measurement of inhibin and activin in early human pregnancy: demonstration of feto-placental origin and role in prediction of early pregnancy outcome. Biol. Reprod., 57, 1490–1494.[Abstract]

Mather, J.P., Moore, A. and Li, R.H. (1997) Activins, inhibins, and follistatins, further thoughts on a growing family of regulators. Proc. Soc. Biol. Med., 215, 209–222.[Abstract]

Mohan, A., Asselin, J., Sargent, I.L., Groome, N. P. and Muttukrishna, S. (2001) Effect of cytokines and growth factors on the secretion of inhibin A, activin A and follistatin by term placental villous trophoblasts in culture. Eur. J. Endocrinology, 145, 505–511.[ISI][Medline]

Muttukrishna, S. (2001) Inhibin, activin and follistatin in human pregnancy. In Muttukrishna, S. and Ledger, W.L. (eds) Inhibin, activin and follistatin in human reproductive physiology. Imperial College Press, London, UK, pp. 119–140.

Muttukrishna, S., Fowler, P.A., Groome, N., Mitchell, G.G., Robertson, W.R. and Knight, P.G. (1994) Serum concentrations of dimeric inhibin during the spontaneous human menstrual cycle and after treatment with exogenous gonadotrophin. Hum. Reprod., 9, 1634–1642.[Abstract]

Muttukrishna, S., George, L., Fowler, P.A., Groome, N.P. and Knight, P.G. (1995) Measurement of serum concentrations of inhibin A ({alpha}ßA dimer) during human pregnancy. Clin. Endocrinology, 42, 391–397.[ISI][Medline]

Muttukrishna, S., Fowler, P.A., George, L., Groome, N.P. and Knight, P.G. (1996) Changes in peripheral serum levels of ‘total’ activin A during the human menstrual cycle and pregnancy. J. Clin. Endocrinol. Metab., 81, 3328–3334.[Abstract]

Muttukrishna, S., Child, T.J., Groome, N.P. and Ledger, W.L. (1997) The source of circulating levels of inhibin A, pro-alpha C containing inhibins and activin A in early pregnancy. Hum. Reprod., 12, 1089–1093.[ISI][Medline]

Muttukrishna, S., North, R.A., Morris, J., Schellenberg, J.C., Taylor, R.S., Asselin, J., Ledger, W.L., and Redman, C.W.G. (2000) Serum inhibin A and activin A are elevated prior to the onset of pre-eclampsia. Hum. Reprod., 15, 1640–1645.[Abstract/Free Full Text]

Muttukrishna, S., Tannetta, D., Groome, N., Redman, C. and Sargent, I. (2001) Secretion of activin A by peripheral blood mononuclear cells (PBMCs) in culture and flowcytometric detection of intra cytoplasmic inhibin/activin subunits and activin receptors. Mol. Cell Endocrinology, 180, 194.

Muttukrishna, S., Jauniaux, E., Pahal, G., Groome, N.P. and Rodeck, C.H. (2002) In vivo distribution of inhibin A, activin A and follistatin in early pregnancy. Abstract, 811, Ceska Gyn. (Suppl.), 2, 76.

Nygren, K.G., Johansson, E.D.B and Wide, L. (1973) Evaluation of the prognosis of threatened abortion from the peripheral plasma levels of progesterone, estradiol and human chorionic gonadotrophin. Am. J. Obs. Gyn., 116, 916–922.[ISI]

Ogasawara, M., Kajiura, S., Katano, K., Aoyama, T. and Aoki, K. (1997) Are serum progesterone levels predictive of recurrent miscarriage in future pregnancies? Fertil. Steril., 68, 806–809.[ISI][Medline]

Ong C.Y., Liao A.W., Spencer K., Munim S. and Nicolaides, K. (2000) First trimester maternal serum free beta human chorionic gonadotrophin and pregnancy associated plasma protein A as predictors of pregnancy complications. Brit. J. Obs. Gyn., 107, 1265–1270.[ISI]

Phillips, D.J., Jones, K.L., Jean-Pierre, Y., Scheerlinck, Hedger, M.P. and de Kretser, D.M. (2001) Evidence for activin A and follistatin involvement in the systemic inflammatory response. Mol. Cell Endocrinology, 180, 155–162.[ISI][Medline]

Phipps, M.G., Hogan, J.W., Peipert, J.F., Lambert-Messerlian, G.M., Canick, J.A. and Seifer, D.B (2000) Progesterone, inhibin and hCG multiple marker strategy to differentiate viable from non viable pregnancies. Obs. Gyn., 95, 227–231.[ISI]

Regan, L. (1997). Sporadic and Recurrent Miscarriage. In Grudzinskas, J.G. and O’Brien, P.M.S. (eds) Problems in early pregnancy. Advances in diagnosis and management. RCOG press, pp. 31–52.

Riley, S.C., Leask, R., Balfour, C., Brennand, J.E. and Groome, N.P. (2000) Production of inhibin forms by the fetal membranes, decidua, placenta and fetus at parturition. Hum. Reprod., 15, 578–583.[Abstract/Free Full Text]

Treetampinich, C., O’Connor, A.E., MacLachlan, V., Groome, N.P., and de Kretser, D. (2000). Maternal serum inhibin A concentrations in early pregnancy after IVF and embryo transfer reflect the corpus luteum contribution and pregnancy outcome. Hum. Reprod., 15, 2028–2032.[Abstract/Free Full Text]

Whittaker, P.G., Stewart, M.O., Taylor, A. and Lind, T. (1989) Some endocrinological events associated with early pregnancy failure. Brit. J. Obs. Gyn., 10, 1207–1214.

Zayed, F., Ghazawi, I., Francis, L. and Alchalabi, H. (2001) Predictive value of human chorionic gonadotrophin beta hCG in early pregnancy after assisted conception. Arc. Gyn. Obs., 265, 7–10.

Submitted on June 27, 2002; accepted on September 17, 2002.