sign in sign up
<<
Home , Recurrent miscarriage

17 Recurrent loss

Rahat Khan, Vikram Talaulikar and Hassan Shehata

DEFINITION increases from 14–21% after one miscarriage to 24–29% after two and 31–33% Recurrent pregnancy loss (RPL) refers to the after three pregnancy losses. The minimum consecutive loss of three or more clinically diagnostic workup of couples experiencing recognized prior to the 20th week RPL consists of a complete medical, surgi- of gestation (excluding ectopic, molar and bio- cal, genetic and family history and a physical chemical pregnancies). RPL is classified into examination (see below). two categories: primary RPL, which consists General causes of RPL are shown in Table 1. of repeated in which a pregnancy has never been carried to viability; and second- ary RPL, in which a live birth has occurred at GENETIC FACTORS some time. Secondary RPL confers a better prognosis than primary RPL1. The highest rate of cytogenetically abnormal fetuses occurs earliest in gestation, with rates declining after the embryonic period (>30 mm INCIDENCE crown–rump length (CRL)).

About 10–15% of all clinically recognized preg- nancies end in miscarriage. Approximately Table 1 General causes of recurrent pregnancy 2% of women experience two and 0.4–1% of loss 2 women experience three consecutive losses . Causes Percentage (%) At less than 6 weeks’ gestation the risk of mis- Genetic factors – 3–5 carriage ranges from 22 to 57%, it declines to chromosomal abnormality 15% at 6–10 weeks and 2–3% after 10 weeks Primary miscarrier 7 of gestation3. (no live births) Secondary miscarrier 50 RISK FACTORS AND ETIOLOGY (1 or more live births) Anatomic causes 5–10 Couples with pregnancy loss usually express Immune mechanisms 50 concern regarding the cause and risk of recur- 10–13 rence. The risk of miscarriage increases with Endocrine 20 maternal age and parity, being 19% at less Infection 1 than 35 years and increasing to 47% in those Unexplained 15 over 35 years. In a similar fashion, the risk of

229 PRECONCEPTIONAL MEDICINE

Parental chromosomal abnormalities ANATOMIC CAUSES

In approximately 3–5% of couples with RPL, Acquired and congenital uterine abnormalities one of the partners carries a balanced struc- are responsible for 10–15% of RPL5 and may tural chromosomal anomaly (versus 0.7% of be associated with fetal growth restriction and the general population), the most common preterm delivery. being balanced reciprocal (60%) and Robert- sonian (40%) translocations. Uterine anomalies

Aneuploidy The most frequent uterine defects include sep- tate, bicornuate and didelphic abnormalities. The risk of (meiotic non-disjunc- The septate is most common and asso- tion, polypoid from fertilization abnormali- ciated with the poorest reproductive outcome ties) increases as the number of previous mis- (miscarriage rate more than 60% in untreated carriages increases. cases)6,7. Other anatomic causes of RPL are exposure related anomalies, Asherman’s syndrome, leiomyomas and endo- Other metrial polyps. A primary endometrial recep- tor defect appears to be responsible for RPL in receptor gene polymorphism some patients. may play a role in RPL and is an active area of investigation4. Maternal diseases including sickle cell anemia, myotonic dystrophy, Mar- Investigation and treatment fan’s syndrome, homocystinuria, factor VIII deficiency, dysfibrinogenemia and Ehler’s Dan- Transvaginal is useful for making los syndrome are all associated with increased a diagnosis of uterine anomalies8. Hystero­ fetal loss. scopy is usually reserved for patients in whom intrauterine pathology is suspected and opera- tive hysteroscopy is necessary. Transvaginal Investigations and treatment ultrasound assessment of the during pregnancy may be useful in predicting preterm Couples with a history of RPL should have birth in cases of suspected . peripheral blood karyotyping, and cytogenetic Magnetic resonance imaging (MRI) is use- analysis of the products of conception should ful for distinguishing between a septate and be performed if the next pregnancy fails. bicornuate uterus8. can provide the couple with a prognosis for future pregnancy, as well as offer familial chromosomal studies and Cervical incompetence appropriate preimplantation genetic diagnos- tic procedures in future pregnancies. In addi- No satisfactory objective test is available for tion, the couple should be informed that they cervical incompetence, and diagnosis is usu- have a 40–50% chance of a healthy live birth in ally made on the basis of a history of late mis- future untreated pregnancies following natu- carriages, preceded by spontaneous rupture of ral conception. membranes and painless cervical dilatation.

230 Recurrent pregnancy loss

Investigation and treatment apart for either lupus or anti- cardiolipin antibodies (aCL) of IgG and/or The Medical Research Council (MRC)/Royal IgM class present in medium or high titers. College of Obstetricians and Gynaecologists In detection of lupus anticoagulant, the dilute (RCOG) trial of elective cervical cerclage Russell’s viper venom (dRVVT) test is more reported a small decrease in preterm birth and sensitive and specific than the kaolin clotting delivery of very low birth weight babies, the time (KCT) or activated partial thromboplas- benefit being most marked in women with tin time (aPTT). three or more recurrent second trimester mis- carriages9. However, no significant improve- ment in perinatal survival was present. Treatment

Currently, several well controlled studies show IMMUNE MECHANISMS that future live birth is significantly improved Both autoimmune and alloimmune mecha- from 50% to 80% when a combination ther- nisms have been proposed as explanations for apy of low dose aspirin (75 mg) plus heparin RPL. (5000 U once or twice a day) is prescribed. A recent randomized trial reported a high suc- cess rate with aspirin alone but included Antiphospholipid syndrome women with low titers of aPL only11.

Antiphospholipid antibodies (aPL) are present in 15% of women with RPL and 33% of women Antithyroid antibodies with systemic lupus erythematosus (SLE)10. In women with RPL associated with untreated Patients with treated dysfunction have aPL, the live birth rate may be as low as 10%. no risk of increased miscarriage12. Although Primary antiphospholipid syndrome (APS), more women with RPL have antithyroid anti- which predominantly affects young women, bodies than in the general population, evidence refers to the association of aPL and adverse that these antibodies actually cause pregnancy pregnancy outcome or vascular . loss is lacking13. Adverse pregnancy outcomes include three or more consecutive miscarriages before 10 weeks’ gestation; one or more morphologically Investigation and treatment normal fetal loss after 10 weeks’ gestation; and one or more preterm birth before 34th week of Because several studies report an increased gestation due to severe pre-eclampsia, eclamp- rate of fetal loss in women with high serum sia or placental insufficiency. When APS exists thyroid peroxidase (TPO) antibody concen- in chronic inflammatory disorders, such as trations, we propose that it should be investi- SLE, it is referred to as secondary APS. gated in women with RPL. Current data suggest that in women with Investigation RPL and thyroid antibodies, treatment with L-thyroxine and/or prednisolone should be To diagnose APS, it is mandatory that the considered, although further controlled stud- patient have two positive tests at least 6 weeks ies are essential14.

231 PRECONCEPTIONAL MEDICINE

Antinuclear antibodies interleukin (IL)-2, interferon (IFN)-γ, IL-12 and IL-18; the main Th2 type cytokines include A connection exists between antinuclear anti- IL-3, granulocyte macrophage colony stimulat- bodies (ANAs) and recurrent miscarriages, ing factor (GM-CSF), CSF-1, IL-10 and trans- with a titer of over 1 : 40 causing concern. forming growth factor (TGF)-β23,24. A study published in 2008 demonstrated that women with a history of unexplained recurrent failed Investigation and treatment in vitro fertilization (IVF) treatment not only have a Th1 bias but also that this polarization Measuring antinuclear and anti-dsDNA anti- is enhanced following hormonal manipula- bodies is not recommended as part of an eval- tions that accompany IVF treatment25. uation of women with RPL. It is as yet unclear as to what should be the Treatment with low dose prednisolone proportion of the Th1 cytokines at the fetoma- could be considered as a treatment modality ternal surface to either damage or benefit any in patients who have raised ANA, but further ongoing pregnancy26–28. studies are needed.

Role of other markers/substrates Alloimmune factors In a study measuring serum concentrations Allogeneic factors may cause RPL by a mecha- of macrophage inhibitory cytokine (MIC)-1 in nism similar to that of graft rejection in trans- asymptomatic women at 6–13 weeks’ gesta- plant recipients. tion who subsequently miscarried or who had (HLA) sharing is a condition in which the already miscarried, MIC-1 concentrations were normal process that allows for creation of a third of those in women who had ongoing maternal blocking antibodies in pregnancy is pregnancies, an observation which suggested decreased. However, no clear evidence as yet a possible predictive as well as therapeutic proves an association between RPL and HLA potential for MIC-129. Recurring miscarriages incompatibility between couples. also have shown an association with elevated serum homocysteine concentrations in other studies30,31. Cytokines and miscarriage

Thomas Wegmann first proposed the immu- Natural killer cells and miscarriage notrophic hypothesis suggesting that a suc- cessful allo-pregnancy was a T helper 2 (Th2) NK cells comprise about 10–15% of periph- phenomenon and demonstrating a Th2 cell eral blood lymphocytes. Two distinct subsets cytokine profile response in normal preg- of human NK cells are possible, depending nancy15,16. Since then a number of human and on the cell surface density of the CD56 mol- animal studies17–20 further confirm the Th2 ecule. Approximately 90% of peripheral blood cytokine predominance associated with a suc- human NK cells are CD56dim and express cessful pregnancy, although some controversy high levels of FcγIII (CD16) as well as perforin. exists21,22. The apparently harmful Th1 cyto- In contrast, a minority (approximately 10%) kines, which can activate natural killer (NK) of NK cells are CD56bright and CD16dim. cells into lymphokine-activated killer (LAK) These CD16dim cells are the primary source cells, include tumor necrosis factor (TNF)-α, of NK cell derived cytokines and thought to

232 Recurrent pregnancy loss be an important regulatory subset32,33. In the NK cell numbers uterus, the NK cells form the largest popula- tion of the leukocytes and are predominantly An abnormal increase in peripheral blood NK the CD56bright variety. cell parameters (either in NK cell absolute val- Studies suggest that uterine NK cell func- ues or in proportion (%) prior to conception tion in preimplantation is to or during early pregnancy) is associated with promote angiogenesis, and thus provide a recurrent miscarriage and with mul- potential mechanism by which the increased tiple implantation failures40,43. Data suggest endometrial uterine NK cell density causes that there may be a significant difference in miscarriage by the final common pathway of subpopulations among uterine NK cells, with excessive oxidative stress34,35. a greater proportion of cells being CD56dim, which may have important functional implica- tions. Some studies using CD57 monoclonal NK cell receptor expression antibodies (mAb) demonstrated elevated NK cell populations in decidua44,45, whereas stud- An imbalance between inhibitory and activat- ies using CD56 failed to document change46,47. ing receptor expression is present in women with implantation failures36. When compared with normal controls, CD158a and CD158b Investigation inhibitory receptor expression by CD56dim/ CD16+ and CD56bright/CD16− NK cells Specific immunological testing should be was significantly decreased, and CD161 acti- conducted as a part of ongoing research in vating receptor expression by CD56+/CD3+ a specialized center. This includes NK cells NK cells was significantly increased in women (number and activation), MIC-1, Th1 and Th2 with implantation failures35. In another study, cytokines, HLA typing, mixed lymphocyte infertile women had a significantly higher antibody tests and mixed lymphocyte culture expression of NK cell activation markers of the reactions. CD69+ type37.

Treatment NK cell cytotoxicity In the absence of strong data to prove the Aoki et al. reported increased preconceptional immune-endocrine nature of abnormalities NK cell activity in women with unexplained in recurrent miscarriage, most of the clinical RPL38, while other studies revealed that infer- therapies used over the years have been of an tile women have higher levels of activated empiric nature. NK cells compared with control multiparous women and that women with elevated levels Progesterone In a subgroup analysis of three of activated NK cells have a poorer IVF treat- trials involving 91 women with recurrent mis- ment and pregnancy outcome39,40. In summary, carriages, treatment significantly despite a few contradictory studies41,42 a sig- decreased the miscarriage rate compared with nificant amount of data points to increased placebo or no treatment48. Despite this the peripheral or local NK cell activity contrib- current RCOG Guideline No. 17 (published in uting towards the pathogenesis of recurrent 2003) states that there is insufficient evidence miscarriage. to evaluate the effect of progesterone or human

233 PRECONCEPTIONAL MEDICINE chorionic gonadotropin supplementation in A study in 2008 by Jerzak et al.56, evaluat- pregnancy to prevent a miscarriage49. ing the effects of vaginal sildenafil on NK cell activity, suggested that NK cell activity was Heparin Heparin, in addition to its anticoag- significantly decreased after vaginal sildenafil ulant effects, suppresses NK cell cytotoxicity therapy in the study women. and antagonizes IFN-γ action by inhibiting its binding to the cell surface50. THROMBOPHILIAS AND Prednisolone therapy A recent study by Thum FIBRINOLYTIC FACTORS et al. demonstrated that prednisolone has a similar in vitro suppression effect on NK cell Retrospective studies have suggested an cytolytic capability to intravenous immuno­ association between inherited thrombophilic globulins (IV)51. In addition, Xu et al. showed defects, fetal loss and late pregnancy complica- that prednisolone had a suppressive effect on tions, with a presumed mechanism being defec- TNF-α (Th1 cytokine) production from pla- tive placentation and microthrombi in the pla- cental tissue44. Furthermore, Quenby et al.35 cental vasculature. Inherited thrombo­philias reported that prednisolone could suppress NK include , protein C and S defi- ciency, antithrombin III deficiency, activated cell levels and reduce the miscarriage rate in prothrombin C resistance (APCR), methylene women with a history of recurrent miscarriage. tetrahydrofolate reductase (MTHFR) C677T and G20210A prothrombin gene . IV immunoglobulins Women with a history Acquired includes anticardio- of recurrent miscarriage have a higher level of lipin antibodies and lupus anticoagulant. NK cell cytotoxicity which can be suppressed In the absence of treatment, factor V Leiden by co-culture of the NK cells with immuno- is associated with an increased risk of mis- globulin-G (IVIg)45. However, women who carriage, compared with a normal factor V have elevated NK cell cytotoxicity and a history genotype. Factor V Leiden is carried by 5% of of recurrent miscarriage or recurrent failed Caucasians, but is rarely found among Blacks. implantation during IVF may have a better Other inherited thrombophilias are rare, and obstetric outcome if they have IVIg infusion no conclusive studies have been conducted to during IVF treatment or early pregnancy52,53. prove their causality in RPL. Moreover, RPL has no significant association with plasmino- TNF-α inhibitors, sildenafil and 1,25-dihy- gen activator inhibitor-I4G/5G polymorphism 54 droxyvitamin D3 Winger et al. used mAb or increased plasminogen activator inhibi- directed against TNF-α along with IVIg to tor activity57. Procoagulant microparticles improve pregnancy rates in their IVF patients. were shown to be associated with early and Concerns about such use, however, relate in late unexplained pregnancy loss in one pilot part to an increased risk of infectious diseases, study58. especially tuberculosis. Evans et al.55 demonstrated that several com- ponents of vitamin D metabolism and signal- Investigations and treatment ing are strongly expressed in human uterine decidua from first trimester pregnancies, A full inherited and acquired thrombophilia suggesting that locally produced 1,25-dihy- screen is recommended in women with RPL. droxyvitamin D3 may exert immunosuppres- The general approach is to treat women with sive effects during early stages of gestation. thrombophilia with a combination of low dose

234 Recurrent pregnancy loss aspirin and low molecular weight heparin. levels are not predictive of future pregnancy Therapy may need to be started before preg- outcome60. nancy occurs and continued to 6 weeks after Treatment with progesterone supplementa- birth (see also Chapter 9). tion does not have a beneficial effect on preg- nancy outcome.

ENDOCRINE Endocrine factors may be responsible for 15–20% of RPL. Diabetic gravida with hemoglobin A1c lev- els in the first trimester of more than 8 are at increased risk of miscarriage and fetal Polycystic ovarian syndrome malformations.

Women with polycystic ovarian syndrome (PCOS) have a miscarriage rate of 20–40% as Investigation and treatment compared to the general obstetric population (10–20%). This may be related to elevated Routine screening for diabetes with the oral serum luteinizing hormone (LH) levels, high glucose tolerance test in asymptomatic women testosterone and androstenedione concentra- with RPL should not be performed unless a tions, or insulin resistance59. random glucose value is elevated. Diabetic women with RPL should be treated in a multidisciplinary joint diabetic clinic. Investigation and treatment

Day 2–5 follicle stimulating hormone (FSH), Hyperprolactinemia LH, prolactin, sex hormone binding globulin, prolactin and transvaginal ultrasound are the Normal circulating levels of prolactin may play recommended investigations in women with an important role in maintaining pregnancy. recurrent miscarriages. Pre-pregnancy suppression of high LH by either clomiphene or metformin among ovu- Investigation and treatment latory women with RPL and PCOS does not improve the live birth rate. A study of 64 hyperprolactinemic women with RPL randomly assigned subjects to therapy with bromocriptine or no therapy61. Treatment defect to lower prolactin concentrations was associ- ated with a higher rate of successful pregnancy It is controversial as to whether such a defect (86% versus 52%). Prolactin levels during exists and whether it is related to miscarriage. early pregnancy were significantly greater in women who miscarried61.

Investigation and treatment Thyroid disease Diagnosis of luteal phase defect based on is not predictive of fertil- Poorly controlled thyroid disease (hypo- or ity status, and single or multiple progesterone hyperthyroidism) is associated with infertility

235 PRECONCEPTIONAL MEDICINE and pregnancy loss. Excess thyroid hormone Decreased ovarian reserve increases the risk of miscarriage62. Women with unexplained RPL have a higher incidence of elevated day 3 FSH and estradiol Investigation and treatment levels than women with known causes of RPL. Day 1–3 FSH or a clomiphene challenge test Routine screening for abnormal thyroid func- tion tests should not be performed in asymp- can be considered in women of any age with tomatic women. Women with overt thyroid RPL. A day 3 FSH level of less than 15 mIU/ml disease should be referred to a specialist. and high estradiol levels more than 80 pg/ml are associated with reduced oocyte numbers.

INFECTION UNEXPLAINED Some infections, including listeriosis, toxo- plasmosis, cytomegalovirus and primary geni- A significant proportion of cases of RPL (15%) tal herpes, cause sporadic pregnancy loss, but remain unexplained, despite detailed inves- no infectious agent has been proven to cause tigations. These women can be reassured RPL63. that the prognosis for a successful pregnancy outcome with supportive care alone is in the Investigation and treatment region of 75%. Treatment offered to couples with unexplained RPL includes the following: Routine cervical cultures for Chlamydia or • Lifestyle modification Weight loss, exer- , vaginal evaluation for bacterial cise, avoiding alcohol, caffeine intake and vaginosis and TORCH (toxoplasma, rubella, cytomegalovirus and herpes simplex) serology smoking. are not useful in the evaluation of RPL, but • Progesterone Large randomized controlled they may be indicated by patient history. studies demonstrating the efficacy of pro- Screening for and treatment of bacterial gesterone treatment are lacking, but the vaginosis in early pregnancy in women with a is widely prescribed to women with history of second trimester miscarriage or pre- RPL. term labor may reduce the risk of RPL. • IVF and preimplantation genetic diagnosis (PGD) Studies evaluating the value of IVF OTHER CAUSES in women with RPL have yielded mixed results. A combination of IVF and PGD at Chemicals the 6–8 cell stage appears promising65. Chemicals which have been associated with • Oocyte donation Ovum donation can over- RPL include nitrous oxide, arsenic, aniline come the problem of poor quality oocytes dyes, benzene, ethylene oxide, lead, pesticides, and has been associated with a live birth mercury and cadmium. rate of 88% in women with RPL66. • Combination therapy A recent observa- Personal habits tional study compared 50 pregnant women treated before and during pregnancy with The association between RPL and smok- prednisolone (20 mg/day), progesterone ing, alcohol use or caffeine consumption is (200 mg/day), aspirin (100 mg/day) and unclear64. folate (5 mg/day) with 52 women who were

236 Recurrent pregnancy loss

not treated; the first trimester pregnancy implantation failure associated with a negative loss rate was 19% in the treated and 63% pregnancy test. The prevalence of ANA and NK in the untreated group. Although this dif- cells did not differ between the two groups. ference was not statistically significant, it The authors concluded that the mechanisms is clinically important and perhaps resulted involved in chemical pregnancies may be the from insufficient study numbers67. result of defective angiogenesis as compared • Complementary therapies Many acupunc- to pregnancies with a negative pregnancy test turists report success in treating women which involve implantation failure. with a history of RPL. Dietary supplemen- tation with vitamin B complex, including folic acid and co-enzyme Q10 may suggest Treatment a reduction in RPL. Reflexology, a holistic therapy, attempts to relieve stress, pain and As not much work has been done on the diag- muscle tension and thus help to reduce nosis and treatment modalities of chemical miscarriages. pregnancies, it is a very challenging area of . A short trial of low dose prednisolone could be the way forward BIOCHEMICAL PREGNANCY LOSS in the treatment of recurrent miscarriage, especially as the safety of prednisolone is well A biochemical or pre-clinical pregnancy loss is established. High quality data on management defined as loss of a biochemically evident preg- of biochemical RPL are limited and, therefore, nancy before it is identifiable on ultrasound. therapeutic intervention is largely guided by Early pregnancy loss occurs in 75% of all the underlying cause. pregnancies, out of which 15–20% are clini- cally recognized. However, the true rate of early pregnancy loss is close to 50%, because Conclusion of the high number of chemical pregnancies that are not recognized in the 2–4 weeks after RPL is an emotionally traumatic experience conception. In a classic study by Wilcox et al. for a couple. Multidisciplinary teams expert in 198868, 221 women were followed up dur- in managing patients with RPL should co- ing 707 total menstrual cycles. A total of 198 ordinate evaluation and management. These pregnancies were recorded; 43 (22%) were should include gynecologists, geneticists, lost before onset of menses and another 20 rheumatologists, hematologists, immunolo- (10%) were clinically recognized losses. gists and reproductive specialists. High qual- ity data on management of RPL are limited; therapeutic intervention is largely guided by Investigations the underlying cause of RPL. In all cases, emo- tional support is important in caring for these No investigative studies have been conducted anxious couples. on the recurrent biochemical pregnancy loss. A US study on 122 women experiencing IVF implantation failure with a negative pregnancy References test and 20 women with chemical pregnancy loss evaluated aPL, ANA and elevated NK 1. Paukku M, Tulppala M, Puolakkainen M, et cells69. Women with chemical pregnancies had al. Lack of association between serum anti- a higher frequency of aPL than women with bodies to and a his-

237 PRECONCEPTIONAL MEDICINE

tory of recurrent pregnancy loss. Fertil Steril obstetrical complications. J Clin Endocrinol 1999;72:427 Metab 2006;91:2587 2. Salat-Baroux J. Recurrent spontaneous abor- 15. Wegmann TG. Fetal protection against abor- tions. Reprod Nutr Dev 1988;28:1555 tion: is it immuno-suppression or immuno- 3. Edmonds DK, Lindsay KS, Miller JF, et al. Early stimulation? Ann Immunol (Inst Pasteur) embryonic mortality in women. Fertil Steril 1984;135:309–12 1982;38:447 16. Wegmann TG, Lin H, Guilbert L, Mosmann 4. Li TC, Tuckerman EM, Laird SM. Endometrial TR. Bidirectional cytokine interactions in the factors in recurrent miscarriage. Hum Reprod maternal–fetal relationship: is successful preg- Update 2002;8:43 nancy a TH 2 phenomenon? Immunol Today 5. Hill JA. Recurrent pregnancy loss. In: Ryan KJ, 1993;14:353–6 Berkowitz RS, Barbieri RL, Dunaif A, eds. Kist- 17. Tangri S, Raghupathy R. Expression of cyto- ner’s and Women’s Health, 7th edn. St kines in placentas of mice undergoing immu- Louis: Mosby,1999:396 nologically mediated spontaneous fetal resorp- 6. Homer HA, Li TC, Cooke LD. The septate tions. Biol Reprod 1993;49:850–6 uterus: A review of management and reproduc- 18. Tangri S, Wegmann TG, Lin H, Raghupathy R. tive outcome. Fertil Steril 2000;73:1 Maternal anti-placental reactivity in natural, 7. Proctor JA, Haney AF. Recurrent first trimes- immunologically mediated spontaneous fetal resorptions. J Immunol 1994;152:4903–11 ter pregnancy loss is associated with uterine 19. Chaouat G, Menu E, Assal-Meliani A, et al. septum but not . Fertil Steril Allopregnancy is a TH 2 like phenomenon: IL 10 2003;80:1212 can prevent CBA×DBA/2 fetal wastage and TH 2 8. Goldberg JM, Falcone T, Attaran M. Sonohys- cytokine can be induced in vivo by tau interferons. terographic evaluation of uterine abnormalities Annecy, France: Colloque Foundation Marcel noted on hysterosalpingography. Hum Reprod Merieux, 1994 1997;12:2151–3 20. Hill JA, Polgar K, Anderson DJ. T helper 1 9. MRC/RCOG Working Party on Cervical Cer- type immunity to trophoblast in women clage. Final report of the Medical Research with recurrent spontaneous . JAMA Council/Royal College of Obstetricians and 1995;273:1933–6 Gynaecologists multicentre randomised 21. Rein DT, Schondorf T, Gohring UJ, et al. Cyto- trial of cervical cerclage. Br J Obstet Gynaecol kine expression in peripheral blood lympho- 1993;100:516–23 cytes indicates a switch to T (HELPER) cells 10. Reindollar RH. Contemporary issues for in patients with preeclampsia. J Reprod Immunol spontaneous abortion. Does recurrent abor- 2002;54:133–42 tion exist? Obstet Gynecol Clin North Am 22. Schust DJ, Hill JA. Correlation of serum cyto- 2000;27:541–54 kine and molecule determinations 11. Empson M, Lassere M, Craig JC, Scott JR. with pregnancy outcome. J Soc Gynecol Invest Recurrent pregnancy loss with antiphospho- 1996;3:259–61 lipid antibody; a systematic review of therapeu- 23. King A, Hiby SE, Gardner L, et al. Recogni- tic trials(1). Obstet Gynecol 2002;99:135 tion of trophoblast HLA class I molecules by 12. Glinoer D, Soto MF, Bourdoux P, et al. Preg- decidual NK cell receptors – a review. Placenta nancy in patients with mild thyroid abnormali- 2000;21(Suppl A):S81–5 ties; maternal and neonatal repercussions. J 24. Raghupathy R, Maksheed M, Azizieh F, Al- Clin Endocrinol Metab 1991;73:421–7 Azemi MMK, Hassan NA, Bandar A. Th1 and 13. Satagnaro- Green A, Glinoer D. Thyroid auto- Th2 cytokine profiles in successful pregnancy immunity and the risk of miscarriage. Best Pract and unexplained recurrent . In: Gupta Res Clin Endocrinol Metab 2004;18:167–81 SK, ed. Reproductive Immunology 2000. Dehli: 14. Negro R, Formoso G, Mangieri T, et al. Levothy- Kluwer Academic Publisher, 2000:149–66 roxine treatment in euthyroid pregnant women 25. Kalu E, Bhaskaran S, Thum MY, et al. Serial esti- with autoimmune thyroid disease: effects on mation of Th1:Th2 cytokines profile in women

238 Recurrent pregnancy loss

undergoing in-vitro fertilization embryo trans- 37. Coulam CB, Roussev RG. Correlation of NK fer. Am J Reprod Immunol 2008;59:206–11 cell activation and inhibition markers with 26. Loke YW, King A. Current topic: interferon NK cytotoxicity among women experiencing and human placental development. Placenta immunologic implantation failure after in vitro 1990;11:291–9 fertilization and . J Assist Reprod 27. Ashkar AA, Croy BA. Functions of uterine Genet 2003;20:58–62 natural killer cells are mediated by interferon 38. Aoki K, Kajiura S, Matsumoto Y, et al. Precon- gamma production during murine pregnancy. ceptional natural-killer-cell activity as a predic- Semin Immunol 2001; 3:235–41 tor of miscarriage. Lancet 1995;345:1340–2 28. Yeaman GR, Collins JE, Currie JK, Guyre PM, 39. Thum MY, Bhaskaran S, Abdalla HI, et Wira CR, Fanger MW. IFN-gamma is produced al. An increase in the absolute count of by polimorphonuclear neutrophils in human CD56dimCD16+ CD69+ NK cells in the uterine endometrium and by cultured periph- peripheral blood is associated with a poorer eral blood polimorphonuclear neutrophils. J IVF treatment and pregnancy outcome. Hum Immunol 1998;160:5145–53 Reprod 2004;19:2395–400 29. Tong S, Marjono B, Brown DA, et al. Serum 40. Fukui A, Fujii S, Yamaguchi E, Kimura H, Sato concentrations of macrophage inhibitory cyto- S, Saito Y. subpopulations kine 1 (MIC 1) as a predictor of miscarriage. and cytotoxicity for infertile patients undergo- ing in vitro fertilisation. Am J Reprod Immunol Lancet 2004;363:129–30 1999;41:413–22 30. Sikora J, Magnucki J, Zietek J, et al. Homocys- 41. Emmer PM, Nelen WL, Steegers EA, Hendriks teine serum concentration and uterine artery JC, Veerhoek M, Joosten I. Peripheral natural color Doppler examination in cases of recur- killer cytotoxicity and CD56(pos)CD16(pos) rent miscarriages with unexplained etiology. cells increase during early pregnancy in women Neurol Endocrinol Lett 2007;28:502–6 with a history of recurrent spontaneous abor- 31. Sikora J, Magnucki J, Zietek J, et al. Homocys- tion. Hum Reprod 2000;15:1163–9 teine, folic acid and vitamin B12 concentration 42. Souza SS, Ferriani RA, Santos CM, Voltarelli in patients with recurrent miscarriages. Neurol JC. Immunological evaluation of patients Endocrinol Lett 2007;28:507–12 with recurrent abortion. J Reprod Immunol 32. Orange JS, Ballas ZK. Natural killer cells 2002;56:111–21 in human health and disease. Clin Immunol 43. Perricone R, Di Muzio G, Perricone C, et al. 2006;118:1–10 Fontana L, De Carolis C. High levels of periph- 33. Cooper MA, Fehniger TA, Turner SC, et al. eral blood NK cells in women suffering from Human natural killer cells: a unique innate recurrent spontaneous abortion are reverted immunoregulatory role for the CD56(bright) from high-dose intravenous immunoglobulins. subset. Blood 2001;97:3146–51 Am J Reprod Immunol 2006;55:232–9 34. Quenby S, Nik H, Innes B, et al. Uterine natural 44. Xu B, Makris A, Thornton C, Hennessy A. Glu- killer cells and angiogenesis in recurrent repro- cocorticoids inhibit placental cytokines from ductive failure. Hum Reprod 2009;24:45–54 cultured normal and preeclamptic placental 35. Quenby S, Kalumbi C, Bates M, Farquhar- explants. Placenta 2005;26:654–60 son RG, Vince G. Prednisolone reduces pre- 45. Ruiz JE, Kwak JY, Baum L, et al. Effect of intra- conceptual endometrial natural killer cells in venous immunoglobulin G on natural killer women with recurrent miscarriage. Fertil Steril cell cytotoxicity in vitro in women with recur- 2005;84:980–4 rent spontaneous abortion. J Reprod Immunol 36. Ntrivalas EI, Bowser CR, Kwak-Kim J, Bea- 1996;31:125–41 man KD, Gilman-Sachs A. Expression of killer 46. Quack KC, Vassiliadou N, Pudney J, Anderson immunoglobulin-like receptors on peripheral DJ, Hill JA. Leukocyte activation in the decidua blood NK cell subsets of women with recur- of chromosomally normal and abnormal rent spontaneous abortions or implantation fetuses from women with recurrent abortion. failures. Am J Reprod Immunol 2005;53:215–21 Hum Reprod 2001;16:949–55

239 PRECONCEPTIONAL MEDICINE

47. Hill JA, Melling GC, Johnson PM. Immunohis- 58. Laude I, Rongieres-Bertrand C, Boyer-Neu- tochemical studies of human uteroplacental mann C, et al. Circulating procoagulant mic- tissues from first-trimester spontaneous abor- roparticles in women with unexplained preg- tion. Am J Obstet Gynecol 1995;173:90–6 nancy loss: a new insight. Thromb ­Haemostat 48. Szekeres-Bartho J, Balasch J. Progestagen 2001;85:18 therapy for recurrent miscarriage. Hum Reprod 59. Okon MA, Laird SM, Tuckerman EM, Li TC. Update 2008;14:27–35 Serum androgen levels in women who have 49. RCOG. The Investigation and Treatment of Couples recurrent miscarriages and their correlation with Recurrent Miscarriage. Guideline No. 17. with markers of endometrial function. Fertil London, UK: RCOG, Revised May 2003 Steril 1998;69:682 50. Dosiou C, Giudice LC. NK cells in pregnancy 60. Ogasawara M, Kajiura S, Kantano K, et al. Are and RPL: endocrine and immunological per- serum progesterone levels predictive of recur- spectives. Endocr Rev 2005;26:44–62 rent miscarriage in future pregnancies. Fertil 51. Thum M-Y, Bhaskaran S, Abdalla HI, Ford B, Steril 1997;68:806 Sumar N, Bansal A. Prednisolone suppresses 61. Hirahara F, Andoh N, Sawai K, et al. Hyperp- NK cell cytotoxicity in vitro in women with a rolactinemic recurrent miscarriage and result history of infertility and elevated NK cell cyto- of randomized bromocriptine treatment trials. toxicity. Am J Reprod Immunol 2008;59:259–65 Fertil Steril 1998;70:246 52. Coulam CB, Goodman C. Increased preg- 62. Anselmo J, Cao D, Karrison T, et al. Fetal loss nancy rates after IVF⁄ET with intravenous associated with excess thyroid hormone expo- immunoglobulin treatment in women with sure. JAMA 2004;292:691 elevated circulating C56+ cells. Early ­Pregnancy 63. ACOG practice bulletin. Management of recur- 2000;4:90–8 rent pregnancy loss. Number 24, February 53. Stricker RB, Steinleitner A, Bookoff CN, 2001. American College of Obstetricians and Weckstein LN, Winger EE. Successful treat- Gynecologists. Int J Gynaecol Obstet 2002;78:179 ment of immunologic abortion with low-dose 64. Christiansen OB, Nybo Anderson AM, Bosch intravenous immunoglobulin. Fertil Steril E, et al. Evidence based investigations and 2000;73:536–40 treatment of recurrent pregnancy loss. Fertil 54. Winger EE, Reed JL, Sherif A, et al. Treatment Steril 2005;83:821–39 with tumor necrosis factor inhibitors and intra- 65. Egozcue J, Santalo J, Gimenez C, et al. Pre- venous immunoglobulin improves live birth implantation genetic diagnosis. Mol Cell Endo- rates in women with recurrent spontaneous crinol 2000;166:21 abortion. Am J Reprod Immunol 2008;60:8–16 66. Trout SW, Seifer DB. Do women with unex- 55. Evans KN, Nguyen L, Chan J, et al. Effects of plained recurrent pregnancy loss have higher 25-hydroxyvitamin D3 and 1,25-dihydroxyvi- day 3 serum FSH and estradiol values? Fertil tamin D3 on cytokine production by human Steril 2000;74:335 decidual cells. Biol Reprod 2006;75:816–22 67. Tempfer CB, Kurz C, Bentz EK, et al. A combi- 56. Jerzak M, Kniotek M, Mrozek J, et al. Sildenafil nation treatment of prednisone, aspirin, folate, citrate decreased natural killer cell activity and and progesterone in women with idiopathic enhanced chance of successful pregnancy in recurrent miscarriage: a matched-pair study. women with a history of recurrent miscarriage. Fertil Steril 2006;86:145–8 Fertil Steril 2008;90:1848–53 68. Wilcox AJ, Weinberg CR, O’Connor JF, et al. 57. Sotiriadis A, Makrigiannakis A, Stefos T, et al. Incidence of early loss of pregnancy. N Engl J Fibrinolytic defects and recurrent miscarriage: Med 1998;319:189–94 a systematic review and meta-analysis. Obstet 69. Roussev R, Coulam CB. Chemical pregnancies. Gynecol 2007;109:1146 J Immunol Ultrasonograph Stud 2003;48:323–8

240