Vol. 16 No 2 | Winter 2014
Recurrent miscarriage
Prof Steve Robson

This article is 10 years old and may no longer reflect current clinical practice.

New pharmacological approaches to a difficult problem.

Recurrent miscarriage, or recurrent pregnancy loss (RPL), is an uncommon, but very distressing problem, for couples. Using the definition of three or more consecutive first-trimester pregnancy losses, without a live birth, it is generally estimated that between one and three per cent of women are affected.1 Using a population-wide estimate of the rate of early pregnancy loss as about 15 per cent for any given pregnancy, then the statistical probability of three such losses in a row is 0.34 per cent (0.15 × 0.15 × 0.15 = 0.0034). The observed proportion is much higher, suggesting that a group of couples will have a true underlying pathology and this is not just bad luck.

There are a number of well-recognised risk factors for recurrent miscarriage (see Box 1). Most of us will initiate investigations to determine if any of these risk factors are present in younger women after they have three miscarriages, but often earlier (perhaps after only two miscarriages) in older women who do not have the luxury of time. The majority of couples with recurrent miscarriage do not have any underlying pathology detected, despite thorough investigation. This is a distressing situation.

When a definite (or likely) cause for an early pregnancy loss is discovered, therapy is commonly initiated. For example, when an inherited thrombophilia is diagnosed, treatment with anticoagulants (heparins or aspirin) is often commenced. Frustratingly, while anticoagulation seems to have an intuitive theoretical underpinning, randomised trials of low-dose aspirin, low molecular weight or unfractionated heparin have not produced convincing results.2 This is a nuisance, since as many as one woman in five who has recurrent miscarriage will screen positive for a thrombophilia and large numbers of studies have now been reported. Fortunately, the evidence seems to be clearer for anti-phospholipid syndrome (APS), a condition identified in as many as 15 per cent women with recurrent miscarriage: the risk of pregnancy loss can be halved by judicious use of a combination of aspirin and heparin.3

In a similar way, the other clinical entities that might predispose to recurrent miscarriage can be dealt with, though not necessarily easily. Anatomical abnormalities of the uterus, including uterine septae or fibroids, may or may not require surgical correction. Balanced translocations in the parents, though accounting for a very small number of recurrent miscarriage cases, may respond to either conservative management or, in selected cases, to the use of pre-implantation genetic screening of embryos with genomic hybridisation techniques in IVF. Good glycaemic control can be sought in women with previously poorly controlled diabetes. And so it goes.

The major challenge occurs in managing those couples where a very thorough investigation has revealed absolutely no pathology that might explain recurrent pregnancy loss. Unfortunately, this is the most common situation that faces us. Older longitudinal studies suggest that the most likely outcome for a couple who have had recurrent miscarriages, and in whom no underlying abnormality has been found, is an ongoing pregnancy next time.4 That said, recurrent miscarriage seems to be a common reason for seeing a specialist, and most of us will be faced with distressed couples in this situation.

No obvious aetiology: a pharmacological solution?

Couples with recurrent miscarriage are often concerned and keen for action, and although it can be difficult to subject ‘tender loving care’1 to clinical trials, reassurance and regular contact definitely play a vital role in management.5 Beyond this, a number of pharmacological therapies are in common use. Aspirin, for example, is an empirical therapy that is often tried. Unfortunately, there is no evidence that aspirin provides any benefit for women with unexplained recurrent miscarriage.6 Similarly, empirical use of heparins in the hope of treating undiagnosed low-grade inflammation or thrombosis in the placental bed does not seem to provide any benefit either.7

‘Immunotherapy’ is a broad term for treatments that aim to modulate a supposedly abnormal maternal immune response to the implanting embryo. These include such treatments as intravenous immunoglobulin (IVIg) infusions, trophoblast cell infusions, paternal cell immunisations and other related therapies. Again – and unfortunately – the evidence for such treatments is weak at the moment.1

One area that might be promising is the use of progesterone in early pregnancy. A recent systematic review of the use of progesterone in women with a history of recurrent miscarriage provided some support, reporting a statistically significant reduction in the rate of subsequent miscarriage in this group.8 The reviewers commented on the small number of women in the four trials included – only 225 in all – and the ‘poor methodological quality’ of the studies, including the fact that oral, intramuscular and vaginal routes of administration were all used.

Box 1. Known risk factors for recurrent miscarriage

  • Increasing maternal age
  • Increasing paternal age
  • Obesity
  • Anti-phospholipid syndrome
  • Lupus anticoagulant
  • Anticardiolipin antibody
  • Anti-B2 glycoprotein-I antibodies
  • Parental aneuploidies
  • Congenital uterine abnormalities
  • Inherited thrombophilias
  • Poorly controlled diabetes

The dialogue between embryo and endometrium

It seems clear that embryo implantation and early fetal growth are not simple processes. Studies in which decidualising endometrial stromal cells (ECSs) and blastocyst-stage embryos were cultured together clearly point to a complex biochemical ‘conversation’ between the embryo and the cells of the endometrium, with abnormal embryos seeming to inhibit the secretion of implantation-promoting factors.9 The interesting conclusion from such studies is that the endometrium seems to be able to detect an abnormal embryo and that decidualisation does not occur so that abnormal embryos are much less likely to implant.

Armed with the knowledge that the human endometrium has the ability to differentiate between normal and abnormal embryos, elegant experiments have been undertaken by Weimer and colleagues using endometrial cell monolayers obtained by biopsy from women of normal fertility and women with a history of recurrent miscarriage.10 When abnormal embryos were placed with the endometrium of normally fertile women, implantation did not occur, as expected. However, the endometrium of women with recurrent miscarriage was receptive to both normal and abnormal embryos. The authors duly concluded that endometrial cells in normally fertile women discriminate between high- and low-quality embryos, whereas the cells from women with recurrent miscarriage don’t discriminate at all. This is clearly a new paradigm.

Moderators of the endometrial-embryo dialogue

Natural killer cells resident within the uterus (uNK cells) appear to be quite different to those found in peripheral blood, whose role is to deal with viral infections and tumours, for example. The uNK cells seem instead to release cytokines involved in angiogenesis, and are likely to play a leading role in the immune function of the endometrium during embryo implantation.11 Studies suggest that women with recurrent miscarriage have greater numbers of uNK cells in the endometrium during the luteal phase12 and high concentrations of uNK cells appear to be associated with lower levels of 11β-HSD1 and decidualisation.

What does all of this mean? There is now experimental evidence that some women with otherwise unexplained recurrent miscarriage are almost ‘hyperfertile’ – their endometrium is receptive to implantation for a long time, and does not appear to be too selective, allowing abnormal embryos to implant. This would explain the common description of women who become pregnant easily, but lose their pregnancy early.

Prednisolone and recurrent miscarriage

There is experimental evidence that prednisolone therapy reduces the concentration of uNK cells in the endometrium13 and trials of prednisolone therapy are underway across Europe. Prednisolone is the active metabolite of prednisone, and prednisone is a synthetic form of cortisol. Cortisol is a steroid released from the adrenal gland, its functions to suppress the immune system and to act on metabolic pathways so as to increase blood glucose levels, hence the term ‘glucocorticoid’. Glucocorticoids down-regulate interleukin receptors, dampening humoral immune responses. Prednisolone binds with glucocorticoid receptors (GRs) and the bonds are irreversible – the resulting complexes migrate into the cell nucleus and strongly influence cell functions.

Prednisolone therapy has been used for women with recurrent miscarriage for two decades, for most of that time without a clear theory of the basis of its action. It has only been in the last few years that the possible link between uNK cell numbers, endometrial hyper-receptivity and recurrent miscarriage has been explored in detail with the realisation that prednisolone may improve pregnancy outcomes by suppressing cytokine production and uNK cytolysis.14

Meta-analysis of the relevant trials concluded that, although corticosteroids do not impart a strong risk of teratogenesis, their use does increase the risk of oral clefts with an increase in risk from about 0.1 per cent to between 0.3 and 0.4 per cent.15 A subsequent prospective study did not report any teratogenic effect from systemic corticosteroid use in the first trimester.16 While corticosteroid use may increase the risk of preterm delivery, gestational diabetes and hypertension in pregnancy, such effects appear to be restricted to women taking steroids in high dose after the first trimester and use in the first trimester only does not appear to increase any of these risks.17,18


Recurrent miscarriage is a distressing condition to deal with, yet it is something that demands action. Investigation will reveal possible causes in less than half of couples, meaning that the majority of women in this situation will not be given a satisfactory explanation for what seems a devastating problem. In these circumstances, empirical treatments such as low-dose aspirin, heparins and various ‘immunotherapies’ have not been shown to provide any real benefit at all. The possibility is open that progesterone treatment might help, but this needs more work.

The true revolution has been the realisation that women with recurrent miscarriage might actually be ‘hyper-fertile’. There is evolving and intriguing experimental evidence that women with otherwise unexplained recurrent miscarriage have an endometrium that is overly receptive and receptive for a longer time to abnormal embryos. Hence the common clinical story of a couple who become pregnant quickly and without problems, only to miscarry each time. There is mounting evidence that uNK cells play a role, perhaps by affecting the decidualisation and by interfering with the endometrial-embryo chemical ‘dialogue’. The use of glucocorticoid treatment – of prednisolone in particular – might well alter this balance, restoring the normal receptivity and selectivity of the endometrium and restricting implantation to healthy embryos. The results of these trials are keenly awaited and perhaps might provide new hope to desperate couples.


  1. Tang A, Quenby S. Recent thoughts on management and prevention of recurrent early pregnancy loss. Curr Opin Obstet Gynecol 2010; 22: 446-51.
  2. McNamee K, Dawood F, Farquharson R. Thrombophilia and early pregnancy loss. Best Pract Res Clin Obstet Gynaecol 2012; 26: 91-102.
  3. Empson M, Lassere M, Craig J, Scott J. Prevention of recurrent miscarriage for women with antiphospholipid antibody or lupus anticoagulant. Cochrane Database Syst Rev 2005: CD002859.
  4. Brigham S, Conlon C, Farquharson R. A longitudinal study of pregnancy outcome following idiopathic recurrent miscarriage. Hum Reprod 1999; 14: 2868-71.
  5. Quenby S, Farquharson R. Predicting recurrent miscarriage: what is important? Obstet Gynecol 1993; 82: 132-8.
  6. Kaandorp S, Di Nisio M, Goddijn M, Middeldorp S. Aspirin or anticoagulants for treating recurrent miscarriage in women without antiphospholipid syndrome. Cochrane Database Syst Rev 2009: CD004734.
  7. Mantha S, Bauer K, Zwicker J. Low molecular weight heparin to achieve live birth following unexplained pregnancy loss: a systematic review. J Thromb Haemostasis 2010; 8: 263-8.
  8. Haas D, Ramsey P. Progestogen for preventing miscarriage. Cochrane
  9. Database Syst Rev 2013; CD003511. Teklenburg G, Salker M, Molokhia M, et al. Natural selection of human embryos: decidualizing endometrial stromal cells serve as sensors of embryo quality upon implantation. PLoS One 2010; 5: e10258.
  10. Weimar C, Kavelaars A, Brosens J, et al. Endometrial stromal cells of women with recurrent miscarriage fail to discriminate between hi and
    low-quality human embryos. PLoS One 2012; 7: e41424.
  11. Tang A, Alfirevic Z, Quenby S. Natural killer cells and pregnancy outcomes in women with recurrent miscarriage and infertility: a systematic review. Hum Reprod 2011; 26: 1971-80.
  12. Tuckerman E, Laird S, Prakash A, Li T. Prognostic value of the measurement of uterine natural killer cells in the endometrium of women with recurrent miscarriage. Hum Reprod 2007; 22: 2208-13.
  13. Quenby S, Kalumbi C, Bates M, et al. Prednisolone reduces preconceptual endometrial natural killer cells in women with recurrent miscarriage. Fertil Steril 2005; 84: 980-4.14
  14. Thum M, Bhaskaran S, Abdalla H, et al. Prednisolone suppresses NK cell cytotoxicity in vitro in women with a history of infertility and elevated NK cell cytotoxicity. Am J Reprod Immunol 2008; 59: 259-65.
  15. Park-Wyllie L, Mazzotta P, Pastuszak A, et al. Birth defects after maternal exposure to corticosteroids: Prospective cohort study and meta-analysis of epidemiological studies. Teratology 2000; 62: 385-392.
  16. Gur C, Diav-Citrin O, Shechtman S, et al. 2004. Pregnancy outcome after first-trimester exposure to corticosteroids: a prospective controlled study. Reprod Toxicol 2004; 18: 93-101.
  17. Laskin C, Bombardier C, Hannah M, et al. Prednisone and aspirin in women with autoantibodies and unexplained recurrent fetal loss. New Engl J Med 1997; 337: 148-154.
  18. Michael A, Papageorghiou A. Potential significance of physiological and pharmacological glucocorticoids in early pregnancy. Human Reprod Update 2008; 14: 497-517.

Leave a Reply

Your email address will not be published. Required fields are marked *