1. Aetna considers the following tests medically necessary for evaluation of members with recurrent pregnancy loss (defined as 2 or more consecutive spontaneous abortions):

   1. Endometrial biopsies for evaluation of luteal phase defect;
   2. Hysterosalpingography, hysteroscopy or sonohysteroscopy/sonohysterography to diagnose uterine anatomic abnormalities;
   3. Karyotype (cytogenetic analysis) of parents to detect balanced chromosomal anomalies;
   4. Karyotype of abortus tissue when a couple with recurrent pregnancy loss experiences a subsequent spontaneous abortion;
   5. Measurement of anti-beta2-glycoprotein I (IgG or IgM) antibodies, anti-cardiolipin (IgG or IgM) antibodies, and lupus anticoagulant, using standard assays, for diagnosis of antiphospholipid syndrome;
   6. Pelvic ultrasound scan to assess ovarian morphology and the uterine cavity;
   7. Prenatal genetic diagnosis for all couples in which 1 partner has been found to have a balanced translocation or inversion;
   8. Tests for inherited thrombophilic disorders: factor V Leiden, prothrombin G20210A mutation, serum homocysteine, and deficiencies of the anticoagulants protein C, protein S, and antithrombin II;
   9. Tests for thyroid stimulating hormone (TSH) and thyroid antibodies.

2. Aetna considers any of the following tests/studies experimental and investigational because they have been shown to be of no value in the evaluation of recurrent pregnancy loss:

   1. Antibodies to phosphatidylserine, phosphatidylethanolamine, or other anti-phospholipid antibodies other than anti-cardiolipin and lupus anticoagulant;
   2. Determination of the percentage of circulating natural killer (NK) cells;
   3. Embryo toxicity assay (ETA);
   4. Genetic association studies of inflammatory cytokine polymorphisms;
   5. Inter-α trypsin inhibitor-heavy chain 4 (ITI-H4) (as a biomarker for recurrent pregnancy loss);
   6. Luteal phase biopsy to determine the status of natural killer (NK)-like cells;
   7. Maternal antiparental antibodies;
   8. Parental human leukocyte antigen (HLA) status;
   9. Pre-implantation genetic screening;
   10. Reproductive immunophenotype (CD3+, CD4+, CD5+, CD8+, CD16+, CD19+, CD56+)
   11. Tests for embryotoxic factor;
   12. Tests for maternal antileukocytic antibodies to paternal leukocytes;
   13. Tests for serum “blocking factor”;
   14. Cytokine polymorphisms analysis (Th1/Th2 intra-cellular cytokine ratio);
   15. X-chromosome inactivation study.

3. Aetna considers any of the following treatments experimental and investigational for recurrent pregnancy loss because they have not been shown to be effective for that indication:

   1. Intravenous immunoglobulin (IVIG) therapy;
   2. Leukocyte immunization (immunizing the female partner with the male partner's leukocytes);
   3. Low-molecular-weight heparin (however, LMWH may be medically necessary in pregnant women with thrombophilic disorders, and for treatment in pregnant women with venous thrombo-embolism - see CPB 0346 - Low-Molecular-Weight Heparins).

This policy is based on the recommendations of the American College of Obstetricians and Gynecologists (ACOG, 2001) and the Royal College of Obstetricians and Gynaecologists (RCOG, 2001).

The ACOG guideline Management of Recurrent Early Pregnancy Loss reached the following conclusions: "Women with recurrent pregnancy loss should be tested for lupus anticoagulant and anticardiolipin antibodies using standard assays.  If test results are positive for the same antibody on two consecutive occasions 6-8 weeks apart, the patients should be treated with heparin and low-dose aspirin during her next pregnancy attempt. Mononuclear cell (leukocyte) immunization and IVIG are not effective in preventing recurrent pregnancy loss" (ACOG, 2001).

An association between the luteal phase defect and recurrent pregnancy loss is controversial.  If a diagnosis of luteal phase defect is sought in a woman with recurrent pregnancy loss, it should be confirmed by endometrial biopsy.  Luteal phase support with progesterone is of unproven efficacy.

Couples with recurrent pregnancy loss should be tested for parenteral balanced chromosome abnormalities.  Women with recurrent pregnancy loss and a uterine septum should undergo hysteroscopic evaluation and resection.  Cultures for bacteria and viruses and tests for glucose tolerance, thyroid abnormalities, antibodies to infectious agents, anti-nuclear antibodies, anti-thyroid antibodies, paternal human leukocyte antigen status, or maternal anti-parental antibodies are not beneficial and, therefore, are not recommended in the evaluation of otherwise normal women with recurrent pregnancy loss.  Couples with otherwise unexplained recurrent pregnancy loss should be counseled regarding the potential for successful pregnancy without treatment.

The Royal College of Obstetricians and Gynaecologists (RCOG) Guidelines on Management of Recurrent Miscarriage (2001) are consistent with ACOG Guidelines.  The RCOG recommends the following work-up for recurrent pregnancy loss:

• A pelvic ultrasound scan to assess ovarian morphology and the uterine cavity  
• Karyotyping of all fetal products 
• Peripheral blood karyotyping in both partners
• Screening tests for anti-phospholipid antibodies (both the lupus anticoagulant and anti-cardiolipin antibodies) performed on 2 separate occasions at least 6 weeks apart.  Discordant results should prompt the performance of a 3rd test. 

The RCOG guidelines conclude that "the place of all other investigations including a search for newly described thrombophilic defects is unproven and such tests should only be performed in the context of research studies."

The American College of Obstetricians and Gynecologists (2001) state that tests for thrombophilias are not required as part of the evaluation of recurrent pregnancy loss, but may be considered in cases of otherwise unexplained fetal death in the 2nd or 3rd trimesters.  "The role of thrombophilia in recurrent pregnancy loss is a controversial subject of current research interests.  Tests for factor V leiden, the prothrombin G20210A mutations, or deficiencies of protein C, protein S, or antithrombin III should be considered in cases of otherwise unexplained fetal death in the second or third trimesters.  However, the role of these heritable thrombophilias in recurrent early pregnancy loss is uncertain at present, and tests for these thrombophilias are not required as part of the evaluation.  Whether antithrombotic treatment improves subsequent pregnancy outcomes in women with evidence of thrombophilia is uncertain."

Investigators have also found evidence of significantly higher serum homocysteine levels among women with a history of recurrent miscarriage (Krabbendam et al, 2005; Hague, 2003).  Routine folate supplementation is recommended during pregnancy to prevent neural tube defects (USPSTF, 2006).  This supplementation should also reduce serum concentrations of homocysteine that may be associated with recurrent pregnancy loss.

A systematic evidence review found insufficent evidence for plasminogen activator inhibitor 4G/5G polymorphism testing in recurrent miscarriage (Augustovski et al, 2006).

The RCOG recommends that in women with recurrent miscarriage who have undergone the above investigations should undergo the following management:

• That all future treatment options are evaluated in randomized controlled trials;
• That treatments of unproven benefit should be abandoned;
• That women with persistently positive tests for anti-phospholipid antibodies are offered treatment with low dose aspirin together with low dose heparin during pregnancy (also the subject of on-going research);
• Those with karyotypic abnormalities should be seen by a clinical geneticist.

Recurrent spontaneous abortion (habitual abortion or miscarriage) is defined as at least 2 or 3 spontaneous abortions prior to 20 weeks gestational age with the same partner.  Two types of immunotherapy have been explored: (i) injections of paternal leukocytes (paternal white blood cell immunization or paternal cell alloimmunization) and (ii) the use of intravenous immunoglobulin (IVIG).

A meta-analysis of randomized controlled trials of immunotherapy for recurrent miscarriage concluded that IVIG and paternal leukocyte injections provided no significant beneficial effect over placebo in preventing further miscarriages (Porter et al, 2006).  The investigators also found no significant benefit for other immunological treatments that have been used for recurrent miscarriage: third party donor cell immunization and trophoblast membrane infusion.  The authors of the meta-analysis concluded that it is questionable whether paternal leukocyte injection is an effective treatment for recurrent miscarriage.  They also concluded that third party donor leukocytes, trophoblast membranes, and IVIGs appear to provide no significant beneficial effect over placebo in preventing further miscarriages.

An American Society for Reproductive Medicine (2002) Committee Opinion concluded: "IVIG as a treatment for recurrent pregnancy loss should be evaluated in patients who are informed, consenting participants in an institutional review board approved randomized clinical trial.  For the management of recurrent spontaneous pregnancy loss IVIG is an experimental treatment."

The University Health Consortium (1999) guidelines on use of immunoglobulin preparations concluded that the use of IVIG for recurrent pregnancy loss is "not recommended."

Embryo toxicity assay (ETA) is a laboratory test performed on a woman who has had recurrent early pregnancy loss.  A blood sample from the woman is used to furnish a culture medium for growing mouse embryos.  The culture is then examined under microscopy to determine if there are any circulating factors in the blood specimen that are toxic to the developing mouse embryos.  There is a lack of adequate evidence in the peer-reviewed published medical literature on the effectiveness of this test in improving clinical outcomes.

The Practice Committee of the American Society for Reproductive Medicine (2004) concluded that the use of IVIG for the management of recurrent spontaneous pregnancy loss is an experimental treatment.

Stephenson and colleagues (2010) conducted a multi-centered, randomized, double-blinded, placebo-controlled trial comparing IVIG with saline in women with idiopathic secondary recurrent miscarriage, defined as a history of at least 1 prior ongoing pregnancy followed by 3 or more consecutive unexplained miscarriages.  Subjects received either IVIG 500 mg/kg body weight or the equivalent volume of normal saline.  Pre-conception infusions were administered 14 to 21 days from the projected next menstrual period.  With documentation of pregnancy, the subject received the same infusion every 4 weeks until 18 to 20 weeks of gestation.  The primary outcome was an ongoing pregnancy of at least 20 weeks of gestation.  A total of 82 patients enrolled, of whom 47 had an index pregnancy.  All ongoing pregnancies resulted in live births.  Thus, the live birth rates were 70 % (16/23) in the IVIG group and 63 % (15/24) in the control group (p = 0.760); odds ratio (OR) 1.37 [95 % confidence interval (CI): 0.41 to 4.61].  Including only clinical pregnancies (embryo with cardiac activity at 6 weeks of gestation), the live birth rates were equivalent, 94 % (16/17) and 94 % (15/16), respectively (p > 0.999); OR 1.07 (95 % CI: 0.06 to 18.62).  Meta-analysis of randomized controlled trials (RCTs) evaluating IVIG for idiopathic secondary recurrent miscarriage revealed live birth rates of 70 % (31/44) in the IVIG group and 62 % (28/45) in the control group (p = 0.503); common OR 1.44 (95 % CI: 0.59 to 3.48).  The authors concluded that this is the largest RCT to date in which IVIG was evaluated in women with idiopathic secondary recurrent miscarriage; no treatment benefit was found.  The meta-analysis, which combined these results with 2 prior RCTs, also showed no significant effect of treatment with IVIG.

In a review on genetics for recurrent pregnancy loss, Sierra and Stephenson (2006) stated that recent research has generated interest in genetic markers for recurrent pregnancy loss such as skewed X-chromosome inactivation and human leukocyte antigen-G polymorphisms.  Assisted reproductive technologies (specifically, pre-implantation genetic diagnosis) have been offered to couples with recurrent pregnancy loss; however, more research is needed before routine use of these new approaches can be advocated.

Stephenson and Kutteh (2007) stated that recurrent pregnancy loss affects up to 5 % of couples trying to establish a family.  Evaluation classically begins after 3 consecutive miscarriages of less than 10 weeks of gestation, but may be warranted earlier if a prior miscarriage was found to be euploid, or if there is concomitant infertility and/or advancing maternal age.  The evaluation begins with an extensive review of medical history and thorough physical examination, followed by a diagnostic screening protocol.  The authors noted that management must be evidence-based; unproven treatments should be avoided.

An American Society for Reproductive Medicine Practice Opinion (ASRM, 2007) concluded that "[a]vailable evidence does not support the use of PGS [preimplantation genetic screening] as currently performed to improve live-birth rates in patients with recurrent pregnancy loss."  This is re-affirmed by the ACOG's Committee Opinion on PGS (2009), which stated that there are no data to support PGS for recurrent unexplained miscarriage and recurrent implantation failures; its use for these indications should be restricted to research studies with appropriate informed consent.

Bombell and McGuire (2008) noted that inflammatory cytokine cascades have been implicated in the pathogenesis of recurrent pregnancy loss (RPL).  Polymorphisms in cytokine genes may affect the risk of RPL, but genetic association studies are often limited by small sample sizes.  Meta-analysis of all available studies can increase the precision of these estimates.  These researchers evaluated and synthesized the available data from association studies of inflammatory cytokine polymorphisms with RPL.  A total of 16 reports of genetic association studies of cytokine polymorphisms with RPL were identified.  Meta-analyses did not identify any significant associations with tumour necrosis factor (-308A, or -238A), interferon-gamma (+874T), interleukin (IL)-1beta (-511T), IL-6 (-174G), or IL-10 (-1082A, or -819T, or -592A).  Significant associations were found with IL-1B (-31T) (2 studies: pooled OR 2.12 (95 % CI: 1.04 to 4.33)) and IL-6 (-634G) (1 study: OR 0.22 (95 % CI: 0.09 to 0.57)).  The authors concluded that available data are inconsistent with more than modest associations between these candidate cytokine polymorphisms and RPL.  They stated that data from future association studies may be added to the meta-analyses to obtain more precise estimates of effect sizes.

Choi and Kwak-Kim (2008) reviewed cytokine gene polymorphism studies in women with recurrent spontaneous abortion (RSA) to provide comprehensive understanding and a direction for the future investigation.  A search of PubMed was made to identify the published data between 2001 and 2007 regarding RSA and cytokine gene polymorphisms.  Either allele and/or genotype frequencies of the following polymorphisms were reported to be significantly different between women with RSA and controls: IFN-gamma +874A-->T, TA (p = 0.01), AA (p = 0.04); IL-6, -634C-->G CG/GG (p = 0.026); IL-10, -592C-->A CC (p = 0.016); IL-1B -511C (p = 0.035), -31T (p = 0.029); IL-1RA, IL1RN*2 (p = 0.002), and IL1RN*3 (p = 0.002).  None of these studies was repeatedly reported by others to be significantly different.  Among these, 4 cytokine polymorphisms (IFN-gamma, +874A-->T; IL-1B -511C; IL-1RA, IL1RN*2, IL1RN*3) were refuted by others and rest of them were studied once.  The authors concluded that multiple cytokine polymorphisms were reported to be associated with RSA.  However, a majority of studies were not confirmed by other investigators or refuted by others.  Inconsistent study results might be related to: (i) the production of these cytokines is partly under genetic controls and other factors affect cytokine levels; (ii) ethnic background, environmental factors, and selection criteria for study populations are different; and (iii) the possibilities exist that multiple cytokine gene polymorphisms or other genes in linkage disequilibrium may play a role in RSA.

Laskin and colleagues (2009) compared live birth rates in women with RPL and either autoantibodies or a coagulation abnormality, treated with low molecular weight heparin plus aspirin (LMWH/ASA) or aspirin (ASA) alone, and placed the results in context with other RCTs with similar cohorts.  The HepASA Trial was an RCT including patients with a history of RPL and at least 1 of the following: anti-phospholipid antibody (aPL), an inherited thrombophilia, or antinuclear antibody.  Treatment groups were stratified by aPL status and history of early versus late pregnancy losses.  Patients received either LMWH/ASA or ASA alone.  The primary outcome was live birth; secondary outcomes included adverse events and bone loss at the spine and femoral neck.  Literature over the past 20 years was reviewed to identify comparable RCT.  Over 4 years, a total of 859 women with RPL were screened: 88 (10.2 %) fulfilled inclusion criteria, became pregnant and were randomized to receive either LMWH/ASA or ASA alone.  Anti-phospholipid antibody were present in 42 (47.7 %) patients in each group.  The trial was stopped after 4 years when an interim analysis showed no difference in live birth rates in the 2 groups, and a lower rate of pregnancy loss in the ASA only group than expected.  In the LMWH/ASA group, 35/45 (77.8 %) had a live birth versus 34/43 (79.1 %) in the ASA only group (p = 0.71).  Neither number of prior losses nor aPL status was correlated with pregnancy outcome.  There were no cases of pregnancy related thrombosis in either group.  Mean change in bone mineral density did not differ by treatment group at either the lumbar spine (p = 0.57) or femoral neck (p = 0.15).  Randomized controlled trials since 2000 for aPL-positive women with RPL and similar inclusion criteria reported a mean live birth rate of 75 % with either LMWH or ASA.  The authors concluded that LMWH/ASA did not confer incremental benefit compared to ASA alone for this population.  Regardless of treatment regimen, number of prior losses, or aPL positivity, almost 80 % of women in the RPL cohort had a successful pregnancy outcome.  These findings contribute to a growing body of literature that contests the emerging standard of care comprising LMWH/ASA for this population.

Kim and colleagues (2011) stated that there are several etiological factors associated with immunology, anatomy, endocrinology, genetic, infection, chromosomal abnormalities, and environmental factors contributing to RPL.  The aim of this study was to identify RPL associated factors in human blood using proteomics.  Since it is difficult to obtain tissues or follicular fluids, these researchers used blood samples from normal and RPL patients to conduct a comparative proteomic study.  Three RPL blood samples and 1 cocktailed blood sample from 3 normal women were used.  These investigators performed 2-DE and selected spots were analyzed with MALDI-TOF/MS.  In the 3 RPL blood samples, 2-DE analysis revealed 549, 563 and 533 spots to be expressed differentially, respectively.  Through a comparative analysis between the control and RPL, 21 spots were shown to be differentially expressed.  Of these, 5 proteins were confirmed by Western blot analysis.  One of these proteins, inter-α trypsin inhibitor-heavy chain 4 (ITI-H4), was weakly expressed at a molecular weight of 120 kDa, but was highly expressed at a modified molecular weight of 36 kDa in RPL patients.  These findings suggested that ITI-H4 expression may be used as a biomarker, which could facilitate the development of novel diagnostic and therapeutic tools.

The ACOG's practice bulletin on "anti-phospholipid syndrome" (2011) stated that obstetric indications for anti-phospholipid antibody testing (anti-beta2-glycoprotein I of immunoglobulin G [IgG] and/or immunoglobulin M isotype, anti-cardiolipin antibody, and lupus anti-coagulant) should be limited to a history of 1 fetal loss or 3 or more recurrent embryonic or fetal losses.

Darmochwal-Kolarz et al (2002) estimated the alterations in the phenotype of lymphocytes of women with unexplained pregnancy failures in comparison with healthy women.  A total of 14 women with unexplained habitual miscarriages and 18 healthy, fertile women with the history of successful pregnancies were included in the study.  The lymphocytes were isolated from peripheral blood and stained with monoclonal antibodies.  The expression of selected surface molecules was estimated using the flow cytometric method.  These researchers found that the percentage of T CD4(+) lymphocytes, CD3(-)16/56(+) cells, and T CD8(+)11b(-) cells was significantly higher in patients with recurrent pregnancy loss in comparison with healthy women.  The percentage of B-1 CD19(+)5(+) lymphocytes was also significantly higher in women with unexplained habitual miscarriages in comparison with healthy women.  Furthermore, they found higher expression of CD25 molecule on T CD3(+) and T CD4(+) lymphocytes in the study group, when compared to controls.  Moreover, the percentages of B CD19(+) and T suppressor CD8(+)11b(+) lymphocytes were lower in women with pregnancy failures in comparison with the control group.  The percentage of T CD3(+) lymphocytes and T CD8(+) cells did not differ in both studied groups.  Similarly, the expression of CD25 antigen and HLA-DR molecule on T CD8(+) did not differ in the study group, when compared to controls.  The authors concluded that these findings suggested that the immunological alterations may be involved in the etiopathogenesis of unexplained recurrent pregnancy loss.  The results of this small study need to be validated by well-designed studies.

Inflammatory cytokine cascades have been implicated in the pathogenesis of recurrent pregnancy loss.  Polymorphisms in cytokine genes may affect the risk of recurrent pregnancy loss, but genetic association studies are often limited by small sample sizes. Meta-analysis of all available studies can increase the precision of these estimates.  In a meta-analysis, Bombell et al (2008) assessed and synthesized the available data from association studies of inflammatory cytokine polymorphisms with recurrent pregnancy loss.  Systematic review and random effects meta-analysis of genetic association studies were performed.  A total of 16 reports of genetic association studies of cytokine polymorphisms with recurrent pregnancy loss were identified.  Meta-analyses did not identify any significant associations with tumor necrosis factor (-308A, or -238A), interferon-gamma (+874T), interleukin (IL)-1beta (-511T), IL-6 (-174G), or IL-10 (-1082A, or -819T, or -592A).  Significant associations were found with IL-1B (-31T) (2 studies: pooled odds ratio (OR) 2.12 (95 % CI: 1.04 to 4.33)) and IL-6 (-634G) (1 study: OR 0.22 (95 % CI: 0.09 to 0.57)).  The authors concluded that available data are not consistent with more than modest associations between these candidate cytokine polymorphisms and recurrent pregnancy loss.  Data from future association studies may be added to the meta-analyses to obtain more precise estimates of effect sizes.

Agrawal et al (2012) determined whether or not interleukin-1 alpha (IL-1α), interleukin-1 beta (IL-1β) and IL-1 receptor antagonist (IL-1RA) polymorphisms are associated with risk of unexplained recurrent pregnancy loss among North Indian women.  This retrospective case-control study examined 200 well-characterized recurrent pregnancy loss cases for IL-1 gene cluster variants, determined by restriction fragment length polymorphism-PCR.  The observed allele, genotype and haplotype distributions were compared with those obtained from 300 ethnically-matched negative controls.  Invariant distribution of IL-1 gene cluster single-nucleotide polymorphisms was observed among recurrent pregnancy loss cases and controls.  Meta-analysis of IL-1β -511, +3953 and IL-1RN 86-bp variable number tandem repeat from the reported literature and this study did not reveal any significant association with the risk of recurrent pregnancy loss.  The authors concluded that no significant difference between recurrent pregnancy loss and control groups was observed at the allele, genotype or haplotype levels when tested for association using the dominant, recessive and additive models of inheritance for IL-1 gene cluster variants.  The authors noted that this is the first report from India pertaining to IL-1 gene cluster variants' association with the risk of recurrent pregnancy loss from North India.  One of the most unfortunate complications of incomplete motherhood is recurrent pregnancy loss particularly of unknown etiology.  Both genetic and environmental factors may play a role in the maintenance of pregnancy.  From a traditional immunological perspective, survival of the semi-allogenic fetus is dependent on suppression of the maternal immune response.  However, the function of immune cells changes during pregnancy, no generalized suppression of the maternal immune response has been recorded.  Cytokines in the interleukin-1 system (IL-1α, IL-1β and receptor antagonist (IL-1RA)) are produced at the fetal-maternal interface during early pregnancy and are believed to influence the immune responsiveness of Th1/Th2 cells and has been implicated in the establishment of successful pregnancy.  This study examined IL-1 gene cluster variants among well-characterized recurrent pregnancy loss patients and compared them with those of healthy controls.  Interleukin-1 gene cluster single-nucleotide polymorphisms were found to be invariably distributed among recurrent pregnancy loss patients and controls.  Meta-analysis of results of IL-1β -511, IL-1β +3953 and IL-1RN 86-bp variable number tandem repeat studies in recurrent pregnancy loss, as reported previously, also did not show any effect of these variations on the risk of occurrence of the disease.

Rao et al (2008) determined the frequency of hypothyroidism in women with RPL in first trimester in the Indian population.  The study included 163 non-pregnant women with RPL in a gestational age up to less than or equal to 12 weeks verified by a pregnancy test or ultrasonography, and a total of 170 age-matched women with at least 1 successful pregnancy and no history of miscarriages were selected as controls.  Levels of thyroid hormones triiodothyronine (T3), thyroxine (T4) and thyroid stimulating hormone (TSH) were estimated in non-pregnant women with RPL and controls.  Hypothyroidism was found in 7 (4.12 %) women with RPL and 1 in control group.  The differences in the levels of serum T3, T4 and TSH between euthyroid and hypothyroid women were found significant in women with RPL in first trimester.  The authors concluded that the findings of this study demonstrated that hypothyroidism has a statistically significant relationship with RPL in the first trimester and suggested that diagnosis of hypothyroidism could help couples with RPL to have a successful outcome in subsequent pregnancies.

In a single-center, retrospective study, Jaslow and colleagues (2010) examined if the frequency of abnormal results for evidence-based diagnostic tests differed among women with RPL based on the number of prior losses (n = 2, 3, or more) and determine whether abnormal results for additional investigative diagnostic tests differed in prevalence among women with different numbers of pregnancy losses.  A total of 1,020 women who had 2 or more consecutive spontaneous pregnancy losses with the same partner were included in this analysis.  Main outcome measures included frequencies of abnormal results for evidence-based diagnostic tests considered definite or probable causes of RPL (karyotyping for parental chromosomal abnormalities; pelvic sonohysterography, hysterosalpingogram, or hysteroscopy for uterine anomalies; immunological tests for lupus anti-coagulant and anti-cardiolipin antibodies; thrombophilic tests for the factor V Leiden mutation; and blood tests for TSH and fasting blood glucose).  These investigators also measured the frequency of abnormal results for 9 additional investigative tests in the same patients (antiphosphatidyl serine antibodies, microbial infection, mid-luteal progesterone, prolactin, functional protein C activity, functional protein S activity, anti-thrombin activity, fasting homocysteine and methylenetetrahydrofolate reductase polymorphisms, and factor II mutation).  The prevalence of abnormal results for evidence-based and investigative diagnostic tests did not differ among women with different numbers of pregnancy losses.  The authors concluded that evaluation of all couples with 2, 3, or more consecutive miscarriages is recommended.

In a case-control study, Ticconi et al (2011) examined the role of anti-thyroid autoantibodies (ATA) in recurrent miscarriage (RM).  A total of 160 women with RM and 100 healthy women were investigated for the presence of serum ATA directed against thyroglobulin (TG-Ab), thyroid peroxidase (TPO-Ab) and TSH receptor (TSHr-Ab), which were determined by either chemiluminescence or radioimmunoassay.  Anti-thyroid autoantibodies were detected in 46 (28.75 %) women with RM and in 13 (13 %) women of the control group (p < 0.05).  The frequencies for TG-Ab and TPO-Ab were higher in RM than in control women.  Among the women of RM group, 91.3 % of ATA+ women were positive also for other autoantibodies.  The majority of study women were euthyroid.  The authors concluded that anti-thyroid autoantibodies, particularly TG-Ab, are associated with RM and could be an expression of a more general maternal immune system abnormality leading to RM.  They stated that ATA could have a role in RM irrespective of thyroid hormone status. 

van den Boogaard et al (2011) conducted a systematic review of the literature on the clinical significance of thyroid dysfunction and thyroid autoimmunity before conception and in early pregnancy.  Relevant studies were identified by searching Medline, EMBASE and the Cochrane Controlled Trials Register.  From a total of 14,208 primary selected titles, 43 articles were included for the systematic review and 38 were appropriate for meta-analyses.  No articles about hyperthyroidism were selected.  Sub-clinical hypothyroidism in early pregnancy, compared with normal thyroid function, was associated with the occurrence of pre-eclampsia [OR 1.7, 95 % CI: 1.1 to 2.6] and an increased risk of perinatal mortality (OR 2.7, 95 % CI: 1.6 to 4.7).  In the meta-analyses, the presence of thyroid antibodies was associated with an increased risk of unexplained subfertility (OR 1.5, 95 % CI: 1.1 to 2.0), miscarriage (OR 3.73, 95 % CI: 1.8 to 7.6), recurrent miscarriage (OR 2.3, 95 % CI: 1.5 to 3.5), preterm birth (OR 1.9, 95 % CI: 1.1 to 3.5) and maternal post-partum thyroiditis (OR 11.5, 95 % CI: 5.6 to 24) when compared with the absence of thyroid antibodies.  The authors concluded that pregnant women with subclinical hypothyroidism or thyroid antibodies have an increased risk of complications, especially pre-eclampsia, perinatal mortality and (recurrent) miscarriage.

The Practice Committee of the ASRM (2012) recommended screening for thyroid or prolactin abnormalities for the evaluation and treatment of RPL.

An UpToDate review on “Evaluation of couples with recurrent pregnancy loss” (Tulandi and Al-Fozan, 2013) states that “Thyroid function should be assessed in women with clinical manifestations or a personal history of thyroid disease.  Screening asymptomatic women for subclinical thyroid dysfunction is controversial.  We feel screening is reasonable since there is evidence of an increased risk of miscarriage in women with subclinical hypothyroidism and in euthyroid women with thyroid peroxidase (TPO) antibodies.  Meta-analyses of case-control and cohort studies have found that the presence of TPO autoantibodies in euthyroid women is associated with an increased risk of spontaneous miscarriage that is two to three times higher than in women without these antibodies.  In addition, meta-analysis of two randomized trials of the effect of thyroid replacement in these women found treatment was associated with a significant reduction in risk of miscarriage (relative risk 0.48, 95 % CI: 0.25 to 0.92); however, there were methodological limitations to these trials”.   Moreover, an UpToDate review on “Evaluation of couples with recurrent pregnancy loss” (Tulandi and Al-Fozan, 2013) does not mention measurement of natural killer cells in the evaluation of RPL.