{"paper_id":"58ce7a26-2ce6-48b2-bef5-f37c7a9f802d","body_text":"Hormonal causes of recurrent pregnancy loss (RPL)\nNicola Pluchino,1 Panagiotis Drakopoulos,1 Jean Marie Wenger,1 Patrick Petignat,1 \nIsabelle Streuli,1 Andrea Riccardo Genazzani2\n1Division of Obstetrics and Gynaecology, University Hospitals of Geneva, Switzerland; 2Division of Obstetrics and \nGynaecology, University Hospitals of Pisa, Italy\nAbSTrAcT\nEndocrine disorders play a major role in approximately 8% to 12% of recurrent pregnancy loss \n(rPL). Indeed, the local hormonal milieu is crucial in both embryo attachment and early preg-\nnancy. Endocrine abnormalities, including thyroid disorders, luteal phase defects, polycystic \novary syndrome, hyperprolactinaemia and diabetes have to be evaluated in any case of rPL. \nMoreover, elevated androgen levels and some endocrinological aspects of endometriosis are \nalso factors contributing to rPL. In the present article, we review the significance of endocrine \ndisease on rPL. \nKey words: Abortion, Hyperprolactinaemia, Luteal phase defect, Miscarriage, Recurrent preg-\nnancy loss\nReview\nHORMONES 2014, 13(3):314-322\nAddress for correspondence:\nPanagiotis Drakopoulos, Boulevard de la Cluse 30,  \n1205, Geneva, Switzerland, Tel.: +30 795532356 \nE-mail: Panagiotis.Drakopoulos@hcuge.ch\nReceived: 01-05-2014, Accepted: 08-05-2014\n1. IntroductIon\nThe development and the maintenance of pregnan-\ncy are dependent on numerous endocrinological events \nthat lead to the successful growth and development \nof the foetus. Although the vast majority of pregnant \nwomen have no pre-existing endocrine abnormalities, \na small percentage of women may develop endocrine \nalterations that could potentially lead to sporadic or \nrecurrent miscarriage. It is estimated that approxi-\nmately 8% to 12% of all cases of recurrent pregnancy \nloss (RPL) are caused by endocrine diseases. 1 Dis-\norders related to inadequate progesterone secretion \nby the corpus luteum, hyperprolactinaemia, diabetes \nmellitus and polycystic ovarian syndrome are some \nexamples of endocrine disorders affecting pregnancy. \nMeanwhile, hyperinsulinaemia and elevated androgen \nlevels have recently been added to the endocrinologi-\ncal abnormalities associated with RPLs.\nThere are two types of miscarriages: sporadic and \nrecurrent. Approximately 30% to 50% of all concep-\ntions and 15% of clinically recognized pregnancies \n(≥6 weeks of gestation) fail to result in a live birth, \nthis due to random foetal chromosomal abnormality \nin 50%-60% of cases.2 RPL comes about through a \nnumber of causes (genetic, anatomical, hormonal, \ninfections, etc.) and is defined by two or more clini-\ncally failed pregnancies detected via ultrasonography \nor histopathologic examination.3 However, as it lacks \na consistent definition, some clinicians continue to \ndefine RPL as consisting of three or more consecu-\ntive pregnancy losses. Indeed, requiring three failed \n\nHormonal causes of recurrent pregnancy loss 315\npregnancies prior to initiating work-up may offer little \nadditional clinical insight compared to testing after the \nsecond loss. Accurate prevalence is not available, but \nit has been estimated that 2%-5% of women experi-\nence RPL with the majority of these cases occurring \nbefore 10 weeks of gestation.4 The pathophysiological \nmechanisms of RPL are as yet poorly understood \nand the aetiologies remain unexplained in up to 50% \nof affected couples. This reproductive disorder is \na challenge to the clinician because genetic, endo-\ncrinological, anatomic, immunologic, thrombophilic \nand iatrogenic data may be required for a complete \nevaluation. Foetal aneuploidy is the most common \ncause of sporadic miscarriage. However, it should be \nnoted that only 4% of couples with RPL have one \npartner carrying a balanced translocation, including \nreciprocal and Robertsonian translocations, resulting \nin unbalanced translocation in the foetus.5\nThe purpose of this article is to review the patho-\nphysiology of endocrinological diseases causing RPL. \n2. luteal phase defIcIency \nProgesterone production triggers morphological \nand physiological changes in the endometrium creat-\ning a suitable environment for the embryo during the \nimplantation window [5-10 days after the luteinizing \nhormone (LH) surge]. In addition, progesterone \nhelps in maintaining early pregnancy. More specifi-\ncally, progesterone affects the proliferation and dif-\nferentiation of stromal cells and augments uterine \nreceptivity through the modulation of locally acting \ngrowth factors and regulation of cytokine production \nin the maternal-foetal interface. Indeed, human and \nanimal studies suggest that progesterone maintains \npregnancy by down-regulation of Th1 cytokines and \nstimulation of Th2 cytokines.6 There is evidence that \nTh2 cytokines favour normal pregnancy, while an \nexcess of Th1 cytokines leads to pregnancy termina-\ntion.\n7 In particular, in the presence of progesterone, \nthe lymphocytes of pregnant women release a 35 \nkD protein named the progesterone-induced block-\ning factor (PIBF), which in turn alters the profile of \ncytokine secretion of activated lymphocytes shifting \nthe balance towards Th2 dominance.\nAll these changes fail to ensue if progesterone \nproduction is lower than the normal minimum. In early \npregnancy, the corpus luteum continues to produce \nprogesterone until the luteal placental shift. Luteal \nphase deficiency (LPD) was originally thought to \nderive from inadequate production of progesterone \nby the corpus luteum and subsequent inadequate \nendometrial maturation to allow proper placenta-\ntion. Those supporting the presence of LPD have \ntheorized that LPD occurs because of poor follicular \ndevelopment, decreased progesterone production by \nthe corpus luteum and a dysfunctional endometrial \nresponse to normal progesterone levels.\n8 However, \nother hypothesized causes of LPD include stress, exer-\ncise, weight loss, hyperprolactinaemia and menstrual \ncycles at the onset of puberty or perimenopause. 9 \nAbnormalities of the luteal phase have been histori-\ncally reported to occur in up to 35% of women with \nRPL.1 However, actual presence of such a defect and \nits relation to miscarriage is a controversial issue and \nthere is no consensus on the best method of diagnosis \n(such as serum progesterone, endometrial biopsy). \nSerum progesterone levels greater than 10 ng/mL in \nthe mid-luteal phase are rarely associated with an ab-\nnormal luteal phase.11,12 Although serum progesterone \nlevels below <12 ng/ml have been associated with an \nincreased risk of miscarriage,13 its levels can fluctuate \nduring measurements since its secretion is pulsatile \nand therefore the interpretation of progesterone levels \nmay be difficult. Consistently low progesterone levels \nmay indicate insufficient progesterone to maintain \npregnancy; alternatively, it may be the result of the \nfailing pregnancy which produces low hCG leading \nto low progesterone levels. In the latter case, low \nprogesterone levels seem to be the mechanism lead-\ning to expulsion of the conceptus rather than the \ncause of abortion. Historically, endometrial biopsy, \nwith evaluation of the morphological changes, was \nconsidered superior to serum progesterone. However, \nhistological analysis is no longer the gold standard \nfor assessment of endometrial maturation. More \nrecently, immunohistochemical molecular markers \n(including cytokines and adhesion molecules) have \nbeen used for cell biological assessment of endome-\ntrial receptivity, while a less traumatic method has \nbeen also proposed.\n14,15 \nProgesterone supplementation after ovulation \nwith or without the use of ovulation-induction agents \ncan also be applied 2 to 3 days after the basal body \n\n316 N. PLuCHINo eT AL\ntemperature increases (or after a positive urinary \nLH test) and continued for up to 7 to 11 weeks of \ngestation.16 Progesterone supplementation can be \nadministered by intravaginal suppositories, intramus-\ncular injection of progesterone in oil, as oral micro-\nnized progesterone, or as oral dydrogesterone. The \nsubject of progesterone administration continues to \nbe controversial. Daya’s meta-analysis demonstrated \nthat hormone treatment to enhance progesterone \nproduction or supplementation is associated with \nan increased chance of a term pregnancy in women \nwith RPL. 17 However, the more recent Cochrane \ndatabase systematic review by Oates-Whitehead, 18 \nwhich analyzed the same three papers used in Daya’s \nmeta-analysis ten years earlier, found no association \nwith improved pregnancy outcome. Currently, a two-\ncentre trial (PROMISE) is taking place, the results \nof which are eagerly awaited.\n2.1. Luteal phase defect: lesson learnt from IVF\nStimulated IVF cycles are associated with a defec-\ntive luteal phase in almost all patients and proges-\nterone supplementation appears to be mandatory \nto improve pregnancy rate.19 The aetiology of luteal \nphase defect IVF has been debated for more than \ntwo decades and different mechanisms have been \nproposed. Recently, it has been postulated that one \nof the main causes of the luteal phase defect in stimu-\nlated IVF cycles is related to the supraphysiological \nlevels of steroids secreted by a high number of corpora \nlutea during the early luteal phase, which directly \ninhibit LH release via negative feedback actions at \nthe hypothalamic-pituitary axis level.16\nStudies in humans have demonstrated that the \ncorpus luteum requires a consistent LH stimulus in \norder to perform its physiological function. LH sup-\nport during the luteal phase is entirely responsible \nfor the maintenance and the normal steroidogenic \nactivity of the corpus luteum. As a result, withdrawal \nof LH unnecessarily causes premature luteolysis.20\nIn addition, supraphysiological concentrations of \nprogesterone in stimulated cycles also cause an ac-\ncelerated transformation to secretory endometrium \nat the time of embryo transfer, which has been shown \nto be detrimental to implantation rates. Immunohis-\ntochemical studies and microarray technology have \nindicated an effect of ovarian stimulation on molecular \nprocesses involved in implantation that are differen-\ntially expressed in stimulated compared to natural \ncycles.21 Understanding and ameliorating the impact \nof hormones on the natural cycle and during ovarian \nstimulation on the intrauterine environment may \nresult in improved implantation rates in the future.\n2.2. Hyperprolactinaemia\nProlactin (PRL) is mainly synthesized and secreted \nby the lactotroph cells of the pituitary, but also by \nother sites, such as the mammary gland, placenta, \nuterus and T lymphocytes. Evidence shows that PRL \nis essential to female reproduction. PRL is commonly \nmeasured in women with RPL, as elevated PRL levels \nare associated with ovulatory dysfunction. Past in vitro \nstudies have shown that PRL plays a critical role in \ncorpus luteum maintenance and progesterone pro-\nduction in rodents, but not in humans. 22 Moreover, \nprogesterone secretion by cultured granulosa cells \nobtained from human ovarian follicles is almost \ncompletely inhibited by high PRL concentrations \n(100 ng/mL), but not by lower concentrations (10 \nto 20 ng/mL).23 These observations suggest the pos-\nsibility that high PRL concentrations in the early \nphase of follicular growth may inhibit progester-\none secretion, resulting in luteal-phase defects. By \ncontrast, more recent researches on rodents have \nrevealed that PRL receptors are involved not only \nin generating but also in maintaining pregnancy. \nHowever, the precise cellular mechanism of PRL \naction in the human ovary remains to be clarified. A \nrandomized control trial of 64 hyperprolactinaemic \nwomen with RPL treated with bromocriptine was \nassociated with a higher rate of successful pregnancy, \nand PRL levels were significantly higher in women \nthat miscarried. 24 Treatment resulted in an 85.7% \nlive birth rate, whereas the untreated cohort had a \n52.4% live birth outcome. Bromocriptine was given \nbefore conception and continued until the end of \nthe 9th week of gestation in the group of patients in \nwhom the serum PRL levels were normalized. On the \nother hand, in a more recent study of 122 subjects \nwith RPL, only three of them had marginally elevated \nPRL levels and one a significantly high level (>100 \nng/ml).\n25 In conclusion, normal PRL levels may play \nan important role in the growth and maintenance of \nearly pregnancy, but further studies are required to \nclarify the role of PRL in the pathogenesis of recur-\n\nHormonal causes of recurrent pregnancy loss 317\nrent miscarriages and to establish whether, in cases of \nhyperprolactaemia, continuation of treatment during \npregnancy may be useful. \n3. thyroId abnorMalItIes\n3.1 Hyperthyroidism\nHyperthyroidism occurs in approximately 0.1%-\n0.4% of pregnancies.26 It seems that excess production \nof thyroid hormone is usually not correlated with \ninfertility or RPL. Pregnant women with untreated \nexcess hyperthyroidism are at increased risk for \nspontaneous miscarriage, congestive heart failure, \nthyroid storm, preterm delivery, pre-eclampsia, foetal \ngrowth restriction and increased perinatal morbidity \nand mortality.27 Treatment of overt Graves’ hyperthy-\nroidism in pregnancy to achieve adequate metabolic \ncontrol has been associated with improved pregnancy \noutcomes. However, hyperthyroidism has not com-\nmonly been reported as an independent cause of \nRPL. Only one retrospective study has suggested that \nexcess exogenous thyroid hormone is associated with \nan elevated rate of foetal loss.28\n3.2 Hypothyroidism \nThe most prevalent cause of hypothyroidism in \npregnant women, affecting approximately 0.5% of \npatients, is chronic autoimmune thyroiditis (Hashi-\nmoto’s thyroiditis).29 Other causes of hypothyroidism \ninclude endemic iodine deficiency, prior radioactive \niodine therapy and thyroidectomy. Untreated hypo-\nthyroidism in pregnancy has consistently been shown \nto be associated with an increased risk for adverse \npregnancy complications as well as detrimental ef-\nfects on foetal neurocognitive development. Specific \nadverse outcomes associated with maternal overt \nhypothyroidism include increased risks for premature \nbirth, low birth weight and miscarriage. 30 Current \nevidence suggests that treated thyroid dysfunction \nis not associated with RPL. For this reason, patients \nshould be euthyroid before attempting pregnancy and \nit is necessary to regularly control the levels of TSH \nduring pregnancy. Thyroid hormones have an impact \non oocytes at the level of the granulosa and luteal cells \nthat interfere with normal ovulation.31 Low thyroxine \nlevels exert a positive feedback effect on thyroid-\nreleasing hormone (TRH). High concentrations in \nTRH have been associated with high PRL levels.  It \nis believed that high PRL levels alter the pulsatility \nof gonadotropin-releasing hormone (GnRH) and \ninterfere with normal ovulation. Therefore, severe \nforms of hypothyroidism rarely complicate pregnancy \nsince they are more closely associated with anovulation \nand infertility. Even if an association exists between \nlow thyroid function and pregnancy loss, there is \nno direct evidence for a causal role. 32 We believe it \nprudent to screen for thyroid disease by measuring \nTSH levels and normalizing thyroid function prior to \nconception when the function is found to be abnor-\nmal. There is disagreement as to the suitable upper \nlimit of normal serum thyroid-stimulating hormone \n(TSH) in order to make the diagnosis of subclinical \nhypothyroidism. The trend regarding the new TSH \nassays is to decrease the upper limit of normal TSH \n(range, 4.5 to 5 mU/L) to 2.5 mU/L. This upper limit \nis recommended by the National Academy of Clini-\ncal Biochemistry guidelines and is based on the fact \nthat 2.5 mU/L  represents more than two standard \ndeviations above meticulously screened euthyroid \nvolunteers.33 \n3.3 Thyroid Autoimmunity \nAutoimmune thyroid disease is the most com-\nmon endocrine disorder in women of reproductive \nage, with an overall prevalence in women of 10% to \n15%.34 The role of thyroid autoantibodies is debatable \nwith regard to whether there is a causal relationship \nwith RPL. In recent years, studies have found an as-\nsociation between thyroid autoimmunity (TA) and \nrecurrent abortions; moreover, it has been suggested \nthat thyroid autoantibodies may be employed as a \nmarker for at-risk pregnancies.35 These studies have \nlinked TA with recurrent miscarriages, although the \nmechanism involved is not completely understood. \nTwo mechanisms have been postulated to explain \nthe possible association between TA and early preg-\nnancy loss. 1) The presence of TA, which reflects a \ngeneralized activation of the immune system and a \ngenerally heightened autoimmune reactivity against \nthe foeto-placental unit.36 This notion is supported by \nthe observation that women with recurrent abortions \nhave an increased number of CD5/20-positive B cells \nas compared with women with a normal pregnancy or \nwith only one abortion.37 In addition, women with RPL \npresent a significant increase in the endometrial Th1 \n\n318 N. PLuCHINo eT AL\ncell population that is associated with hypersecretion \nof INFg and reduced secretion of IL-4 and IL-10. 38 \nDominant pro-inflammatory Th1 immune responses \nare related to recurrent spontaneous abortions. Previ-\nous observations suggest that there are activated T \ncells in the uteri of women with anti-thyroid antibodies \nwhich may secrete cytokines that hamper successful \npregnancy. These adverse effects can be mediated \ndirectly through T cells or indirectly through other \ncells, such as natural killer cells. 39 2) The presence \nof TA, which may act as an infertility factor and may \ndelay conception increasing the rate of aging related \nmiscarriage.40 There are no clear recommendations \nas to whether euthyroid women should be tested for \nTA, but it seems that there is no benefit in treating \nthese women with thyroid replacement therapy.41 Se-\nlenium is a trace element which is essential in thyroid \nhormone synthesis. Selenium substitution decreased \nTPO antibody levels in euthyroid subjects and women \nwith recurrent pregnancy loss had lower selenium \nlevels in their hair than controls.42 However, there is \nno RCT regarding the role of selenium substitution \nin women with RPL.\nFurther studies are required to determine whether \nall women with positive thyroidautoantibodies should \nnot be started on thyroxin therapy during their preg-\nnancies to decrease the miscarriage rate.\n4. dIabetes MellItus\nPregestational diabetes, including type 1 and type \n2 diabetes as well as other rare types of diabetes, com-\nplicates from 0.5% to 1% of all pregnancies.43 Studies \nshow that patients suffering from this clinical condi-\ntion run a significantly increased risk of spontaneous \nabortion, preterm labour, hypertensive disorders and \noperative deliveries.44,45 The main underlying cause \nis lethal embryonic malformations, the prevalence of \nwhich is increased in the case of poorly controlled \ndiabetes during the periconceptional period. 46,47,48 \nGlucose is teratogenic at high levels, and rates of \ncongenital foetal anomalies are directly related to \nglycaemic control in the first trimester. Current evi-\ndence shows that well-controlled diabetes is not a risk \nfactor for RPL and attention should first be given to \noptimal metabolic control of diabetic women during \nthe preconceptional period.49,50\n5. polycystIc ov ary syndroMe\nIt has been estimated that 40% of pregnancies in \nwomen with PCOS will result in spontaneous loss.51 \nHowever, diagnostic criteria for this heterogeneous \ndisorder have not been uniform, resulting in a wide \nrange of reported prevalence of both miscarriage and \nRPL in PCOS patients.52 The majority of these stud-\nies have used polycystic ovary morphology alone to \ndefine PCOS and the results are extremely variable \ndue to a variety of diagnostic and selection criteria \nemployed. Polycistic ovarian morphology (PCO) per \nse is not a predictive of pregnancy loss in women \nwith PRL.53 However, patients with PCOS may have \nseveral underlying contributing and interrelated \nfactors, which have been reported in women with \nRPL, regardless of whether they have PCOS. These \ninclude obesity, hyperinsulinaemia, insulin resistance, \nhyperhomocysteinaemia, high levels of plasminogen \nactivator inhibitor-1 factor, hyperandrogenaemia \nand poor endometrial receptivity. 54,55 It is thought \nthat obesity acts on female reproductive function \nthrough hyperinsulinaemia and, consequently, through \nits effect on androgen production. Some authors \nhave argued that insulin resistance is a key factor in \nexplaining the association between obesity, PCOS \nand recurrent miscarriages. 56 Moreover, evidence \nhas shown a possible association between insulin \nresistance, hyperhomocysteinaemia and the risk of \nPCOS.57 Recent studies have highlighted the pres-\nence of hypofibronolysis associated with high levels \nof PAI-1 as being a potential cause of RPL in women \nwith PCOS.58 The effects of elevated PAI-1 may also \nbe increased by elevated homocysteine, eventually \nleading to thrombosis. Moreover, plasma PAI-1 levels \nare associated with dyslipidaemia, hyperinsulinaemia \nand hypertension, three factors that contribute to the \nestablishment of hyperhomocysteinaemia.59 Hence, \nPCOS involves several confounding factors that \nmay contribute, individually or in a combination, to \nthrombosis and, eventually, may lead to RPL. \nMetformin treatment of PCOS patients decreases \ninsulin resistance, thus improving ovulation cycles \nand, therefore, conception rates in infertile women.60 \nMetformin has shown benefit in reducing the risk \nof miscarriage in women with a history of RPL and \nan abnormal glucose tolerance test result,61 but it is \nuncertain whether it decreases the rate of miscar-\n\nHormonal causes of recurrent pregnancy loss 319\nriage in PCOS patients as no proper RCT has been \nconducted.\n6. hyperandrogenIsM\nHyperinsulinaemia and hyperandrogenaemia \nare closely associated. Elevated androgens have \ndetrimental effects on endometrial development and \ndecrease oocyte and embryo viability, or they may have \nan indirect effect via the insulin pathways, perhaps \nvia the insulin-like growth factor.62 The presence of \nan independent link between hyperandrogenaemia \nand RPL remains contentious. Several studies have \ninvestigated the androgen levels of women with RPL \nwith conflicting results as to whether or not an associa-\ntion exists. In the past, two studies have shown that \nandrogen levels in the follicular phase are higher in \nwomen dealing with RPL than in normal fertile con-\ntrols.63 The apparent controversy is mainly attributed \nto the considerable variation in the specific androgens \nmeasured and to the phase of the menstrual cycle in \nwhich the measure was applied. The free testosterone \nand free androgen index (FAI) are considered to \nbe the most sensitive methods and the assessment \nshould be applied in the early follicular phase.64 The \nmost recent large-scale study with measurement of \nthe FAI in the early follicular phase demonstrated \na significantly increased risk of miscarriage with \nincreasing FAI (FAI >5). 65 The prevalence of an \nelevated FAI was found to be 11% in the group of \npatients with RPL. Thus, hyperandrogenaemia seems \nto contribute to the pathology of RPL, but further \nstudies should be conducted in order to determine \nwhether therapeutic intervention to reduce the FAI \nimproves the outcome in this group of women. \n7. reduced ov arIan reserve \nWomen with altered ovarian reserve markers, such \nas low AMH levels and a lower number of antral fol-\nlicles (AFC) counted by transvaginal utrasonography \n(TVUS), usually experience a poor response to ovar-\nian stimulation for assisted reproductive technology \n(ART). Whether the reduction in the primordial \nfollicular pool is associated with an alteration in oo-\ncyte quality is a debated matter. Diminished ovarian \nreserve (DOR), defined as altered ovarian reserve \nmarkers with regular menstrual cycles, can be seen in \nthe general population of young women conceiving \nnaturally and is not necessarily considered a pathologi-\ncal entity.66 Diminished ovarian reserve can also result \nfrom a partial destruction of the primordial follicular \npool due to surgical interventions on the ovaries \n(oophorectomy, cystectomy), to chemotherapies, \nmedical conditions such as autoimmune oophoritis or \nto genetic factors such as permutations in the FMR1 \ngene. Moreover, ovarian aging leads to a decrease \nin the ovarian reserve associated with an increase in \nfoetal aneuploidy and pregnancy losses, which renders \nthe study of a direct relationship between DOR and \npregnancy loss difficult. Several publications however \nreport an increased first trimester miscarriage rate in \nwomen with DOR.67-69 The evidence from the current \nmedical literature is questionable for several reasons: \n1) most studies are carried out in an infertile popula-\ntion, 2) the study sizes are small, 3) none of studies \ndifferentiates between DOR of different origins. \nFurther studies in a non-infertile population with \na characterization of the aetiology of DOR should \nbe conducted before concluding on the association \nbetween DOR and pregnancy loss.\n9. conclusIons\nThe definition, diagnosis and treatment of patients \nwith a history of RPL remains difficult. The majority \nof sporadic losses before the 10 th week of gestation \nresult from random numeric chromosome errors. \nHowever, endocrine disorders play an important role \nin RPL, especially in the early stages of gestation. \nLarge, well-defined clinical trials focused on appro-\npriate testing and treatment of endocrine disorders \nin RPL are essential to provide additional data on \nthis challenging clinical condition. \nconflIct of Interest\nNone.\nreferences\n 1. Smith mL, Schust DJ, 2011 endocrinology and recurrent \nearly pregnancy loss. Semin reprod med 29: 482-490.\n 2. Jacobs PA, Hassold T 1987 Chromosome abnormali-\nties: origin and etiology in abortions and livebirths. In: \nV ogel F, Sperling K, (eds) Human genetics. Berlin: \nSpringer-Verlag; pp, 233-244.\n\n320 N. PLuCHINo eT AL\n 3. Practice Committee of the American Society for re-\nproductive medicine, 2012 evaluation and treatment of \nrecurrent pregnancy loss: a committee opinion. Fertil \nSteril 98: 1103-1111.\n 4. mills JL, Simpson JL, Driscoll Sg, et al, 1988 Inci-\ndence of spontaneous abortion among normal women \nand insulin-dependent women whose pregnancies were \nidentified within 21 days of conception. N engl J med \n319: 1617-1623.\n 5. goddijn m, Joosten JH, Knegt AC, et al, 2004 Clini-\ncal relevance of diagnosing structural chromosome \nabnormalities in couples with repeated miscarriage. \nHum reprod 19: 1013-1017.\n 6. Kalinka J, Szekeres-Bartho J, 2005 The impact of dy-\ndrogesterone supplementation on hormonal profile and \nprogesterone-induced blocking factor concentrations in \nwomen with threatened abortion. Am J reprod Immunol \n53: 166-171.\n 7. Szekeres-Bartho J, Balasch J, 2008 Progestagen therapy \nfor recurrent miscarriage. Hum reprod update 14: 27-\n35.\n 8. Porter TF, Scott Jr, 2005 evidence-based care of recur-\nrent miscarriage. Best Pract res Clin obstet gynaecol \n19: 85-101.\n 9. Arredondo F, Noble LS, 2006 endocrinology of recur-\nrent pregnancy loss. Semin reprod med 24: 33-39.\n 10. Insler V , 1992 Corpus luteum defects. Curr opin obstet \ngynecol 4: 203-211.\n 11. Hensleigh PA, Fainstat T, 1979 Corpus luteum dys-\nfunction: serum progesterone levels in the diagnosis \nand assessment of therapy for recurrent and threatened \nabortion. Fertil Steril 32: 396-400. \n 12. Cumming DC, Honore LH, Scott Jz, 1985 The late luteal \nphase in infertile women: comparison of simultaneous \nendometrial biopsy and progesterone levels. Fertil Steril \n43: 715-719. \n 13. Arck PC, rücke m, rose m, et al, 2008 early risk \nfactors for miscarriage: a prospective cohort study in \npregnant women. reprod Biomed online 17: 101-113.\n 14. Van der gaast mH, Beier-Hellwig K, Fauser BC, et al, \n2003 endometrial secretion aspiration prior to embryo \ntransfer does not reduce implantation rates. reprod \nBiomed online 7: 105-109.\n 15. Altmäe S, esteban FJ, Stavreus-evers A, et al, 2014 \nguidelines for the design, analysis and interpretation \nof ‘omics’ data: focus on human endometrium., Hum \nreprod update 20: 12-28.\n 16. Karamardian Lm , g rimes DA, 1992 Luteal phase \ndeficiency effect of treatment on pregnancy rares. Am \nJ obstet gynecol 167: 1391-1398.\n 17. Daya S, 1989 e fficacy of progesterone support for \npregnancy in women with recurrent miscarriage: a \nmeta-analysis of controlled trials. Br J obstet gynaecol \n96: 275-280.\n 18. oates-Whitehead  rm, Haas Dm, Carrier JA, 2003 \nProgestogen for preventing miscarriage. Cochrane \nDatabase Syst rev 4: CD003511.\n 19. Fatemi Hm, Popovic-Todorovic B, Papanikolaou e, et \nal, 2007 An update of luteal phase support in stimulated \nIVF cycles. Hum reprod update 13: 581-590.\n 20. Fauser BC, Devroey P, 2003 reproductive biology and \nIVF: ovarian stimulation and luteal phase consequences. \nTrends endocrinol metab 14: 236-242.\n 21. Boomsma Cm, Kavelaars A, eijkemans mJ , et al, 2010 \novarian stimulation for in vitro fertilization alters the \nintrauterine cytokine, chemokine, and growth factor \nmilieu encountered by the embryo. Fertil Steril 94: \n1764-1768. \n 22. Frasor J, Park K, Byers m, et al, 2001 Differential roles \nfor signal transducers and activators of transcription \n5a and 5b in PrL stimulation of eralpha and erbeta \ntranscription . mol endocrinol 15: 2172-2181.\n 23. mcNatty KP, Sawers rS, 1975 relationship between \nthe endocrine environment within the graafian follicle \nand the subsequent secretion of progesterone by human \ngranulosa cells in culture. J endocrinol 66: 391-400.\n 24. Hirahara F, Andoh N, Sawai K, Hirabuki T, u emura \nT, minaguchi H, 1998 Hypecprolacrinemic recurrent \nmiscarriage and results of randomized bromocriptine \ntreatment trials. Fertil Steril 70: 246-252.\n 25. Li TC, Spuijbroek mDeH, Tuckerman e, Anstie B, \nLoxley m, Laird Sm, 2000 endocrinological and en-\ndometrial factors in recurrent miscarriage.Br J obstet \ngynaecol 107: 1975-1980.\n 26. glinoer D, 1998 Thyroid hyperfunction during preg-\nnancy. Thyroid 8: 859-864.\n 27. millar LK, Wing DA, Leung AS, Koonings PP, montoro \nmN, mestman JH, 1994 Low birth weight and preec-\nlampsia in pregnancies complicated by hyperthyroidism. \nobstet gynecol ; 84: 946-949.\n 28. Anselmo J, Cao D, Karrison T, Weiss re, refetoff S, \n2004 Fetal loss associated with excess thyroid hormone \nexposure. JAmA 292: 691-695.\n 29. Allan WC, Haddow Je, Palomaki ge, et al, 2000 ma-\nternal thyroid deficiency and pregnancy complications: \nimplications for population screening. J med Screen 7: \n127-130.\n 30. Abalovich m, gutierrez S, Alcaraz g, maccallini g, \ngarcia A, Levalle o, 2002 overt and subclinical hypo-\nthyroidism complicating pregnancy. Thyroid 12: 63-68.\n 31. Wakim AN, Polizotto SL, Buffo mJ, marrero mA, \nBurholt Dr, 1993 Thyroid hormones in human fol-\nlicular fluid and thyroid hormone receptors in human \ngranulosa cells. Fertil Steril 59: 1187-1190.\n 32. Clifford K, rai r, Watson H, regan L, 1994 An in-\nformative protocol for the investigation of recurrent \nmiscarriage: preliminary experience of 500 consecutive \ncases. Hum reprod 9: 1328-1332.\n 33. Baloch z, Carayon P, Conte-Devolx, et al, 2003 guide-\nlines Committee, National Academy of Clinical Bio-\nchemistry. Laboratory medicine practice. guidelines. \nLaboratory support: for the diagnosis and monitoring \n\nHormonal causes of recurrent pregnancy loss 321\nof thyroid disease. Thyroid 13: 3-126.\n 34. Poppe K, Velkeniers B, glinoer D, 2007 Thyroid disease \nand female reproduction. Clin endocrinol 66: 309-321.\n 35. marai I, Carp H, Shai S, Shabo r, Fishman g, Shoenfeld \nY , 2004 Autoantibody Panel Screening in recurrent \nmiscarriages. Am J reprod Immunol 51: 235-240. \n 36. Stagnaro-green A, roman SH, Cobin rH, el-Harazy \ne, Alvarez-marfany m, Davies TF, 1990 Detection of \nat-risk pregnancy by means of highly sensitive assays \nfor thyroid autoantibodies. JAmA 264: 1422-1425.\n 37. roberts J, Jenkins C, Wilson r, et al, 1996 recurrent \nmiscarriage is associated with increased numbers of \nCD5/20 positive lymphocytes and an increased incidence \nof thyroid antibodies. eur J endocrinol 134: 84-86.\n 38. Twig g, Shina A, Amital H, Shoenfeld Y , 2012 Patho-\ngenesis of infertility and recurrent pregnancy loss in \nthyroid autoimmunity. J Autoimmun 38: J275-281.\n 39. Konova e, 2010 The role of NK cells in the autoim-\nmune thyroid disease-associated pregnancy loss. Clin \nrev Allergy Immunol 39: 176-184. \n 40. Lejeune B, grun JP, de Nayer P, Servais g, glinoer \nD, 1993 Antithyroid antibodies underlying thyroid \nabnormalities and miscarriage or pregnancy induced \nhypertension. Br J obstet gynaecol 100: 669-672.\n 41. rushworth FH, Backos m, rai r, Chilcott IT, Baxter \nN, regan L, 2000 Prospective pregnancy outcome in \nuntreated recurrent miscarriers with thyroid autoanti-\nbodies. Hum reprod 15: 1637-1639.\n 42. Prummel mF, Wiersinga Wm, 2004 Thyroid autoim-\nmunity and miscarriage. eur J endocrinol 150: 751-755.\n 43. gabbe Sg, graves Cr, 2003management of diabetes \nmellitus complicating pregnancy. obstet gynecol  102: \n857-868.\n 44.  Sibai Bm, 2000 risk factors, pregnancy complications, \nand prevention of hypertensive disorders in women \nwith pregravid diabetes mellitus. J matern Fetal med \n9: 62-65.\n 45. melamed N, Hod m , 2009 Perinatal mortality in preges-\ntational diabetes. Int J gynaecol obstet 104: Suppl 1: \n20-24.\n 46. Lucas mJ, Leveno KJ, Williams mL, raskin P, Whalley \nPJ, 1989 e arly pregnancy glycosylated hemoglobin, \nseverity of diabetes, and fetal malformations. Am J \nobstet gynecol 16: 426-431.\n 47. Kitzmiller JL, Wallerstein r, Correa A, Kwan S, 2010 \nPreconception care for women with diabetes and preven-\ntion of major congenital malformations. Birth Defects \nres A Clin mol Teratol 88: 791-803.\n\t48.\tGutaj\tP ,\tZawiejska\tA,\tWender‑Ożegowska\tE,\tBrązert\t\nJ, 2013 maternal factors predictive of first-trimester \npregnancy loss in women with pregestational diabetes. \nPol Arch med Wewn 123: 21-28.\n 49. rai r, regan L, 2006 recurrent miscarriage. Lancet \n368: 601-611.\n 50. Nicholson W, Baptiste-roberts K, 2011 oral hypo-\nglycaemic agents during pregnancy: The evidence for \neffectiveness and safety. Best Pract res Clin obstet \ngynaecol 25: 51-63.\n 51. rai r, Backos m, rushworth F, regan L, 2000 Poly-\ncystic ovaries and recurrent miscarriage--a reappraisal. \nHum reprod 15: 612-615.\n 52. rotterdam eSHre/ASrm-Sponsored PCoS Consen-\nsus Workshop group, 2004 revised 2003 consensus \ndiagnostic criteria and long-term health risks related \nto polycysic ovary syndrome. Fertil Steril 81: 19-25.\n 53. Christiansen oB, 2006 evidence-based investigations \nand treatments of recurrent pregnancy loss. Curr opin \nobstet gynecol 18: 304-312.\n 54. Wang JX, Davies mJ, Norman rJ, 2002 obesity increases \nthe risk of spontaneous abortion during infertility treat-\nment. obes res 10: 551-554.\n 55. Angioni S, Sanna S, magnini r, melis gB, Fulghesu \nAm, 2011 The quantitative insulin sensitivity check \nindex is not able to detect early metabolic alterations \nin young patients with polycystic ovarian syndrome. \ngynecol endocrinol 27: 468-474.\n 56. Tian L, Shen H, Lu Q, Norman rJ, Wang J, 2007 Insulin \nresistance increases the risk of spontaneous abortion \nafter assisted reproduction technology treatment. Clin \nendocrinol metab 92: 1430-1433.\n 57. Schachter m, raziel A, Friedler S, Strassburger D, Bern \no, ron-el r, 2003 Insulin resistance in patients with \npolycystic ovary syndrome is associated with elevated \nplasma homocysteine. Hum reprod 18: 721-727.\n 58. Sun L, Lv H, Wei W, zhang D, guan Y , 2010 Angioten-\nsin-converting enzyme D/I and plasminogen activator \ninhibitor-1 4g/5g gene polymorphisms are associated \nwith increased risk of spontaneous abortions in polycystic \novarian syndrome. J endocrinol Invest 33: 77-82.\n 59. Bastard JP, Piéroni L, Hainque B, 2000 relationship \nbetween plasma plasminogen activator inhibitor 1 and \ninsulin resistanceDiabetes metab res rev 16: 192-201.\n 60. Tang T, Lord Jm, Norman rJ, Yasmin e, Balen AH, \n2012 Insulin-sensitising drugs (metformin, rosiglita-\nzone, pioglitazone, D-chiro-inositol) for women with \npolycystic ovary syndrome, oligo amenorrhoea and \nsubfertility. Cochrane Database Syst rev 5: CD003053.\n 61. zolghadri J, Tavana z, Kazerooni T , et al, 2008 rela-\ntionship between abnormal glucose tolerance test and \nhistory of previous recurrent miscarriages, and beneficial \neffects of metformin in these patients: a prospective \nclinical study. Fertil Steril 90: 727-730.\n 62. okon mA, Laird Sm, Tuckerman e, Li TC, 1998 Serum \nandrogen levels in women who suffer recurrent miscar-\nriage and their correlation with markers of endometrial \nfunction. Fertil Steril 69: 682-690.\n 63. Bussen S, Sutterlin m, Steck T, 1999 endocrine ab-\nnormalities during the follicular phase in women with \nrecurrent spontaneous abortions. Hum reprod 14:18-20.\n 64. rotterdam eSHre/ASrm-Sponsored PCoS consensus \nworkshop group, 2004 r evised 2003 consensus on \ndiagnostic criteria and long-term health risks related \n\n322 N. PLuCHINo eT AL\n 67. Trout SW, Seifer DB, 2000 Do women with unexplained \nrecurrent pregnancy loss have higher day 3 serum FSH \nand estradiol values? Fertil Steril 74: 335-337.\n 68. Levi AJ, raynault mF, Bergh PA, et al, 2001 repro-\nductive outcome in patients with diminished ovarian \nreserve. Fertil Steril 76: 666-669.\n 69. elter K, Kavak zN, gokaslan H, et al, 2005 Antral fol-\nlicle assessment after down-regulation may be a useful \ntool for predicting pregnancy loss in in vitro fertilization \npregnancies. gynecol endocrinol 21: 33-37.\nto polycystic ovary syndrome PCoS. Hum reprod 19: \n41-47.\n 65. Cocksedge KA, Saravelos SH, Wang Q, Tuckerman \ne, Laird Sm, Li TC, 2008 Does free androgen index \npredict subsequent pregnancy outcome in women with \nrecurrent miscarriage? Hum reprod 23: 797-802.\n 66. massé V , Ferrari P, Boucoiran I , et al, 2011 Normal \nserum concentrations of anti-mullerian hormone in a \npopulation of fertile women in their first trimester of \npregnancy. Hum reprod 26: 3431-3436.","source_license":"CC0","license_restricted":false}