Multiple Pregnancy Failures: An Immunological Paradigm



Leif Matthiesen, Department Obstetrics and Gynecology, Helsingborg Hospital, 251 87 Helsingborg, Sweden



Recurrent spontaneous abortion (RSA), three or more pregnancy losses prior to 20 weeks, occurs in about 1% of all pregnancies, 50% of RSA cases remain unexplained and unresolved. Recently, immune pathways have been implicated in the pathophysiology of RSA. Immune tolerance of the fetal–placental unit and placental angiogenesis are mandatory for a successful pregnancy outcome. Unscheduled dysregulation of the placental vasculature is thought to be the pathophysiologic mechanisms underlying an array of pregnancy complications like infertility, miscarriage, pre-eclampsia, and fetal growth restriction and death. Investigations on mechanisms and management of RSA are mired by substandard design and lack of optimal randomized clinical trials and have resulted in disagreement on guidelines for evaluation and treatments for patients with multiple pregnancy losses of unknown etiology. The present review focuses on evidence-based research discussion with immunologic causes, and immune-regulatory therapies recommended for helping patients with a history of RSA. We highlight data that might support revalidation of low molecular weight heparin as a protective therapy in RSA. Newly launched growth factors, GM-CSF, and potentially novel agents to suppress inflammatory rejection, including regulatory T cells, human chorionic gonadotropin, and M-CSF/IL-10, may work in concert with tender-loving-care therapy and give hope to couples with multiple pregnancy losses.


Multiple pregnancy losses or recurrent spontaneous abortion (RSA), classified as three or more losses in clinically identified pregnancies prior to 20 weeks, occurs in about 1% of all pregnancies.[1] Both the heterogeneous etiologies and their respective prevalences are disputed.[2-5] Some conditions have been associated with high risk of pregnancy losses like presence of the antiphospholipid syndrome [6] and parental chromosomal abnormalities.[7] Other conditions including acquired or inherited uterine anatomical abnormalities,[8, 9] endocrine imbalance,[10, 11] thrombophilias,[12] and immunological factors [13-17] have a more complex influence on early pregnancy wastage (Fig. 1). The proportion of these factors that contribute to RSA differs in the reports owing to variations in design of investigating the patients suffering multiple miscarriages.[18, 19] After thorough evaluations, approximately 50% of couples with RSA remain unresolved and no probable explanation is disclosed leaving you without suggestive specific therapeutical approaches against a cause.[8, 20, 21]

Figure 1.

History, age, and pregnancy outcome. The figure illustrates the likely hood of miscarriages based on the history of prior abortions and age.

Implantation and angiogenesis

Immune tolerance of the allogeneic fetus is mandatory for a successful pregnancy outcome.[14] Both innate and adapted immune responses contribute to a cross-talk with the fetal–placental unit to create a balance between anti- and pro-inflammatory decidual environments.[22-24] The maternal blood is in direct contact with syncytiotrophoblasts at the intervillous space and in the decidual tissue where the extravillous cytotrophoblasts anchor the placenta and further differentiate into endovascular cytotrophoblasts that invade spiral arteries to remodel their endothelial lining.[22, 25, 26] In normal pregnancies, the remodeling of the maternal uterine arteries allows for a dramatic increase in blood flow needed to support the fetal growth. Dysregulation of the placental vasculature (anti-angiogenesis) is thought to be the pathophysiologic mechanism underlying an array of pregnancy complications like infertility, miscarriage, pre-eclampsia, fetal growth restriction, and death.[22, 27, 28]

A successful pregnancy outcome is also supported by particular MHC class I molecules, hormones, complement regulatory proteins, several immunoregulatory molecules (indolamine 2,3-dioxygenase, Fas:FasLigand), regulatory T cells (CD4+ CD25+ Foxp3+), IL-10, regulatory macrophages, and growth factors expressed at the placental–decidual interface.[9, 11, 13, 15, 17, 24, 29, 30] These mechanisms act in concert to maintain tolerance of the placenta and fetus and strengthen the angiogenesis at the placental bed.[31]

The management of recurrent miscarriages of unknown etiology by prophylactic use of anticoagulants, low-dose aspirin (ASA) or low molecular weight heparin (LMWH), is based on the immunopathologic evaluations showing concomitant inflammation and thromboembolism in the placenta and the decidua.[32] Pre-clinical and clinical research aiming to identify immunological readouts to explain recurrent miscarriages and to aid in clinical practice is still in infancy. This is further compounded by huge variations in clinical protocols and poor design including insufficient number of patients and outdated and overspread inclusion criteria.[33-40] Indeed, attempts to perform a meta-analysis failed because of ‘heterogeneity’ across studies both from statistical and clinical perspectives, which made them inappropriate to create a pooled risk ratio across studies. Also, none of the studies were blinded or controlled by a placebo-equivalent for LMWH injection.[41] Finally, so far no study has been designed with a primary focus on a biological entity that scientific ASA or LMWH might target (Table 1).

Table 1. Research on the use of Low Molecular Weight Heparin in Recurrent Spontaneous Miscarriage
Potentially efficient
But limited evidence
Open-randomized studies dominate the literature
Historical controls
‘Mixed’ populations
Small groups
Lack of blinded randomized studies
Lack of placebo-equivalent to low molecular weight heparin
Lack of pre-clinical studies describing mechanisms

Pre-clinical investigation on proposed interplay between inflammation and coagulation in miscarriage suffers similar obstacles, probably due to the variety of intrinsic factors underlying pregnancy loss.[8, 20, 21] Identification of an immune cause for recurrent miscarriage is hampered by the sparse documentation of early immunologic events in human pregnancy. Currently, our knowledge is derived from animal studies, often in knockout models where deficiency in a single molecule is required to impart adverse activity of normal pathways.[13, 14] The role of cytokine dysregulation in humans as a cause for RSA is controversial either because interpretation of the data is based on conclusions drawn from in vitro studies with tissue samples taken after the fetal demise or because investigations are based on peripheral blood lymphocytes responding in vitro to trophoblast antigens [42] (Table 2).

Table 2. Obstacles in Clinical Research of Mechanisms, Guidelines, and Treatment Protocols in Recurrent Spontaneous Miscarriages
Sparse documentation on early implantation in human pregnancy
Knowledge rely on animal studies
Variations among study designs
Pioneering literature and treatment protocols obsolete – but remain used
Peripheral reflections of dysregulation at the maternal–fetal interface and vice versa
Lack of consensus on definition of RSM
Guidelines vary – too many!
Definitions vary
Number of miscarriages vary
Consecutive versus non-consecutive RSM
Too many tests
Lack of evidence-based recommendations
Too many ineffective treatment protocols

Guidelines on the evaluation of multiple pregnancy losses

Guidelines on RSA are poorly implemented in clinical practice. Obstacles for the diagnosis of RSA are disagreement on the number of recurrent miscarriages and whether they have to be consecutive or not; biochemical pregnancies as established by fetal heart activity detected by ultrasound; very early or late gestational age at the time of miscarriage; complete lack of randomized clinical trials on necessary tests to be carried out; and finally treatment protocols. Established high-risk factors are maternal age > 40 years, parental chromosomal abnormalities, and antiphospholipid syndrome, respectively.[3, 4, 18-21]

Hypothetically, endocrine factors may have effects on pregnancy outcome. Luteal insufficiency associated with RSA includes low progesterone values; however, this might be the cause of an already aborted pregnancy more than the cause of it. The prevalence of thyroid dysfunction is similar in both the RSA group and the general population, which suggests that thyroid status is not critical. On the other hand, presence of thyroid autoantibodies has recently been shown to be associated with increased risk of RSA, but no RCT has been carried out showing an effect of levothyroxine on pregnancy outcome.[10, 11]

Evaluation of the uterine cavity is regarded as an important examination in repeated miscarriages, for example by hydrosonography, to identify the presence of congenital abnormalities or fibroids. In one study, it was shown that septate, bicornuate, or arcuate uterus was more often present in RSA than normal pregnant women. No RCT has been undertaken to show the influence of surgical correction or myomectomy in relation to RSA.[43, 44]

Increased risk of arterial or venous thromboembolism, thrombophilia, indicates the presence of acquired or inherited risk factors with a proven association with RSA. Combination of autoantibodies to cardiolipin or beta2 glycoprotein1 and the presence of lupus anticoagulant in RSA or late pregnancy complications like pre-eclampsia, growth retardation, and fetal death 12 weeks apart establish the diagnosis of antiphospholipid syndrome.[5, 6, 9, 45] The treatment with low-dose aspirin and LMWH is probably useful but the duration of therapy, dose, and use of effective drugs is not clear by placebo-controlled studies.[46]

A meta-analysis indicates that the physiologic pro-coagulant state in pregnancy increases the risk of thromboembolism in women with inherited thrombophilia. Mutation in factor V Leiden, prothrombin gene, low levels of anti-thrombin, and increased levels of protein C and –S may indicate a higher risk of RSA or intrauterine fetal death. However, as of now, no study has commented on the effect of anticoagulant treatment in women with inherited thrombophilia and RSA. The common understanding is that prophylactic LMWH treatment is recommended only to women with history of thromboembolism.[12, 47]

Establishing the gestational week at which the miscarriage take place is of importance. Recurrent early miscarriage during the first 6 weeks of pregnancy is more likely to be the result of chromosomal rearrangements.[48] In contrast, late first trimester miscarriage is likely to a result of abnormal trophoblast invasion resulting in defective angiogenesis in the placental bed.[49] In the latter case, it has been suggested that LMWH might have a beneficial effect owing to pro-angiogenesis.[8, 27, 50, 51]

Present and future treatment options for recurrent spontaneous miscarriage of unknown etiology

Evaluation of couples with RSA should only include tests that have proven benefits and support evidence-based treatment protocols or determinants. In as many as half of the cases of RSA, there is no consensus on the probable cause. For this group, none of the treatment through RCT has proven effective and the best approach may be the dedicated tender-love-care approach.

In the past, couples with unexplained RSA, a variety of immunotherapy, have been offered. A number of interventions have been proposed, many of them proven ineffective, and should be used only in experimental settings like RCT. Corticosteroids have been used owing to their immunosuppressive effects because of accepted phenomenon of pro-inflammatory condition and its association with miscarriage. However, this therapy in pregnancy increases the side effects of maternal diabetes and hypertension as well as the risk of premature rupture of membranes and delivery without improving pregnancy outcome.[52] Likewise, progesterone is in use because of its anti-inflammatory responses.[53] Although animal models of early pregnancy loss demonstrate mechanistic underpinnings, the results in humans are controversial at best.[54] High dose of pooled immunoglobulins (IVIg) given as intravenous infusion is used in rheumatology to block an array of inflammatory mechanisms believed as causes of diseases, but the potential immunomodulatory effects have never been well documented.[55] In addition, the therapy is expensive, invasive, time-consuming, and laced with serious risks.[56] The use of IVIg in RSA is fairly well documented but the results remain inconclusive to justify its use among these patients. Recently, a large RCT at the University of Chicago prematurely concluded its study owing to confusing nature of interim results, leaving the use of IVIg still a question.[57-59]

As mentioned above, the suggestion to use LMWH is based on the presence of inflammation and thrombosis and infarctions in the placenta and decidua of patients with both early and late pregnancy complications like RSA, intrauterine growth retardation, fetal death, and pre-eclampsia.[14, 16, 26, 28, 32] In this respect, Quenby et al.[50] showed that LMWH promoted differentiation of extravillous cytotrophoblasts with sustained migration capacity. Furthermore, Kalkunte et al. showed in a three-dimensional tube formation test using trophoblasts and endothelial cells on matrigel with addition of serum from pre-eclamptic patients that LMWH had the capacity to reverse and normalize disrupted angiogenesis.[27, 28, 31] We and others have shown that LMWH is pro-angiogenic and anti-inflammatory by blocking complement activation, antigen/antibody binding, adhesion and chemotaxis, and production of inflammatory cytokines.[9, 16, 27, 28] We hypothesize that early and late pregnancy complications may have similar pro-inflammatory pathophysiology leading to apoptosis, inhibition of trophoblast motility, and differentiation of trophoblast. The clinical picture of either early fetal demise or late pregnancy complications depends on the degree of inflammation and thereby damage of trophoblast. Therefore, it is urgent to investigate the potential beneficial effects of LMWH on trophoblast function to improve pregnancy outcome in high-risk pregnancies (Table 3).

Table 3. Effect of Low Molecular Weight Heparin in Reproduction
Increase Tissue Factor Pathway Inhibitor secretion from endothelium and trophoblasts and thereby blocking pro-inflammatory cytokines and activation of coagulation cascade
Protection against apoptosis of trophoblast
Increase expression of adhesion molecules, integrins, and matrix metalloproteinases
Anti-inflammatory by blocking production of IL-1, TNF-α, and IL-6
Anti-complement activation
Blocks antigen/antibody binding
Blocks adhesion and chemotaxis of pro-inflammatory cells

Recently, granulocyte–macrophage colony-stimulating factor (GM-CSF) has been launched to be used in the settings of in vitro fertilization (IVF). GM-CSF improves apposition, adhesion, differentiation, and invasion of the embryo and thereby optimal placental development.[60-62] Protection from apoptosis and control of the immune response in very early pregnancy improve IVF success. A defective upregulation of GM-CSF is reported in women with recurrent miscarriages with abnormal placentation.[63, 64] One advantage using GM-CSF is that it does not require invasive medication commonly used in RSA because it is used in the IVF protocols prior to embryo transfer.

Human chorionic gonadotropin (hCG) has been evaluated both as predictor of pregnancy outcome and also as therapeutic option.[65, 66] Recombinant hCG has shown promising results on in vitro studies compared with hCG extracted from pregnant mere′s serum on stabilizing corpus luteum, possibly due to its effect as a pro-angiogenetic hormone. In addition, hCG prevents T-cell activation, downregulates pro-inflammatory cytokines, and may recruit regulatory T cells (Tregs). hCG also stimulates corpus luteum progesterone production, decidualization, and inhibit uterine muscle contractions.[67] However, well-defined RCT are needed to include hCG as a therapeutic in patients with RSA.[68]Interestingly, several clinical trials are underway for the use of hCG on oocyte maturation as well as to study the pregnancy outcomes following IVF (

A novel challenging therapeutical approach is Tregs which is entering the clinic in graft versus host disease, type 1 diabetes, and kidney transplantation, although still phase I and II trials are needed.[69, 70] The therapeutical approach lies in immunosuppressive effects of these cells to avoid rejection reactions and act as regulators of self-tolerance. In pregnancy, locally expanded Tregs are fetus-specific to suppress abortion, and in contrast, in recurrent miscarriages and pre-eclampsia, a defect in function and number are seen.[29, 30, 70] Low levels of circulating CD4+ CD25+ Foxp3+ Tregs have been shown to predict miscarriage risk in newly pregnant women with a history of failure.[71] These observations suggest a role for Tregs treatment in RSA. Interestingly, a clinical trial is underway evaluates the effect of intrauterine injection of hCG before embryo transfer in IVF/ICSI on the implantation and pregnancy rates. This based on the rationale that intrauterine hCG injection will attract regulatory T cells and improve implantation ( Identifier: NCT01030393).

Several cellular compartments act during pregnancy to secure maternal tolerance of the allogeneic fetus and placenta. Decidual macrophages mature during early stages of pregnancy stimulated by progesterone and persist during rest of gestation.[72] Recently, Svensson et al. showed that macrophage colony-stimulating factor (M-CSF) and interleukin (IL)-10 act in concert to sustain this macrophage polarization during the entire length of pregnancy as cellular suppressors of inflammation inhibiting cytotoxic T cells and with a capacity to clear infection.[73-75] This could hypothetically open up an avenue for future development of an immunoregulatory therapy consisting of M-CSF and IL-10 that should be given to high-risk pregnancy patients (Table 4).

Table 4. Present and Possible Future Treatment Options in Recurrent Spontaneous Miscarriages
Low molecular weight heparin
Acetylsalicylic acid
Granulocyte–macrophage colony-stimulating factor
Human chorionic gonadotropin
Intravenous immunoglobulin
Lymphocyte Immunotherapy

Conclusion thoughts

Recurrent spontaneous miscarriage is a difficult condition in terms of its accurate diagnosis, and the mechanisms underlying the condition are complex but may represent a regulatory link with other pregnancy complications. Clinical studies on immune-mediated pregnancy loss are difficult to interpret because of substandard design. Trials on interventional therapy are restrained by too small index groups and lack of appropriate controls and limit the ability to make accurate conclusions. This simply tells us that pioneering research data on immune evaluation tests and immunomodulating therapies are still beyond our reach.

Despite this gloomy situation, the prognosis for patients with RSA is good. Our understanding of cellular and molecular interactions at the maternal–fetal interface is rapidly expanding and will surely give us an insight into the mechanisms of recurrent pregnancy losses and propel us to develop novel therapeutic approaches.