Abortion

(A)      INTRODUCTION

(B)      DEFINITIONS

(C)     PHYSIOLOGY OF IMPLANTATION - IMMUNOLOGICAL ASPECTS

(D)     ETIOLOGY

(E)      PATHOGENESIS

(F)      CLINICAL FEATURES

(G)     CONCLUSION

(H)     REFERENCES

(A) INTRODUCTION

A spontaneous abortion is defined as any recognized, involuntary pregnancy loss occurring before fetal viability, 24 weeks. Approximately 80% of all spontaneous abortions occur before 12 weeks and are called early abortions. The rest occur between the thirteenth and the twenty-fourth week and are called late abortions. Majority of early abortions correspond to anembryonic pregnancies or blighted ova, whereas most of the abortions with a fetus present occur in the second trimester.

It has been seen that only 57% of all conceptions advance beyond 20 weeks. Of those lost, 75% occur before implantation and only 25% are clinically recognizable¹. Recent investigations have found an overall pregnancy loss of 31% with 22% occurring before implantation².

(B)DEFINITIONS

Blighted ova—Early pregnancy losses in which fetal development is not observed with ultrasound and fetal tissue is absent on the histologic examination of the products of conception.

Early fetal demise—Early pregnancy losses in which fetal development is clearly observed by ultrasound and fetal tissue is found on the histologic examination.

Complete abortion—Cases in which ovum along with chorion/placenta is completely expelled.

Incomplete abortion—Cases in which placenta or chorion is wholly or partly retained within the uterus.

Primary aborters—Those without any previous live births. Primary miscarriage tends to be partner specific.

Secondary aborters—Those who have had one or more children.

Approximately 25% of pregnant women will threaten miscarriage but a smaller percentage actually complete it.

Most spontaneous abortions occur in the first trimester, with the median between 8 and 9 weeks gestation3,4. If a fetal heart is detected in the first trimester (5-14 weeks), 95-98% gestate successfully^3,4,5. Most of those who abort show either absent fetal heart or an empty gestational sac. The frequency of chromosomal anomalies in material from recurrent aborters is approximately half that reported for sporadic aborters⁶.

(C)     PHYSIOLOGY OF IMPLANTATION—IMMUNOLOGICAL ASPECTS

         Many Pathological conditions of pregnancy encountered in clinical obstetrical practice e.g. miscarriage. IUGR & pre-eclampsia, are all due to failure of the normal controlling mechanisms on trophoblast migration, leading to insufficient invasion and vascular adaptation with the final result of poor perfusion of the feto-placental unit.

            Allorecognition

         NK cells tend to be cylolytic to largest cells which are deficient in MHC class I molecules. Thus, T cells recognise ‘non-self’, NK cells recognise ‘missing self’ of the absence of self. Neoplastic transformation or viral infection can lead to downregulation of class I molecules and it is possible that NK cells will provide the defence in these situations.

         Therefore, the phylogenetically more primitive NK cells have been designed for internal surveillance and to monitor the integrity of self while T cells have been developed to guard against non-self external aggressors.

            Immunology of the maternal fetal interaction

         The defensive cells present at the feto-maternal interface in the human uterus are macrophages and NK cells which are cells of innate system. B cells are absent and T cells sparce. Unlike the artificial situation of an organ transplant, the intermingling of cells from genetically different individuals occurs naturally during placentation. Also unlike a transplant, the mammalian conceptus is not rejected.

         Despite the original definition of NK recognition as ‘Non MHC restricted’ it's now generally accepted that NK cells are capable of allorecognition and that human NK cells can broadly discriminate between some MHC class I antigens. Human extravillous trophoblast expresses class I HLA-G and may be HLA-C also, but not the usual class I HLA-A, B like other somatic cells. Perhaps these class I antigens provide the recognition molecules by which decidual NK cells detect the allogeneic nature of trophoblast. A recent discovery is that NK cells express type II membrane glycoproteins which contain a C-type lectin-like domain.⁷

         Some of these NKreceptors have been found to recognise oligosaccharide ligands which deliver stimulatory or inhibitory signals to the NK cell⁸. A search for trophoblast-specific carbohydrate antigens might prove very fruitful. There is already evidence that sialylated type I blood gp related carbohydrates are preferentially expressed on certain human extravillous trophoblast sub-populations⁹. Thus, the concept of ‘self’ and ‘non-self’ in terms of NK cells may be very different to how it is generally perceived in classical T cell immunology. Decidual NK cells also produce colony stimulating factors, among other cytokines, so the possibility arises that during evolution trophoblast has managed to utilise certain aspects of the maternal cytokine response to sustain its own development. These cytokines will provide either positive or negative signals to ensure the correct degree of trophoblast migration within the uterine environment.

         Thus, although the human placenta is genetically foreign to the mother, the characteristics of its interaction with the uterus are more akin to those observed between unrelated primitive invertebrates than between mammalian allograft & host as seen in contemporary transplantation immunology.

            Development of trophoblast

         The formation of the human placenta begins at the time the blastocyst penetrates the maternal endometrium to become rapidly embedded in the uterine stroma.

            Early trophoblast development

         The first element to differentiate in the morula is the trophoectoderm. As the morula is converted into a blastocyst, trophoectoderm cells form a layer encircling blastocyst with the inner cell mass at one pole. The blastocyst inserts itself between the epithelial cells of the uterine mucosa on the 6th-7th day after ovulation.

         By the 8th day, trophoblast has differented into an outer multinucleated syncytiotrophoblast (primitive syncytium) and an inner layer of primitive mononuclear cytotrophoblast.

         By the 9th day, vacuoles or lacunae appear in the syncytium and these rapidly enlarge and fuse to communicate with each other. The establishment of a potential uteroplacental circulation occurs when maternal venous capiliaries are eroded by the syncytium so that blood can seep into the lacuna system. The lacunae will eventually become the intervillous space of the definitive placenta.

         By 12 to 13 days after fertilization, the blastocyst is completely embedded in the decidual stroma and the uterine surface epithelium has grown over the overlying defect.

         After this initial phase of nidation, trophoblast differentiation may be considered to occur along two main pathways-villous & extravilllous. Villous trophoblast ultimately covers all the chorionic villi of the definitive placenta and is concerned with transport of oxygen and nutrients from mother to child. Extravillous trophoblast migrates deep into the uterine mucosa as far as the myometrium.

            Villous trophoblast

         The earliest sign of villous development is the appearance of buds of cytotrophoblast protruding into the primitive syncytium in the second week of gestation. Penetration of solid cytotrophoblast buds by mesenchymal cells convert them into secondary villi. Tertiary villi are formed when the mesenchymal core is supplied by embryonic blood vessels.

         The embryonic vessels link with the vessels in the mesoderm of the chorionic plate, and ultimately with vessels in the fetus so that by the fourth week the feto-placental circulation is established. The mesodermal core of the chorionic villi is derived from cells of the inner cell mass which originate from the caudal margin of the embryonic disc.¹⁰

         During the second and third months of pregnancy, rapid extension and branching of the villi result in the villous tree characteristic of the mature placenta. In the first 3 months there is an obvious two-layered trophoblast covering of the villi. The inner cytotrophoblast cells sit on a basement membrane covered by the outer syncytiotrophast layer. Syncytiotrophoblast lines the entire intervillous space and thus comes into contact with maternal blood in a manner similar to endothelial cells. In the first trimester, the core of the villus contains loose mesenchymal cells, embryonic blood vessels and fetal macrophages (Hofbauer cells) Hofbauer cells appear very early in gestation (by the third week) and increase in number throughout the first trimester. As gestation proceeds, the villi become-smaller, the cytotrophoblast layer appears less prominent as the cells become widely separated and the syncytiotrophoblast becomes more irregular and narrow with clustering of the nuclei. The stroma contains the same elements, but the fetal capillaries become much more prominent and come to lie directly under the basement membrane at the periphery of the villus. By the last few weeks of gestation, thinned areas of syncytiotrophoblast form only a narrow rim over the fetal capillaries to become vasculosyncytial membranes, which are specialized zones for the facilitation of gas transfer across the placenta.

            Extravillous trophoblast

         (a) Interstitial trophoblast—After the second week, the budding cytotrophoblast which will ultimately form the villous tree breaks through the primitive syncytium. At the tips of these early villi, cytotrophoblast cells remain as a solid core known as the cytotrophoblast cell columns which fix the villi to the decidua. These are known as anchoring villi. The cytotrophoblast cell columns spread laterally and fuse with neighbouring columns to form a cytotrophoblast shell which encircles the entire embryonic sac. Once this shell has come into physical contact with decidua, isolated interstitial trophoblast cells stream into the uterine mucosa from the shell. After invasion into decidua has occurred the interstitial trophoblast cells become isolated fusiform pleomorphic cells from round uniform cohesive cells.

         On the maternal side of the cytotrophoblast shell in the region of the original primitive syncytium, two irregular narrow strips of fibrinoid necrosis appear known as Rohr’s layer and Nitabuch’s layer. Apart from these small foci of necrosis at the placental-decidual junction, the migration of trophoblast into maternal tissues is remarkable for the lack of tissue destruction or inflammatory response. The invading interstitial trophoblast cells apparently move towards the decidual spiral arteries as trophoblast is always concentrated around the arteries with a much sparser number in the intervening areas. Arteries in the decidua basalis which are surrounded by trophoblast show endothelial swelling and a characteristic destruction of the muscular media which is replaced with ‘fibriroid’ material. This disruption of vessel walls is not found in decidual vessels away from the implantation site. By 8 weeks of pregnancy interstitial trophoblast has extensively colonised the full thickness of the uterine mucosa to reach the decidual-myometrial junction. As the cells move deeper into the decidua, the trophoblast cells become rounder and multinucleated - these are known as placental bed giant cells.

         During the second trimester diffuse infiltration of the inner myometrium by placental bed giant cells like trophoblast is seen causing destruction of the musculo-elastic tissue of the intramyometrial segments of the spiral arteries.

            (b) Endovascular trophoblast—Cytotrophoblast shell provides the origin not only for the interstitial trophoblast but also for endovascular trophoblast. Where the shell comes to lie over the openings of the maternal spiral arteries, the trophoblast cells form a plug seemingly occluding the lumen. From these plug-like areas of the shell, trophoblast moves in a retrograde manner down the spiral arteries ‘like wax dripping down a candle.’¹¹

         Endovascular trophoblast is only seen in vessels which have already been surrounded by interstitial trophoblast and undergone medial disruption. Eventually, trophoblast cells become incorporated into the fibrinoid material in the wall, although the relative contribution of endovascular trophoblast and interstitial trophoblast to this process is uncertain, and may differ in the different segments of the vessel. Interestingly, the walls of the decidual veins are not similarly invaded by trophoblast. Syncytial knots, detached from chorionic villi into the intervillous space, are deported into the maternal circulation via these veins to be trapped in the blood vessels of the lungs where they are eventually lysed.

            Immunohistology of trophoblast

         Many phenotypic differences are evident among the trophoblast sub populations, particularly between villous and extravillous trophoblast. Multinuclear cell formation is the terminal state of both villous and extravillous developmental pathways-but these two end products are not identical. While many placental proteins are detected in syncytiotrophoblast, placental bed giant cells are largely unproductive. In addition, syncytiotrophoblast is HLA class I-(W6/32-) and placental bed giant cells are class I+ (W6/32+), indicating that the immunology of the trophoblast-maternal blood interface is likely to be different from the trophoblast-uterine interface.

            Second and third trimesters

         As pregnancy proceeds, there is a shift in the distribution of changes in the placental bed. In the second half of pregnancy, there is no further trophoblast penetration of the myometrium although vascular transformation by trophoblast appears to continue throughout gestation.¹² For the first 2 months of gestation, villi cover the entire surface of the chorionic sac. As the conceptus expands, the villi beneath the decidua capsularis degenerate, leaving a layer of trophoblast and mesenchyme known as chorion leave. The chorion leave fuses with the decidua parietalis with obliteration of the uterine lumen and regression of the uterine surface epithelium-transport of substances and even maternal cells in the event of infection may occur across this chorio-decidual junction.

            Genomic imprinting and trophoblast development

         Genomic imprinting is the phenomenon whereby genes have differential expression depending on the parental source from which they are inherited. The parental gene that is silent is said to be imprinted. GYNOGENOMES (both sets of haploid chromosomes are derived from the mother) created by replacement of the male pro-nuclei with female pronuclei after fertilization leads to the development of relatively normal sized embryos with poor development of extraembryonic membranes and the placenta. These conceptuses remain viable until the early somite stage, and then undergo involution.

         In contrast ANDROGENOMES (paternally derived chromosomes) created by transplantation of male pronuclei into ova in which the female pronuclei have been removed result in conceptuses with severely stunted embryos, but with well developed extra embryonic trophoblast tissues. Thus, the paternally transmitted genome appears to provide the necessary genetic information for trophoblast growth whereas the maternally transmitted genome is associated with the development of the embryo itself. The classic human example of genome’s imprinting is complete hydatidiform mole. These tumors contain two paternally derived sets of haploid chromosomes (diploid androgenomes).¹³ There is no associated fetus, grape-like distension of chorionic villi together with excessive placental tissue and trophoblast proliferation.

         Genomic imprinting has been viewed as a mechanism where the paternal genes, by having a positive effect on the growth of the placenta, give the fetus a greater chance of survival while the maternal genes, by limiting the growth of the placenta, will reduce the nutritional burden of each pregnancy and thus promote the mother’s chances of a successive pregnancy.^14,15

         In humans four genes are known to be imprinted, H19, IGF-II, WT1 and SNRPN¹⁶. At present it is not known what imprinted genes are important for trophoblast development or for implantation.¹⁷

         The gene encoding insulin-like growth factor II (IGF-II) is maternally imprinted.¹⁸

         Immunohistological studies on human placentae have shown that those trophoblast cells with mitotic activity express receptors for IGF, so the paternally expressed gene encoding this growth factor could be one important element in trophoblast development.

         H19 gene is paternally imprinted.¹⁹ It is expressed by normal trophoblast²⁰ as well as by trophoblast derived from complete moles,²¹ placental site trophoblast tumours and choriocarcinomas.²² Since complete moles are androgenetic, it would appear that the parental source of certain imprinted genes can sometimes be reversed. Loss of imprinting of IGF-II and H19 is also commonly associated with choriocarcinomas.²³

         Wilms’ tumour suppressor gene WT1 has no defined role in placental development.

         SNRPN (small nuclear ribonucleoprotein polypeptide N) is the fourth imprinted human gene.²⁴ To date, there is no evidence that imprinting of this gene affects the reproductive process.

         No data are available for MHC imprinting in the human placenta. Normal trophoblast development is likely to be subjected to the opposing influence of maternally expressed genes. The X chromosome may be especially important in this regard. The paternal X chromosome is preferentially inactivated in human trophoblast²⁵whereas X inactivation is random in somatic tissues, affecting the paternal and maternal X chromosome equally.²⁶ Since the male trophoblast has a single maternal X and female trophoblast has the paternal X preferentially inactivated, this could be a device whereby some X-linked genes important for trophoblast development could behave as though they were paternally imprinted.

         The gene encoding tissue inhibitor of metalloproteinases (TIMP) which may control trophoblast degradation of extra-cellular matrix and hence trophoblast invasion is X linked. The expression of this gene is a further example of how the mother can exert a balancing influence on trophoblast invasiveness.

         Thus the concept of imprinting as a mechanism designed to control maternal fetal interaction during the evolution of the placental form of reproduction is an attractive one.

            Role of retroviruses in trophoblast development

         Retroviruses are RNA enveloped animal viruses which contain a single-stranded RNA genome and an RNA dependent DNA polymerase to transcribe the viral RNA into a double-stranded insertion element for integration into host DNA.^27,28 Genetic mapping studies have identified three genes in the retroviral genome. In order from the 5’ to 3’ end, these are :

         1.   a gag gene which codes for viral structural core proteins.

         2.   a pol gene which codes for reverse transcriptise.

         3.   an env gene which codes for the major envelope proteins. Flanking these are the 5’ and 3’ long terminal repeat (LTR) region which contain sequences essential for viral expression. The inserted retroviral genome is known as the provirus, which can be expressed with the production of viral particles. These viral particles can be classified morphologically into four types (A, B, C & D) that can be transmitted horizontally like other viruses (exogenous transmission). However, an important characteristic of retroviruses is that they can remain unexpressed in the host cell, and be transmitted vertically in the genome though the genuline (endogeneous transmission).

         Many of these viruses appear to be selectively expressed in the placenta.²⁹

         Kato et al (1987)³⁰ have demonstrated that three polyadenylated RNAs of 9, 7.3 and 3.5 Kb long of HERV-R are abundant in first trimester and their human chorionic villi, representing 0.03-0.05% of the total mRNA for this tissue.

         Recent in situ hybridisation studies by Boylet al (1993),³¹ who probed with ERV-3 (HERV-R) envelope DNA generated by PCR amplification of genomic ERV-3 sequences, confirm that the syncytiotrophoblast layer of term placentae is the major site of expression. Presence of retroviral-encoded proteins in the placenta has been demonstrated-which include reverse transcriptase, 75 K Da protein and others. In addition to proteins, the presence of actual retroviral particles in the placenta has long been documented.^32,33 The earliest reports are electron microscopic demonstration of C-type viral particles.^34-42 These viruses are normally located on the convoluted plasma membrane at the basal border of the syncytiotrophoblast in apposition with either the underlying villous cytotrophoblast or basement membrane. Viral particles have never been observed at the brush border of the syncytiotrophoblast in contact with maternal blood in the intervillous space. Placenta is especially permissive for retroviral replication⁴³-incidence being 35-80%.

         It can be concluded that :

         1.   Retroviral transcripts are more abundant in chorionic villous tissue than in associated embryos.

         2.   Placental extracts contain retroviral reverse transcriptase.

         3.   Antisera raised against various retroviral-related proteins react against placental proteins.

         4.   The reverse is also the case in that antisera raised against trophoblast membrane antigens recognise shared determinants on retroviral infected cells.

         5.   Pregnant women show cell mediated & humoral immunity against retroviral proteins, indicating that these proteins are immunogenic in vivo.

         6.   Even complete viral particles may be expressed since C-type particles are frequently observed in human placenta.

         Further data is required to determine how retroviruses are potentially capable of playing an important role in placenta development.

         The factors why retroviruses are especially transcriptionally active in the placenta include hormonal and immunological.²⁷ Estrogen (17B-estradiol) has been reported to induce the production of retroviral proteins and RNA-directed situations of immunological activity. The implantation site represents their mutual recognition of each others could, in same way, trigger retroviral expression in the placenta as well as the uterus.

         It has been suggested that the fusion of villous cytotrophoblast to form the overlying layer of syncytiotrophoblast could be mediated by some retrovirual fusion protein since retroviruses are frequent inducers of syncytium formation.⁴⁴

         Besides affecting the host cell’s own expression of class I antigen, endogenous retroviruses may also encode their own transplantation antigen. In conclusion, it will be appreciated that endogenous retroviruses are potentially capable of playing an important role in placental development.

(D)     ETIOLOGY

         Causes of spontaneous abortions include :

         1.   Genetic abnormalities 50-60%

         2.   Endocrine abnormalities 10-15%

         3.   Chorioamniotic separations 5-10%

         4.   Incompetent cervix 8-15%

         5.   Infections 3-5%

         6.   Abnormal placentation 5-15%

         7.   Immunologic abnormalities 3-5%

         8.   Uterine anatomic abnormalities 1-3%

         9.   Unknown reasons < 5%

            Genetic causes

         Genetic anomalies are the most frequent and important causes of early pregnancy losses. The majority of genetically originated abortions occur before 8 weeks and are blighted ova. Chromosomal abnormalities are found in approximately 80% of blighted ova and 5% to 10% of the abortions in which a fetus is present. The abnormal zygote found in a blighted ovum results from an error in maternal or paternal meiosis I or II, from super fecundation of an egg by two spermatozoids, or from a chromosomal division in the absence of cytoplasmic division.

         The chromosomal abnormalities most frequently found in abortions are⁴⁵—

            I. Autosomal trisomy

            Trisomy 16 is most frequently found in abortus material (@ 60%), and in these cases the sac is completely empty. The trisomy predominantly affects chromosomes 16, 21 and 22, although it may affect any other of the autosomes.

            II. Triploidy

            This anomaly consists of a mean chromosomal count of 69 and occurs in 15% to 20% of all abortions with abnormal chromosomes. In many cases the sac is empty, but if a fetus is present, it has obvious abnormalities (omphalocele syndactyly, cleft lip and palate, etc.). In about 50% of the cases hydropic degeneration of the placenta is present. In triploidy without molar degeneration, the extra chromosomal set is probably maternal.

            III. Monosomy X

            It may be the result of the loss of an X chromosome at the time of fertilization, or it may be the result of nondisjunction during either male or female meiosis. It consists of @ 25% of all abortions with chromosomal abnormalities. About one of every 15 fetuses with 45, X Karyotype will not be aborted and will be identified at birth as an individual with Turner’s syndrome. Some of the abortion specimens with 45, X Karyotypes are sacs containing a small umbilical cord that inserts into an amorphous mass of embryonic tissue.

            IV. Tetraploidy

            This condition probably results from a failure of cytoplasmic division after a chromosome division in the germinal cells. The mean chromosome count is 92, found in 3-6% of blighted ova with abnormal chromosomes. In these cases abortions usually occur very early in pregnancy, and the embryo can not be recognized in the specimen.

            V. Structural rearrangement of the chromosomes

         This group consists of unbalanced translocations and inversions and accounts for 3% to 5% of abortions with abnormal chromosomes. Couples that have abortions with structural rearrangement of the chromosomes should have karyotype analysis because 5% will be carriers of the rearranged chromosomes.

         There are cases of early fetal death in which extensive investigations, including karyotype, give negative results. In these cases the lack of fetal development may be the result of a lethal molecular mutation that can not be identified with the present methods of analysis.

         Animal experiments using insertional mutagenesis (insertion and integration within the genome of a DNA fragment, which causes disruption of the function of one or several genes) have provided evidence indicating that inactivation of the genes containing the information for collagen synthesis causes an arrest in development and embryonic death⁴⁶. Fetal death may also be the result of mutations affecting genes that control the expression of other genes at the transcriptional level (Homeo box genes) or mutations that cause excessive concentration of products and embryonic cell toxicity.

            Endocrine Abnormalities

            I. Progesterone deficiency

         Most of the evidence incriminating progesterone deficiency in early pregnancy loss comes from studies demonstrating that luleal phase deficiency occurs more frequently in patients with recurrent abortions than in control patients^47-49. A rigorous diagnosis of corpus luteum defect requires histologic confirmation of endometrium out of phase by 2 or more days during the secretory period of the menstrual cycle. Because endometrial biopsies are not obtained during pregnancy, the only possible method of documenting a corpus luteum deficiency during gestation is by measuring the serum progesterone concentration⁵⁰. How ever, progesterone production by the corpus luteum is pulsatile and is characterized by marked fluctuations in serum levels⁵¹. As a result, many studies have shown a poor correlation between serum progesterone levels and endometrial biopsy findings.

         Low serum progesterone concentration is seen in patients with blighted ova⁵². Posing a problem in diagnosing progesterone deficiency. Because blighted ova are so frequent, patients with low progesterone concentrations in early pregnancy must be examined for blighted ova rather than treated for corpus luteum deficiency.

         Measurement of 17 hydroxy progesterone a substance produced exclusively by the corpus luteum is a better marker than progesterone. In a recent study, measurements of serum 17-hydroxy progesterone in patients conceiving after the use of ovulation-inducing agents could discriminate patients that miscarried from those that did not. Unfortunately measurement of this substance did not differentiate aborters from non aborters in patients that did not receive ovulation induction.

         Commonest indications for progesterone supplementation are in patients (i) with symptoms of threatened abortion, (ii) conceiving after ovulation induction, (iii) with a serum level < 15 ng/ml, (iv) with a hisory of early pregnancy losses.

            II. Thyroid deficiency

         Usually causes preterm labor after 24 wks. rarely causes early pregnancy loss.

            III. Diabetes

         Diabetes with both an elevated blood glucose and hemoglobin Aic in the first trimester have a significantly increased risk of spontaneous abortion, whereas those with good metabolic control had a risk similar to that of control subjects⁵³.

            IV. Increased Androgen

         A rare endocrine etiology for early pregnancy losses is maternal hyperandrogenicity⁵⁴ with elevated serum levels of testosterone and dehydroepiandrosterone sulfate (DHEAS). Pregnancy loss usually occurs as an early fetal demise at 14 weeks gestational age apparently because of corpus luteum dysfunction.

            V. Polycystic ovary syndrome

         Spontaneous abortions occur more frequently in patients with polycystic ovary syndrome (PCO) than in normal controls⁵⁵. The elevated serum luteinizing hormone concentration seems to have a deleterious effect on the corpus luteum⁵⁶. Pituitary suppression with GnRH agonists followed by hCG administration has been found to be useful in the prevention of this type of miscarriage⁵⁷.

            Infections

         Account for approx 3-5% of cases with early pregnancy loss. They can be :

         1.   Ascending infections facilitated by some degree of cervical incompetence. These infections are usually caused by a mixture of anaerobic and aerobic organisms with predominance of group B streptococci and E. coli. Ascending infections usually recur in subsequent pregnancies, suggesting the presence of a defect in the cervix facilitating uterine invasion by the vaginal flora. Histologic examination of the placenta shows severe chorioamnionitis.

         2.   Hematogenous infections in minority of cases with varicella, parvovirus, rubella, toxoplasmosis, herpes simplex, treponema, listeria, chlamydia, and mycoplasma detected mostly on fetal autopsy or in histologic or bacteriologic examination. Mycoplasma hominis and Ureaplasma urealyticum are occassionally found in cultures of aborted products of conception⁵⁸ which is in contrast with the high frequency of positive cervical and endometrial ureaplasma cultures found in patients with a history of repeated spontaneous abortions or infertility⁵⁹. Diagnosis and treatment of mycoplasma infection before pregnancy prevents recurrent abortion⁶⁰.

            Anatomic abnormalities of the uterus

         Cause approximately 10% to 15% of all abortions with adequate fetal development. Can be divided into

                        Acquired                                                                            Developmental

                  1.   Uterine synechiae                                        1.   Mullerian abnormalities

                  2.   Incompetent cervix                                            a. Septate uterus

                                                                                                b. Bicornuate uterus

                                                                                          2.   DES exposure in utero

                                                                                          3.   Incompetent cervix.

            I. Uterine synechiae

         The association between uterine synechiae and early pregnancy losses has been known since the original work of Asherman in 1947. Uterine synechiae are bandlike structures between the walls of the uterus causing minimal to almost complete obliteration of the uterine cavity. Histologically, these bands are made of fibrous tissue, myometrium, and endometrium. The endometrium around these adhesions is usually atrophic with distorted gland openings.

         In the majority of cases, synechiae are the result of intrauterine infection combined with surgical trauma after the retention of products of conception following abortion or delivery. Schenker and Margalioth⁶¹ found that in 66% and 22% patients, intrauterine adhesions were associated with postabortal and postpartum curettage, respectively. Also 14% of patients with synechiae were found to have a history of multiple pregnancy losses.

         Diagnosis is made by hysterosalpingogram or by direct hysteroscopic vision. Treatment requires surgical division of the fibrous bands, placement of an intrauterine device to avoid contact between the ends of the adhesions and estrogens to stimulate endometrial growth

            II. Mullerian abnormalities

         Septate and bicornuate uteri are associated with early pregnancy losses because of inadequate blood supply to the conceptus when the pregnancy is implanted in the relatively avascular septum. Another mechanism of pregnancy loss in these patients is incompetent cervix, an abnormality that is frequently present in patients with abnormal uterine anatomy. Hence it has been suggested that a prophylactic cervical cerclage should be used in all pregnant patients with congenital uterine anomalies⁶². Other abnormalities such as double uterus and unicornuate uterus are manifested more commonly by preterm labor, characteristically occurring later with each successive pregnancy.

            III. DES exposure in utero

         Approximately 70% of women exposed in utero to DES (diethyl stilboestrol) have a small, T-shaped uterus and an abnormally high frequency of poor pregnancy outcomes⁶³. These patients tend to lose their pregnancies between 20 to 28 weeks because of incompetent cervix. Patients exposed in utero to DES should have a cervical cerclage between 12 and 16 weeks to improve the chances of a normal pregnancy outcome.

            IV. Incompetent cervix

         It is usually the result of cervical trauma, most commonly overzealous mechanical dilation during a pregnancy termination or a diagnostic curettage. Deep cervical lacerations during vaginal delivery and extensive conization for the treatment of cervical dysplasia also cause incompetence. It is also seen in patients with mullerian fusion abnormalities and in patients exposed to DES in utero.

         Diagnosis is made during pregnancy by seeing bulging membranes and painless cervical dilation on perspeculum examination. Diagnosis is suspected in patients with a history of second trimester pregnancy losses and an increased amount of vaginal discharge—USG shows funnelling of membranes into the endocervical canal or diameter of the internal os > 23 mm. In nonpregnant patients, increased compliance of the cervix is demonstrated by passing a No. 8 Hegar or a No. 15 Pratt dilater or a Foley’s catheter filled with 1 ml of water through the internal cervical os.

         Cervical cerclage is placed in such patients at between 12 and 16 weeks transvaginally by Shirodkar or McDonald technique and transabdominally in patients with little cervix left after extensive conization.

            Subchorionic hematomas and

            chorioamniotic separations

         Bleeding occurs between the amnion and the chorion (chorioamniotic separation) or between the chorioamnion and the decidua (Subchorionic hematoma).

         The cause of such bleeding and membrane separation is unknown. Most patients with a severe initial episode do well, and the pregnancy continues without problems after a few days of spotting, but persistent bleeding has a guarded prognosis. Diagnosis is usually made by ultrasound and the estimation of the hematoma size by ultrasound seems to be of prognostic value⁶⁴.

            Defective placentation

         During normal placentation the spiral arteries undergo adaptive changes characterized by loss of the normal musculoelastic arterial wall and replacement by fibrinoid material containing trophoblastic cells. Thence, the narrow, thick-walled arteries are transformed into wide-open, tortuous vascular channels that provide the necessary blood flow for the developing conceptus. Abnormal placentation is synonymous with the lack of these changes. Patients with recurrent abortions caused by abnormal placentation that are able to prolong a pregnancy beyond the second trimester remain at high risk for preeclampsia, preterm labor, and fetal growth retardation. Abnormal placentation occurs with similar frequency in patients with chromosomally normal and abnormal fetuses but is never found in patients with blighted ova.

            Immunological causes

         Alloimmune Etiology—The conceptus acquires genetic material from both its parents. Thus the conceptus inherits and express on the surface of its cells its father’s histocompatibility antigens that are foreign (i.e., allogeneic & alloantigenic) to the mother.

            Defence mechanisms

         The pre-implantation embryo expresses only low levels of paternal histocompatibility antigens which include products of major histocompatibility gene compel such as HLA and non-MHC or minor histocompatibility (H) transplantation antigens⁶⁵.

         This pre-implantation embryo is protected by its small size which reduces the probability of encounter with any maternal effecter cells that have strayed into tubal or peritoneal fluid.

         i.    by lack of direct contact with lymphatic-containing, maternal tissues

         ii.    by its zona pellucida which is shed just before implantation⁶⁶.

         However if cells such as macrophages in tubal or peritoneal fluid become activated-secretion of cytokines such as interleukin-1, tumor necrosis factor alpha and granulocyte macrophage colony stimulating factor (GM-CSF) can lead to embryo damage and... failure may occur e.g. in endometrosis^67-69.

         Radiographic contrast dye used for HSGs may paralyse activated macrophages and this might explain why some couples unexpectedly achieve pregnancy after this type of investigation.

         At the time of implantation expression of H antigens is shut off so that only embryonic antigens are expressed.

         During first 5 post implantation days, there is re-expression of minor paternal H antigens followed on the fifth post-implantation day by MHC expression^65,70. There is paternal MHC expression on the human embryo from 6-20 days post implantation⁷¹ and this is confirmed by immuno-histological study of first trimester embryos⁷².

         The conceptus is in fact 2 grafts in one. The fetal tissue (graft # 1) is enclosed within a sac of membranes lined by fetal trophoblast cells that form the feto-maternal interface and placenta. Conceptus behaves in a manner different from conventional allgrafts and this is primarily due to its trophoblast (graft # 2).

         Fetal trophoblast is a unique tissue the development of which is dependent upon presence of paternal genetic material. Where the genes of the trophoblast are entirely paternal, hydatidiform moles and choricarcinoma can develop⁷³. These neoplastic growths are not rejected by the female even when high levels of paternal HLA antigen are expressed by the tumor⁷⁴.

         Trophoblast may be divided into :

         1.   Non MHC expressing cells founds primarily at the interface between the fetal placental capillaries and maternal blood^75-76.

         2.   MHC expressing interstitial and chorionic membrane trophoblast in contact with deciduas expresses a modified from of class I MHC antigen that lacks individual HLA determinant.

            Defence mechanisms protecting the fetal

            trophoblast and blood supply

         An important feature of most trophoblast tissue is its extreme resistance to rejection by antigen specific immune effector mechanisms⁷⁶ Trophoblast cells are highly resistant to killing by antigen specific cytotoxic76cells and by antibody + cytolytic null type cells (antibody dependent cell-mediated killing).

         Null type cells include :

         i.    Natural killer cells (NK) that lyse transformed and virus-infected cells selectively and without antigen specificity.

         ii.    Natural cytotoxic (NC) cells that lyse cells of solid tumors resistant to NK cells.

         iii.   Macrophages

         iv.   Lymphokine activated Killer cells (LAKs).

         The spontaneous activity of NC and NK cells may be boosted by cytokines released by T cells that mediate delayed type hypersensitivity and generation of LAKS is dependent upon such factors.

·         Trophoblast, while resistant to killing by NK & NC cells, is quite sensitive to killing by LAKS^77-79.

·         NK, NC and activated macrophages produce a cytotoxin called tumor necrosis factor-alpha (TNF-a) that damages a trophoblast cell line (Clark et al, unpublished data).

         Stimulation of TNF-a release can also occur via a non T cell dependent mechanism through exposure of macrophages to bacterial products such as endotoxin⁸⁰. TNF-a acts on vascular endothelium to promote clotting (the thromobosis produces nutritional death of the conceptus).

            Role of decidua

            (a) Suppression

         Prior to implantation, a novel population of hormone induced suppressor cells develops^81,82, in the uterine lining. There is no antigenic specificity and these cells can not release any soluble suppressor factors unlike classic suppressor T cells⁸¹. The relevance of these cells to spontaneous abortion is three fold viz.

         i.    They persist during first 4 days after implantation.

         ii.    During this period of time, the decidua has the ability to exclude macrophages and prevent the expression of delayed type hypersensitivity. Whereas after formation of the placenta, these suppressor T-like cells disappear and expression of DTH becomes possible⁸³.

         iii.   Rate of loss has been found to be increased after injection of anti-body to CD8, presumably by interfering with the hormone induced suppressor T-like cell population⁸⁴.

         By the fifth post-implantation day, a population of small sized suppressor cells replaces the CD8 + suppressor population. These cells are small cells with cytoplasmic granules, lack conventional T cell markers but possess surface receptor for the Fc end of IgG⁸⁵ which are recruited/activated by soluble products from trophoblast cells. These cells release a potent immunosuppresive molecule that is homologous with transforming growth factor beta (TGF-b)⁸⁶ -which in turn inhibits NK cell activation, LAK generation and the response of NC cells to activation by interleukin 3 and inhibits the cytotoxic respiratory burst of monocyte macrophage cells^86,87. TGF-b has the potential to block all of the natural effector mechanisms capable of attacking the trophoblast. TGF-b and TNF-a antigonize each other with respect to effects on the immune response⁸⁸.

            (b) Stimulation

         Presence of growth factors can stimulate placental trophoblast growth. These are cytokines-viz interleukin 3 (a T cell derived molecule) and colony stimulating factors CSF-1⁸⁹ and CSF-GM⁹⁰ produced by T cells and a variety of stromal cells participating in the decidual response.

            Role of HLA sharing in recurrent abortion

         The major histocompatibility complex (MHC) of which HLA is the human variety consists of several linked loci that code for cell surface proteins important in immune function. Class I antigens (HLA loci A, B & C) serve as targets for recognition by cytotoxic T cells and as co-elements in recognition of viral antigens.

         Class II antigens (HLA D Loci) code for surface proteins restricted to a small number of cell-types involved in immune recognition and response. These immune response associated antigens (Ia antigens) associate with a variety of foreign molecules and allow their recognition by T helper/DTH cells.

         In allogeneic recognition reactions, of which the fetomaternal relationship may be one example, Ia differences cause the strongest proliferation of responding T cells in the host, Assumably, Ia is not expressed on the trophoblast due to this reason⁷⁶.

         Since HLA antigens represent strong antigens leading to graft rejection-it was rather surprising to find an increased frequency of HLA antigen sharing among couples with reduced pregnancy success (recurrent abortion)⁹³. It was proposed that the defect in abortion was lack of adequate stimulation of a helpful immune response in the mother due to a hypoantigenic conceptus⁹⁴. 25% of habitually aborting couples showed abnormal reactivity of the wife’s lymphocytes against the husband’s cells in vitro.

         It has been suggested that HLA-D sharing is a harbinger of sharing at a locus called TLX where TLX represents an antigen shared between lymphocytes and trophoblast. If husband and wife share the same TLX, the conceptus will not be antigenic and will not stimulate a protective immune response in the mother. It is proposed that deliberate immunization with husband or third-party lymphocytes will stimulate the protective blocking antibody response that is missing in aborting patients^95-100. This antibody inhibits an MLC reaction by recognizing the antigen as the receptor in the responding T cell^101,102; while it is called a blocking antibody.

         It seems likely that TLX sharing, if it does in fact occur, includes all of the important antigenic determinants (epitopes) on the molecule so that the father and fetus are still foreign but not sufficiently different to elicit an intrauterine immune response. Alternately, what may be shared is an antigen (TLX) which performs a helper function in ensuring an adequate induction of immunity to a minor H antigen on the embryo, possibly an embryonic antigen. Unless the husband is TLX homozygous and shares an antigen in common with his partner in which case a 100% failure rate would be predicted, even after 7-10 recurrent abortions there would still be a 50% or greater chance of success based on simple Mendelian genetics.

         Primary Aborters—No live births, abort only with one mate, have significant HLA sharing, and abort due to deficient immune response to antigens on the fetal trophoblast (possibly TLX). The response is humoral¹⁰³ and is proposed to stimulate local suppressor cell activity in decidua^104,105. These antibodies tend to block allorecognition by maternal T cells in vitro.

         Secondary Aborters —One or more live births, no HLA sharing, may abort with several partners and may possess antibody to paternal cells as detected by a variety of methods^106-108. Abortion is attributed to a toxic effect of the antibody. These patients may have second rather than first-trimester losses, possess anti-phospholipid antibodies that interfere with coagulation (so called lupus anticoagulant) and occasionally have clinically diagnosable autoimmune disease^109,110. It has been suggested that some first trimester aborters and primary aborters may also have such antibodies, and if blocking activity is already present, treatment with aspirin and prednisone (as is done for patients with lupus anticoagulant) is more effective than immunization¹⁰⁰. Antibody to blood group antigen p has also been linked to abortion of p+ fetuses¹¹¹.

         Revised classification may be :

         1.   Primary aborters without evidence of autoantibodies or blocking factors.

         2.   Secondary aborters without evidence of autoantibodies or blocking factors, but possibly possessing antibodies against antigens of the husband.

         3.   Primary and secondary aborters with evidence of autoantibodies or blocking factors^113,114.

         4.   Genetic aborters¹¹³.

         First two groups may be suitable for immunotherapy.

            Immunotherapy

         Success rates as high as 95% has been noted with immunotherapy when patients with pre-existing blocking antibodies¹⁰⁰ were excluded.

         Atternative forms of treatment for recurrent abortion

            Treatment                                                                                                                               Success Rate

         1.   Human Placental trophoblast¹¹⁵                                                                    75%

         2.   hCG¹¹⁶                                                                                                        94%

         3.   Corpus lutum extract¹¹⁷                                                                                84%

         4.   Progesterone¹¹⁸                                                                                           84%

         5.   Tender loving care¹¹⁹                                                                                    72%

         6.   Cerclage¹²⁰                                                                                                  85%

            Antiphospholipid antibodies

         Antiphospholipid antibodies comprise a family of autoandibodies which have a well-established association with fetal loss^121,122. They are most commonly associated with fetal loss, thrombocytopenia or thrombotic events without evidence of auto immune disease viz. the antiphospholipid syndrome.

         Primary Antiphospholipid Syndrome—must include one clinical and one serological feature.

            Clinical features

         1.   Recurrent venous or arterial thrombosis

         2.   Recurrent fetal loss

         3.   Thrombocytopenia

            Serological features

         1.   IgG aCL>20 GPL

         2.   LA

         3.   IgM aCL>20 MPL + LA

         I MPL or GPL is equivalent to 1 ug of affinity purified aCL IgM or IgG.

         Diagnostic Tests—The antiphospholipid antibodies are a diverse family of autoantibodies which share in common a reactivity with negatively charged phospholipids.

         There are three clinically significant members—viz. biological false positive test for syphilis.

·         Lupus anticoagulant

·         Anticardiolipin antibodies.

            I. Biological False Positive Test For Syphilis

         A biological false positive test can result from :

         1.   Antibodies produced in response to infection by a number of non treponemal pathogens¹²³. In this situation the BFP is likely to be transient, reflecting recent activity of the pathogen. These antibodies are not associated with thromobosis or fetal loss.

         2.   Autoantibodies produced by patients with autoimmune discuses, particularly patients with APS and/or SLE. These autoantibodies can persist for many years and are associated with both thrombosis and fetal loss.

            The BFP reaction requires

         1.   that an phospholipid based screening assay such as the VDRL is persistently positive for 6 months.

         2.   that the absence of treponemal infection is confirmed by a non-phospholipid-based assay such as the TPHA¹²⁴.

            II. Lupus Anticoagulant

         It comprises autoantibodies of either IgG or IgM class which prolong phospholipid-dependent coagulation assays by reacting with negatively charged phospholipids¹²⁵. Coagulation assays used to screen for LA include the activated partial thromboplastin time (APTT), Kaolin clotting time (KCT), dilute Russell Viper venom test (dRVVT) and dilule tissue thromboplastin assay (dTta)¹²⁶.

         The minimum criteria for the detection of LA proposed by Triplett & Brandt (1989) are :

         1.   A prolongation of a phospholipid-dependent screening test such as the APTT.

         2.   Demonstration that the abnormality is due to an inhibitor rather than a factor deficiency.

         3.   Proof that the inhibitor is directed against phospholipids¹²⁶.

         In order to confirm the presence of an inhibitor the screening test is repeated using a mixture (1 : 1) of the patient’s plasma with normal plasma. If the abnormality is due to an inhibitor the test will remain prolonged. If the abnormality is due to a factor deficiency the normal plasma will act as a source of the factor and the test result will correct to normal.

         Similarly a platelet neutralization procedure (PNP) is used to confirm the antiphospholipid nature of an inhibitor, Lysed platelets are added to the abnormal plasma, and the screening test (APTT or dRVVT) is repeated. An abnormality caused by LA will correct in a PNP whilst an abnormality due to a factor inhibitor will not.

            III. Anticardiolipin Antibodies

         It is detected by solid phase immune assay (enzyme-linked immuno sorbent assay (ELISA) or radio immuno assay (RIA)^127,128. The aCL assay has proved to be 200-400 times more sensitive than the VDRL test and detected 90% of LA in a population¹²⁷.

            IV. Antibodies to other phospholipids

         Phospholipid antigens other than cardiolipin have been used in solid-phase assays for aPL because cardiolipin being found exclusively in the mitochondria is unlikely to be the physiological antigen for aCL as it is not exposed to circulating aPL¹²⁹.

         Phosphatidyl serine has been used as an alternative antigen as it is located in membranes of endothelial cells and platelets. However, PS is found in the interior leaflet of non-activated cell membranes and is only exposed to circulating antibodies after cell activation causes its transfer to the exterior leaflet. Some use phosphatidyl ethanol amine which is in both the exterior and interior leaflets of cell membranes and thus is exposed to circulating aPL.

            Anticardiolipin antibody cofactor, B2 Glycoprotein

         The requirement of a cofactor to facilitate the binding of aCL to phospholipids has been described recently^130-132. The cofactor, which is present in normal serum, has been identified as B2-glycoprotein (B2-GP1 - also called apolipoprotein H)¹³¹.

         The invitro properties of B2 - GP1 are :

         1.   Inhibitor of intrinsic phase of coagulation.

         2.   Inhibitor of platelet activation.

         3.   Phospholipid binding protein^134-136.

         The formation of a trimolecular complex involving B2 - GP1, aCL and phospholipid prevents the normal function of B2 - GP. Also, Matsuura et al (1990)¹³² have demonstrated that, unlike aCL of autoimmune origin, aCL resulting from syphilitic or other infections bind to phospholipid in the absence of B2 - GP1. This difference in the cofactor requirements of aCL from various origins explains why only aCL of autoimmune origin are associated with thrombosis and recurrent fetal loss.

         Prevalence of aPL, both LA and aCL, in the general obstetric population is 2%.

         There is a definite correlation between the presence of aCL and LA. Among patients with SLE and LA, 59% will have aCL and in those with SLE and aCL, 45% will have LA. Incidence of aPL in women with three or more first trimester miscarriages varies between 14%¹³⁷ and 42%¹³⁸. Parke et al (1991)¹³⁹ found a positive frequency of aPL in women with recurrent miscarriage as 16%, women who had undergone normal pregnancy as 7% and women who had never been pregnant as 3%. Late fetal losses are also associated with aPL-incidence being 30-40%. Parazzini et al (1991)¹⁴⁰ concluded that there is no apparent justification for considering aPL to be a risk factor for fetal loss among women who present with spontaneous miscarriage or fetal death and have no previous spontaneous fetal loss. Lockshin (1987)¹⁴¹ noted that fetal death occurred in 77% women with SLE with aCL alone versus 50% patients with LA alone. He concluded that aCL was more sensitive and specific for predicting fetal death than LA.

(E)      PATHOGENESIS

            I. Effect on Platelets

         Early workers described thrombocytopenia as a frequent finding in patients with aPL¹⁴². Also thrombosis associated with APS is congruent with aPL binding to and disrupting the function of platelet membrane phospholipids. However, the negatively charged phospholipids, with which aPL react, are present only in the inner leaflet¹⁴³. Therefore, non activated platelets should not be antigenic for aPL. Upon activation the distribution of phospholipids in the platelet membrane is altered with the transfer of the negatively charged PS to the exterior leaflet of the membrane. This permits the platelet to participate in the processes of coagulation and may make them antigenic for aPL.

            II. Effect on the Vascular Endothelium

         (1) Inhibition of prostacyclin production—aPL which reacts with membrane phospholipid would inhibit the production of prostacyclin by the vascular endothelium and promote thrombosis¹⁴⁴. aPL inhibits the release of arachidonic acid from membrane phospholipid¹⁴⁴.

         (2)Inactivation of the protein C/protein S/Thrombomodulin pathway—Protein C acts as an anticoagulant by inhibiting the activated coagulation factors Va, VIIIa and platelet bound Va-Xa complex¹⁴⁵. Two steps in this pathway are phospholipid dependent and may be inhibited by aPL.

         a.   Protein C is activated by thrombo modulin - a protein present on the surface of vascular endothelial cells. Thrombomoduln must be bound to phospholipid to yield optimum protein C activation¹⁴⁶. It has been demonstrated that aPL can inhibit the invitro activation of protein C by thrombomodulin^146,147.

         b.   Once activated, protein C requires a co-factor, designated protein S, which facilitates the binding of activated protein C to the platelet membrane¹⁴⁸. Once bound to the platelet membrane, the protein C/protein S complex then inhibits coagulation factors Va and Xa. aPL can interact with phospholipids and inhibit the protein S dependent anticoagulant activity of activated protein C.

         (3) Inhibition of antithrombin III anticoagulant pathway—Endothelial cells express upon their surface heparin like molecules-glycosaminoglycans (GAGS)-which activate antithrombin III¹⁴⁹. The GAG heparan sulphate is believed to be the physiological activator of ATIII and is particularly important as an anticoagulant in the microcirculation¹⁵⁰. It has been proposed that aPL could cross-react with GAGS and inhibit the activation of ATIII.

            Clinical Associations with APS

         The hall mark of the presence of aPL is the triad of arterial and venous thrombocytopenia and fetal loss.

            Medical Disorders

         (I)      Neurological Disorders

                  1.   Transient ischaemic attacks

                  2.   Amaurosis fugax

                  3.   Migraine like headaches

                  4.   Acute ischaemic encephalopathy

                  5.   Multi-infarct dementia

                  6.   Degenerative myelopathy.

         (II)      Cardiac Disorders

                  1.   Valvular lesions

                  2.   Valvular or chamber thrombosis

                  3.   Coronary artery occlusion

         (III)     Adrenal gland involvement

                  1.   Addison’s disease

         (IV)    Skin

                  1.   Live do reticularis

         (V)     Lungs

                  1.   Pulmonary hypertension

         Snedden’s Syndrome comprises the clinical triad of live do reticularis, cerebrovascular occlusion and labile hypertension in the presence of aPL¹⁵¹. Evidence for an underlying collagen disorder can be found in 30-40% of individuals with aPL.

         Drugs such as procainamide, guanidine, phenytoin, chlorpromazine, valproic acid, amoxycillin, hydrallazine and propanolol have been reported to induce aPL but drug induced aPL is generally not associated with thrombotic complications or fetal loss.

            Obstetric disorders

         1.   First trimester miscarriage.

         2.   Later fetal loss with evidence of growth retardation.

         3.   Later fetal loss without evidence of growth retardation.

         4.   Placental abruption.

         5.   Pre-eclampsia-often severe and early onset.

         6.   Chorea gravidarum.

         Bird Sall et al (1992)¹⁵² found that 33% of women who had a shill birth due to placental abrupion had aPL.

            Indications for investigation of the presence of antiphospholipid

            antibodies in an obstetric population

         1.    All autoimmune diseases

         2.    Thrombocytopenia

         3.    Previous arterial or venous thrombotic event

         4.    BFP VDRL

         5.    Recurrent (>3) first trimester miscarriages

         6.    All fetal losses after 20 weeks of pregnancy

         7.    Placental abruption (previous or current pregnancy)

         8.    Fetal growth retardation (previous or current pregnancy)

         9.    Severe early onset pre-eclampsia (previous or current pregnancy)

         10. Chorea gravid arum.

            Management

            Specific therapy

         I.    Women with antiphospholipid antibodies without previous loss or significant medical disease.

               —Managed with careful monitoring alone

         II.    Antiphospholipid syndrome with fetal loss

               A.   Low level antibodies (aCL<60 GPL; KCT<250s)

                     Most patients of this group deliver a normally grown infant at term without intervention.

                     These women are treated with low dose aspirin—monitoring their LA level during pregnancy and observe closely.

               B.   High litre antibodies (aCL>60 GPL; KCT>250s).

                     These patients require active treatment with close monitoring of fetus and mother.

                     Treatment is with low dose aspirin + (i) Heparin 10,000 IU s/c BD in patients with previous thrombotic event.

         (ii)   Corticosteroids to reduce KCT<200 s in patients with no thrombotic event.

            Various therapies for APS

            (A) Single Agent Therapy

            1. Corticosteroids :

         Mechanism of action—Immune suppression by inhibiting the production of Interleukin 2 by T4 cells.

         Dosage—40-60 mg/day

         The steroid dose is titrated against the KCT value to maintain a KCT value below 200s, usually commencing with prednisone 40 mg and measuring KCT values monthly. For maintenance low dose prednisone 10-20 mg/day is used. Administration of corticosteroids will suppress the LA but not aCL.

         Side effects : Cushinoid features, acne, adrenal insufficiency, diabetes mellitus, oral candida, hypertension, osteoporosis.

         Overall live birth rate—44%.

            2. Heparin :

         Mechanism of action—Facilitates the action of ATIII.

         Dosage—Subcutaneous heparin sufficient to increase APTT to 1.5-2 times normal in those with normal APTT, or if APTT prolonged already heparin dose sufficient to achieve thrombin time of >100s.

         Usually a dosage of 10,000 IU twice daily is used in women with previous thromboembolis or those with aPL in high titre.

         Side effects : Bruising, thrombocytopenia, osteoporosis.

         Overall live birth rate—77%.

            3. Aspirin :

         Mechanism of action—Inhibits cyclooxygenase in the platelet which preferentially lowers platelet thromboxane, leaving endothelial prostacyclin synthesis relatively intact.

         Dosage—75-80 mg per day.

         Overall live birth rate—83%.

            (B) Combination therapy

         1.   Coricosteroid + low dose aspirin

               Over all live birth rate—68%

         2.   Heparin plus low-dose aspirin

               Overall live birth rate—88%

         3.   Azathioprine plus corticosteroid

               (75-100 mg)

         4.   Immunoglobulin prednisone plus low dose aspirin

               Overall live birth rate—71%.

            General Care

         Woman with APS is at significant risk of hypertension, venous thrombosis, pulmonary embolism or thrombocytopenia. Besides she may develop activation of her autoimmune disease.

            (C) Asymptomatic Antinuclear Antibodies

         Significantly greater prevalence of low-titer antinuclear antibodies (ANA) has been reported in patients with unexplained fetal losses before viability than in normal control subjects¹⁵³. High frequency of positive ANA titres also occurs in patients with fetal losses caused by nonimmunologic factors such as uterine anatomic malformations and luteal phase defect^154,155. ANA titre is between 1 : 20 and 1 : 160 and the fluorescent pattern is usually speckled or homogenous. Occasionally there will be a positive history of nasal or oral ulcerations, skin sensitivity, or unexplained recurrent musculo-skeletal pain. A positive ANA tite in a patient with prior early pregnancy losses indicates the need for further investigations of autoimmune factors viz. LAC; ACA and BFP-ST. If any of these tests is positive, autoimmunity is most probably responsible for pregnancy losses.

(F)      CLINICAL FEATURES

         The clinical types of abortions which obstetrician usually encounters regarding early pregnancy losses are

         1.   The patient with first trimester vaginal bleeding sub grouped into

               a.   Threatened abortion

               b.   Inevitable abortion :

                     Complete

                     Incomplete

         2.   The patient with fetal death or second trimester abortion—Missed Abortion

         3.   The patient that has a history of multiple early pregnancy losses—Recurrent Abortion.

            Threatened Abortion

         The most common symptom of a patient with impending abortion is vaginal bleeding. In most patients there is an interval of several days between the onset of symptoms and the actual miscarriage. However, in some the symptoms progress rapidly so these patients should be examined within a reasonable time.

         At the time the ovum is becoming embedded in the uterus, that is, at the time the first menstrual period is missed or soon after, a slight implantation hemorrhage is not uncommon. This diagnosis can only be made in retrospect and the safe rule is to regard any bleeding, no matter how little, which is not explained by a lesion on the vagina or cervix as evidence of threatened abortion. The bleeding is indicative of some degree of separation of the chorion from the decidua and it varies in amount, duration and type. At first the discharge is bright red. when it changes to dark brown it means that active bleeding has ceased and that old blood in the uterus is undergoing dissolution.

         Sometimes the patient also complains of backache and mild lower abdominal discomfort due to uterine contractions. However the pain is usually mild in these cases.

            Inevitable Abortion

         Abortion becomes inevitable if, in addition to the clinical features of threatened abortion, there are painful uterine contractions, dilatation of the cervix, or extrusion of some part of the conceptus through the os. Ballooning of upper vagina, tenderness of the uterus and pyrexia are other suggestive signs.

         Many abortions occur with a minimum of warning and upset, others are characterized by recurrent haemorrhage and this may produce serious shock. Retention of the products in the canal can itself produce shock.

         Incomplete Abortion—The bleeding does not get progressively less but varies from day to day, becoming heavy from time to time. It continues intermittently for weeks and months and may be accompanied by periodic uterine colic. A normal menstrual rhythm is not re-established.

         In long standing cases, the attachment of the chorion to the uterus becomes organized into fibrous tissue and becomes a placental polyp.

         On examination the uterus is slightly enlarged and often softened. Internal os is patulous—a very important sign indicating that there is something within the uterus.

            Complete Abortion

         The bloody discharge decreases progressively and usually ceases within 7-10 days. The first menstrual period occurs 4-6 weeks later.

            Clinical assessment of patients with

            first trimester vaginal bleeding

         Firstly the gestational age of the pregnancy should be estimated by clinical dating and by assessment of uterine size. A large for date uterus may indicate a Hydatidiform-mole whereas a small uterus suggests a blighted ovum. Presence of a tender adnexal mass suggests an ectopic pregnancy. Also, the pelvic examination may show cervical changes if the process is advanced.

         For pregnancy of less than 6 weeks GA, serum hCG is done to decide whether an ultrasound should be done and how to interpret its results. It is possible to see a gestational sac inside the uterus using TVS only when S.hCG is 1000 MIU/ml or more.

         Patients with first trimester bleeding and serum hCG levels below this critical value should have a repeated quantitative hCG evaluation 3 days later.

• If the hCG value doubles—pregnancy is likely to be intrauterine and there is a high probability of a normal outcome.
• If the hCG value does not double and initial progesterone and estradiol concentrations are low (progesterone < 15 ng/ml estradiol < 200 mg/ml), then the pregnancy is abnormal, either a blighted ovum or an ectopic.

         The predictive accuracy of low values for hCG, progesterone, and estradiol is 90-95% and most patients exhibiting this combination have spontaneous abortions or tubal pregnancies. The prediction of a normal outcome based on favorable hormone concentrations early in gestation is accurate in 80% of the cases. Also sonographic evidence of normal fetal cardiac activity in these patients indicates a low risk for abortion. Simpson et al¹⁵⁶ demonstrated that the fetal loss rate after documenting the presence of fetal heart motion by USG at 8 weeks is only 3.2%. This finding has been confirmed by other investigators¹⁵⁷.

         Patients with first trimester vaginal bleeding as well as reassuring ultrasound and hormonal findings should be told that the possibility of spontaneous abortion is small (2.5% to 3.2%). Bed rest + progesterone supplementation in cases with serum value < 15 ng/ml is recommended in these cases.

         Patients with blighted ova need karyotyping of the products of conception. If the karyotype of the blighted ovum reveals an autosomic trisomy, the patient will be at higher risk for a subsequent trisomy offspring and genetic amniocentesis should be recommended in subsequent pregnancies. If the karyotype shows structural rearrangement of the chromosomes, parents should go for karyotyping to rule out the possibility that one is a carrier of a translocation or an inversion.

         Cases with inevitable abortion require evaluation of the uterus under general anesthesia. However patients in shock should be resuscitated first.

            Missed Abortion

         Sometimes the fetus dies in utero but the uterus fails to respond normally by expelling it. The fetus then becomes macerated or mummified, the liquor amnii is absorbed and the placenta becomes pale and thin. Carneous (Blood) Mole is one variant caused by multiple hemorrhages in the choriodecidual space. A mass of partly organized blood clot and chorion is formed eventually after absorption of the dead fetus.

         Patient usually has complained of slight uterine bleeding following symptoms and signs of a normal pregnancy. The discharge clears up temporarily and this pregnancy is apparently progressing normally until it becomes clear after some weeks of observation that the uterus is not growing, indeed it becomes smaller and harder in consistency. Sooner or later hemorrhage recurs or there may be an intermittent brown discharge. Breast signs retrogress and symptoms such as nausea disappear. Diagnosis is usually made by ultrasound.

         Complications include disseminated intravascular coagulation caused by products of placental degeneration but is unlikely to occur for at least 3 weeks after the death of the pregnancy. Stimulation of the uterus can precipitate the trouble presumably by raising the intrauterine pressure to drive thromboplastin into the circulation. So blood tests for fibrinogen content and for clot stability before and during the induction of abortion are must. Normal level of fibrinogen is 350-450 mg%. The danger level is < 100 mg%.

         Treatment of afibrinogenaemia is only evacuation of uterus which causes correction in 12-14 hours.

         Histologic and microbiologic examination of the placenta is a fundamental part of the evaluation of these patients as the most common reasons for early fetal demise include chromosome abnormalities, antiphospholipid syndrome, ascending infection, subchorionic hematomas and abnormal placentation. Placenta will show extensive acute inflammatory changes in patients with ascending infection and typical lesions in patients with chronic villitis caused by cytomegalic virus infection. Thrombosis of fetal and maternal vessels will be seen in patients with protein C deficiency—possibility of fetal chromosome abnormalities can also be diagnosed.

         Fetal autopsy to rule out genetic syndrome, fetal blood for karytyping as well as for bactrial cultures for mycoplasma, ureaplasma, chlamydia and listeria may be done.

         ANA titer, anticardiolipin antibody, lupus anticoagulant and a TORCH titer may also help searching for the cause. HSG is recommended in unexplained cases a few weeks after their miscarriage to rule out a uterine anomaly. Induction with laminaria tents for cervical dilation followed by prostaglandin suppositories or high dose oxytocin are preferable over uterine evacuation and curettage—so as to get an intact fetus and placenta.

         Recurrent Abortions—Incidence is 1%. Three consecutive pregnancies ending in spontaneous abortion constitute the definition of recurrent abortion.

            Clinical Assessment and Diagnostic Work-up

         Majority of patients will have recurrent anembryonic abortions or blighted ova in this group. Following history and findings are suggestive of recurrent blighted ova :

         i.    Finding of molar degeneration in the histological examination of the products of conception of a prior abortion is suggestive of triploidy.

           ii.    Previous abortions showing empty sacs on ultrasound examination is suggestive of trisomy 16.

           iii.   History of a malformed infant in the family suggests balanced translocation in one of the parents.

           iv.   Interval infertility is again suggestive of balanced translocation in the parents.

           v.   History of repetitive abortions occurring before 12 weeks.

         (a) The probabily of a successful pregnancy in such cases is 62%¹⁵⁸ without any treatment.

         (b) Recurrent abortions with early fetal demise—Need evaluation of following factors.

               i.    Anatomic abnormalities of uterus and cervix best assessed by HSG (Hysterosalpingography) surgery, cervical cerclage can correct the defect.

               ii.    Corpus luteum deficiency :

                     Results in progesterone deficiency leading to pregnancy loss in first trimester especially between 8 and 12 weeks when the production of progesterone switches from the corpus luteum to the developing placenta. Typically, these patients have uterine contractions for several days preceding the onset of bleeding and the abortion. Diagnosis is made by endometrial biopsy or serum progesterone estimation during the secretory phase of the cycle.

               iii.   PCOD & Hyperandrogenism diagnosed by serum luteinizing hormone/follicle-stimulating hormone (LH/FSH) ratio and total testosterone concentrations. Other endocrinal problems like thyroid gland dysfunction should be treated. GnRH agonists for pituitary suppression followed by induction of ovulation with hCG is helpful in PCOD. For hyperandrogenism prednisone is used.

               iv.   Growth retarded but chromosomally normal fetuses. A biopsy of placental implantation area (obtained by curetting at the time of delivery) shows the lack of physiologic changes in the spiral arteries.

               v.   Autoimmune disorders diagnosed by ANA titer, lypus anticoagulant, VDRL, anticardiolipin and SS-A antibodies. Testing for HLA sharing or antipaternal antibodies has no role in establishing a prognosis. Low dose aspirin + prednisone 40-60 mg/day or low dose aspirin + subcutaneous heparin 5000 U 12 hrly is recommended.

               vi.   Repeated ascending infections due to group B streptococci, mycoplasma, ureaplasma and chlamydia can cause repetitive early pregnancy losses in patients who are carriers of the organism causing the infection and in patients with some degree of cervical in competence.

         Cerclage operation + antibiotic treatment is recommended in these cases.

(G)     CONCLUSION

         Differentiation of all abortion cases into embryonic and anembryonic pregnancies before abortion occurs with the advent of ultrasonography has simplified the counselling of patients with threatened abortions and of patients who had one or more spontaneous abortions. Couples with blighted ova do not require extensive workup, whereas patients who have aborted cytogenetically normal fetuses need as extensive search for nongenetic factors responsible for the pregnancy loss.

         Eventually it can be concluded that the obstetrician should investigate each pregnancy loss until the underlying cause is discovered. Otherwise, the diagnosis and management of patients with repetitive early pregnancy losses will be affected by inadequate information about the nature of their prior abortions.

(H)     REFERENCES

                 1.    Miller JF, Williamson E, Glue J, et al : Fetal lose after implantation : A prospective study. Lancet 1980; 2 : 554.

                 2.    Wilcox AJ, Weinberg CR, O’Connor JF, et al : Incidence of early loss of pregnancy. N Engl J Med 1988; 319 : 189-194.

                 3.    Regan L 1988 A prospective study of spontaneous abortion. In : Beard RW, Sharp F (eds) Early pregnancy failure, mechanims and treatment. Peacock Press, England, p 23-42.

                 4.    Grudzinskas JG, Stabile I, Campbell S 1988. Early pregnancy failure : biochemical and biophysical assessment. In : Beard RW, Sharp F (eds) Early pregnancy failure, mechanisms and treatment. Peacock Press, England, p 183-192.

                 5.    Simpson JL, Mills JL, Holmes LB et al 1987 Low fetal loss rates after ultrasound-proved viability in early pregnancy. Journal of the American Medical Association 258 : 2555-2557.

                 6.    Boue J, Boue A, Lazar P 1975. Retrospective and prospactive epidemiological studies of 1500 Karyotypes spontaneous human abortions. Teratoloty 12 : 11-26.

                 7.    Chambers, WH, Adamkiewicz, T and Houchins. JP (1993) Type-II integral membrane preteins with characteristics of C-type animal lectivs expressed by natural killer (NK) cells. Glycobiology 3 : 9-14.

                 8.    Bezouska, K., Yuen, CT, O’Brien, J, Childs, RA, Chai, W, Lawson, AM, Drbal, K, Fiserova, A, Pospisi, M and Feizi, T (1994). Oligosaccharide ligands for NKR-P1 protein activate NK cells and cytotoxicity. Nature 372 : 150-157l.

                 9.    King, A and Loke, YW (1988). Differential expression of blood group related carbobydrate antigens by trophoblast subpopulatins. Placenta 9 : 513-521.

              10.    Crane, JP and Cheung, SW (1998) An embryonic model to explain cytogenetic inconsistencies observed in chorionic villous versus fetal tissue. Prenatal Liganosis 8 : 119-129.

              11.    Ramsey, EM and Donner, MW (1980) Placental Vasculature and circulation. Georg Thieme, Stuttgart, pp. 12-13.

              12.    Pijenenbarg R (1994) Trophoblast invasion. Reproductive medicine Review 3 : 53-73.

              13.    Lawler, SD and Fisher, RA (1993) The contribution of the paternal genome : hydatidiform mole and choriocarscinoma. In : The Human Placenta, Eds. CWG, Sargent, IL and StarKey, PM, Blackwell Scientific, Oxford, pp. 82-112.

              14.    Moore, T and Haig, D (1991) Genomic imprinting is mammatian development : a parental tug-of-war. Trends in Genetics 7 : 45-49.

              15.    Haig, D (1993) Genetic conflicts in human pregnancy. Quarterly Review of Biology 68 : 495-532.

              16.    Jinno, Y, Yum, K, Nishikawa, K, Kubota, T Ogawa, O Reeve, AE and Niikawa, N (1994) Mosaic and polymorphic imprinting of the WTI gene in humans. Nature Genetics 6 : 305-309.

              17.    Goshen, R, Ben-Rafael, Z Gon. KB, Lustig, O, Tannos, V, de-Groot, N and Hochberg, AA (1994). The role of genomic imprinting in implantation. Fertility and Sterility 62 : 903-910.

              18.    GiannouKakis, N, Deal, C Paquette, J, Good year, CG and Polychromakos, C (1993) Parental genomic imprinting of the human 1GF2 gene. Nature Genetics 4 : 98-101.

              19.    Zhang, Y and Tycko, B (1992) Monoallelic expression of the human H19 gene. Nature Genetics 1 : 40-44.

              20.    Rachmilewitz, J, Gileadi, O Elder-Geva, T, Schneider, ., De Groot, N and Hochberg, A (1992) Transcription of the H19 gene in differentiating cytotrophoblast from human placenta. Molecular Reproduction and development 32 : 196-202.

              21.    Mutter, GL, Stewart CL, Chaponot ML and Pomponio RJ (1993) Oppositely imprinted genes H19 and insulin-like growth factor 2 are co-expressed in human androgenetic trophoblast. American Journal of Human Genetics 53 : 1096-1102.

              22.    Ariel I, Lustig O, Oyer CE, Elkin M, Gonik B, Rachmilewitz, J Biran, H Goshen R, De Groot N and Hochberg A (1994) Relaxation of imprinting in trophoblastic disease. Gynecology and Oncology 53 : 212-219.

              23.    Hashimoto K, Azuma C, Koyama M, Ohashi K, Kamiura S, Nobunaga T, Kimura T, Tokugawa Y, Kanai T and Safi F (1995). Loss of imprinting in choriocarcinoma. Nature Genetics 9 : 109-110.

              24.    Reed ML and Leff SE (1994) Maternal imprinting of human SNRPN, a gene deleted in Prader-Willi syndrome. Nature Genetics 6 : 163-167.

              25.    Harrison KB (1989) X-chromosome inactivation in the human cytotrophoblast, Cytogenetics and cell Genetics 52 : 37-41.

              26.    Lyon MF (1993) Epigenetic inheritance in mammals. Trends in Genetics 9 : 123-128.

              27.    Hellman A, Weislow OS, Twardzik DR and Fowler AK (1979) Type C retrovirus activation and possible functions in the normal and tumour-bearing host. cancer Research 39 : 2902-2907.

              28.    Coffin J. M. (1990) Retroviridae and replication. In : Virology, Eds. Fields B. N. and Knipe D. M., Raven Press, New York pp. 1437-1500.

              29.    Johnson P. M., Lyden T. W. and Mwenda J. M. (1990) Endogenous retroviral expression in the human placenta. American Journal of Reproductive Immunology 23 : 115-120.

              30.    Kato N, Pfeifer-Ohlsson S, Kato M, Larsson E, Rydnest J, Ohlsson R and Cohen M (1987) Tissue-specific expression of human provirus ERV3 mRNA in human placenta : two of the three ERV3 mRNA contain human cellular sequences. Journal of Virology 61 : 2182-2191.

              31.    Boyd, MT, Bax, CM, Bax, BE, Bloxam, DL and Weiss, RA (1993) The human endogenous retrovirus ERV-3 is upregualted in differentiating placenatal trophoblast cells virology 196 : 905-909.

              32.    Parem, S (1979) C-type virus expression in the placenta. Current Topics in Pathology 66 : 175-189.

              33.    Kalter, SS (1983) Viral expression in the trophoblast. In : Biology of Trobhoblast, Eds. Loke, YW and Whyte, A, Elsevier, Amsterdam, pp. 627-662.

              34.    Kalter, SS, Helmke, RJ, Heberling, RL, Panigel, M, Fowler, AK, Strickland, JE and Hellman, AJ (1973) C-type particles in normal human placentas. Journal of the National Cancer Institute 50 : 1081-1083.

              35.    Benveniste, RE, Lieber, MM, Livingston, D, Sherr, CJ and Todaro, G (1974) Infertious C-type virus isolated from a baboon placenta. Nature 248 : 17-20.

              36.    Dalton AJ, Hellman, A, Kalter, SS and Helmke, RJ (1974) Ultrastructural comparision of placental virus with several type-C oncogenic viruses. Journal of the National Cancer Institute 52 : 1379-1381.

              37.    Vernon, ML McMahon, JM and Hackett, JJ (1974) Additional evidence of C particlesx in human placentas. Journal of the National Cancer Institute 52 : 987-989.

              38.    Feldman, D (1975) An electron microscopic study of virus particles in rhesus monkey placenta. Proceedings of the National Academy of Sciences USA 72 : 118-121.

              39.    Kalter, SS, Heberling, RL, Smith, GC and Helmke, RJ (1975) C-type viruses in chimpanzee (Pan sp.) placentas. Journal of the National Cancer Institute 55 : 735-736.

              40.    Seman, G, Levy, BM, Panigel, M and Dmochowski, L (1975) Type-C virus particles in placenta of the cottontop marmoset (sanguinus oedipus). Journal of the National Cencer Institute 54 : 251-252.

              41.    Imamura, M, Phillips, PE and Mellors, RC (1976). The occurrence and frequency of type-C virus-like particles in placentas from patients with systemic lupus erythmatosus and from normal subjects. American Journal of Pathology 83 : 383-389.

              42.    Dirksen, ER and Levy, JA (1977) virus-like particles in placentas from normal individuals and patients with systemic lupus erythematosus. Journal of the National Cencer Institute 59 : 1187-1192.

              43.    Chandra, S, Liszczak, T, Korol, W and Jensen, EM (1970). Type-C particles in human tissues. I. Electron microsupic study of embryonic tissues in vivo and vitro. International Journal of Cancer 6 : 40-45.

              44.    Owens, R J, Tanner, CC, Mulligan, MJ, Srinivas, RV and Compans, RW (1990) oligopeptide inhibitors of HIV-induced syncytium formation. AIDS Research and Human Retroviruses 6 : 1289-1296.

              45.    Etiben B, Bartels I, Bahr-Porsch S, et al : Cytogenetic analysis of 750 spontaneous abortions with the direct preparation method of chorionic villi and its implications for studying genetic causes of pregnancy wstage. Am J Hum Genet 1990; 47 : 656-663.

              46.    Lohler J, Timpl R, Jaenisch R : Embryonic lethal mutation in mouse collagen alpha-1 (I) gene causes rupture of blood vessels and is associated with erythropoietic and mesochipnat cell death. Cell 1984; 38 : 597.

              47.    The PT, Byrd JR, McDonough PG : Etiology and subsequent reproductive performance of 100 couples with recurrent abortion. Fertil Steril 1979; 32 : 389.

              48.    Botella-Llusa J : The endometrium in repeatede abortion. Int J Fertil 1962; 7 : 147.

              49.    Balasch J, Creus M, Marquez M, et al : The significance of luteal phase deficiency on fertility : A diagnostic and therapeutic approach. Hum Reprod 1986; 1 : 145.

              50.    Hernandez Horta JL, Gordillo-Fernandez J, Soto de Leon B, et al : Direct evidence of luteal insufficiency in women with habitual abortion. Obstet Gynecol 1977; 49 : 705-708.

              51.    Filicori M, Butler JP, Crowley WF : Neuroendocrine regulation of the corpus luteum in the human. J Clin Invest 1984; 73 : 1638.

              52.    Yuen BH, Livinston JE, Poland BT, et al : Human chorionic gonadotropin, estradiol, progesterone, prolactin and B-Scan ultrasound monitoring of complications of early pregnancy. Obstet Gynecol 1981; 57 : 207-214.

              53.    Mills JL, Simpson JL, Driscoll SG, et al : Incidence of spontaneous abortion among normal women and insulin-dependent diabetic women whose pregnancies were identified within 21 days of conception. N Engl J Med 1988; 319 : 1617-1623.

              54.    Badarau L, Gavrilita L, Iuschevici C : A microscopic study of placenta and uteroplacental membrances in repeated abortion due to hyperandrogenism. Rev Fr Gynecol Obstel 1972; 67 : 685-697.

              55.    Sagle M, Bishop K, Ridley N, et al : Recurrent early miscarriage and polycystic ovary. Br Med J 1988; 297 : 1027-1028.

              56.    Regan L, Owen EJ, Jacobs HS : Hypersecretion of Luteinizing hormons, infertility, and miscarriage. Lancet 1990; 336 : 1141.

              57.    Johnson P, Pearce JM : Recurrent spontaneous abortion and polycystic ovary disease. Comparison of two regimens to induce ovulation, br Med J 1990; 300 : 154-156.

              58.    Sompolinsky D, Soomon F, Elkinal L, et al : Infections with mycoplasma and bacteria in induced midtrimester abortion and fetal loss. Am J Obstet Gynaecol 1975; 121 : 610-616.

              59.    Stray-Pedersen B, Eng J, Reikvan TM : Uterine T-mycoplasma colonization in reproductive failure. Am J Obstet Gynaecol 1978; 130 : 307.

              60.    Qinn PA, Chewchuck AB, Suber J, et al : Efficacy of antibiotic therapy in preventing spontaneous pregnancy loss among couples colonized with genital mycopleasma. Am J Obstet Gynaecol 1983; 145 : 239-244.

              61.    Schenker JG, Magalioth EJ : Intrauterine adhesions : An updated appraisal. Fertil steril 1982; 37 : 593-610.

              62.    Golan A, Langer RM, Wexler B, et al : Cervical cerclage : Its role in the pregnant anomalous uterus. Int J Fertil 1990; 35 : 164.

              63.    Kaufman RH, Noller K, Adam E, et al : Upper genital tact abnormalities and pregnancy outcome in diethylstilbestrol exposed progeny. Am J Obstet Gynaecol 1984; 148 : 973.

              64.    Pedersen JF, Mantoni M :Large intrauterine haematomata in threatened miscarriage : Frequency and clinical consequences. Br J Obstet Gynaecol 1990; 97 : 75.

              65.    Billington WD, Bell SC 1983. Immunology of mouse trophoblast. In : Loke YW, Whyte A (eds) Biology of trophoblast. Elsevier, Amsterdam, p 571-595.

              66.    Warner CM, Brownell MS, Ewoldsen MA 1988. Why aren’t embryos immunologically rejected by their mothers ? Biology of Reproduction 38 : 17-29.

              67.    Morcos RN, Gibbons WE, Findley WE 1985 Effect of peritoneal fluid on in vitro clearage of 2-cell mouse embryos : possible role in infertility associated with endometriosis. Fertility and Sterility 44 : 678-683.

              68.    Halme J, Becker S, Haskill S 1987. Altered maturation and function of peritoneal macrophages : possible role in pathogenesis of endometriosis. American Journal of Obstetrics and Gynaecology 156 : 783-788.

              69.    Hahn DW, Carraher RP, Folesy RG, McGuire JK 1986. Experimental evidence for failure to implant as a mechanism of infertility associated with endometriosis. American Joural of Obstetrics and Gynaecology 155 : 1109-1113.

              70.    Rhagupathy R, Singh B, Barrington-Leigh J, Wegmann TG 1981. Ontogeny and turnover Kinetics of paternal H-2K antigenic determinants on the allogeneic murine placenta. Journal of Immunology127 : 2074-2079.

              71.    Rugh R 1968 The mouse. Its reproduction and development. Burgess Publishing Co, Minneapolis, Minnesota, USA.

              72.    Butterworth BH, Khong TY, Loke YW, Robertson WB 1985. Human cytotrophoblast populations studied by monuclonal antibodies using single and double biotinavidin peroxidase immunocy to chemistry 33 : 977-983.

              73.    Szulman AE 1988. The biology of trophoblast disease : Complete and partial hydatidiform males. In : Beard RW, Sharp F (eds) Early pregnancy failure, mechanism and treatment. Peacock Press, England, p 309-316.

              74.    Sasagawa IM, Ohmomo V, Kanazawa K, Takeuchi S 1987. HLA expression by trophoblast of invasive moles. Placenta 8 : 111-118.

              75.    Bulmer JN, Jolmson PM 1985. Antigen expression by trophoblast populations in the human placenta and their possible immunological relevance. Placenta 6 : 127-140.

              76.    Head JR, Drake BL, Zuckermann FA 1987. Major histocompatibility antigens on trophoblast and their regulation : implications in the maternal-fetal relationship. American Journal of Reproductive Immunology and Microbiology 15 : 12-18.

              77.    Drake BL, Head JR 1988. Murine trophoblast cells are susceptible to lymphokine-activated Killer (LAK) cell lysis. American Journal of Reproductive Immunology and Microbiology 16 : 114.

              78.    Clark DA, Croy BA, Rossant J, Chaouat G 1986a Immune presensitization and local intrauterine defence as determinants of success or failure of murine interspecies pregnancy. Journal of Reproduction and Fertility 77 : 633-643.

              79.    Clark DA, Chaput A, Tutton D 1986b. Active suppression of host-versus-graft reaction in pregnant mice. VII. spontaneous abortion of allogeneic DBA/S X CBA/J fetu;ses in the uterus of CBA/J mice correlates with deficient non-T Suppressor cell activity. Journal of Immunology 136 : 1668-1675.

              80.    Kornbluth RS, Edginton TS 1986. Tumor necrosis faitor production by human monocytes is a regulated event : induction of TNF-alpha-mediated cellular cytotoxicity by endotoxin. Journal of Immunology 137 : 2585-2591.

              81.    Brieley J, Clark DA 1987. Characterization of hormone dependent suppressor cells in the uterus in pregnant and pseudopregnant mice. Journal of Reproductive Immunology 10 : 201-218.

              82.    Daya S, Clark DA, Devlin C, Jarrell J, Chaput A 1985. Preliminary characterization of two types of suppressor cells in the human uterus. Fertility and sterility 44 : 778-785.

              83.    Redline RW, Shea CM, Pappioannou VE, Lu CY 1998. Defective anti-listerial responses in deciduoma of pseudopregnant mice. American Journal of Pathology 133 : 485-497.

              84.    Clark DA, Banuratt DK, Manuel J, Fulop G, Croy BA 1989b SCID mice in reproductive biology. In : Phillips RA (ed) contemporary topics in immunobiology. Springer, Heidelber (inpress).

              85.    Slapsys RM, Younglai E, Clark DA 1988. A novel suppressor cell is recruited to decidua by fetal trophoblast-type cells. Regional Immunology I : 182-189.

              86.    Clark DA, Falbo M, Rowley RB, Banwatt D, Stedronska-clark J 1988. Active suppression of host-versus-graft reaction in pregnant mice. IX. Soluble suppressor activity obtained from allopregnant mouse decidua that blocks the response to interleukin 2 is related to transforming growth factor beta. Journal of Immunology 141 : 3833-3840.

              87.    Tsmunawaki S, Sporn M, Ding A, Nathan C 1988. Deactivation of macrophages by transforming growth factor-beta. Nature 34 : 260-262.

              88.    Ranges GE, Figari IS, Espevik T, Palladino MA Jr 1981. Inhibition of cytotoxic T cell development by transforming growth factor beta and reversal by recombinant tumor necrosis factor alpha. Journal of Experimental Medicine 166 : 991-998.

              89.    Bartocii A, Pollard JW, Stanley ER 1986. Regulation of colony stimulating factor 1 during pregnancy. Journal of Experimental Medicine 164 : 956-961.

              90.    Athnasakis I, Bleackley RC, Paetkau V, Guilbert L, Barr PJ, Wegmann TG 1987. The immunostimulatory effect of T cells and T cell lymphokines on murine fetally derived placental cells. Journal of Immunology138 : 37-44.

              91.    Thornes BJ, Mermod JJ, Vassalli P 1987. Phagocytosis and inflammatory stimuli induce GM-CSF MRNA in macrophages through post-transcriptional regulation. Cell 48 : 671.

              92.    Zucali JR, Dinarello CA, Oblon DJ, Gross MA, Andersson L, Weiner RW 1986. Interleukin 1 stimulates fibrobcasts to produce granulocyte-macrophage colony-stimulating activity and prostaglandin E2. Journal of Clinical Investigation 77 : 1857.

              93.    Thomas ML, Harger JH, Wagener DK, Robin BS, Gill TJ 1985. HLA sharing and spontaneous abortion in humens. American Journal of Obstetrics and Gynaecology 8 : 1053-1058.

              94.    Taylor C, Faulk WP 1981. Prevention of recurrent abortion with Leukocyte transfusions. Lancet ii : 68-70.

              95.    Rokline RE, Kitzmiller JL, Carpenter CB, Garvoy MR, David JR 1976. Maternal fetal relation. Absence of an immunological blocking factor from the sera of women with chronic abortions. New England Journal of Medicine 295 : 1209-1213.

              96.    Fizet D, Bousqauet J 1983. Absence of a factor blocking cellular cytotoxicity reaction in serum of women with recurrent abortion. British Journal of Obstetrics and Gynaecology 90 : 453-456.

              97.    P ower DA, Mather DA, MacLeod AM, Lind T, Catto GRD 1986 maternal antibodies to paternal B-lymphocytes in normal and abnormal pregnancy. Americal Journal of Reproductive Immunology and Microbiology 10 : 10-13.

              98.    Takukawa K, Kamazawa K, Takeuchi S 1986. Production of blocking antibodies by vaccination with husband’s lymphocytes in unexplained recurrent aborters : the role in successful pregnancy. American Journal of Reproductive Immunology and Microbiology 10: 1-9.

              99.    Beer AE 1988. Pregnancy outcome in couples with recurrent abortions following immunological evaluation and therapy. In : Beard RW, Sharp F (eds) Early Pregnancy loss, mechanisms and treatment. Peacock Press, England, p 227-354.

            100.    Unander M 1988 Discussion. In : Beard RW, Sharp F (eds) Early pregnancy failure, mechanisms and treatment. Peacock Press, England, p 400-403.

            101.    Singal DP, Butter L, Liao SK, Joseph S 1984. The fetus as an allograft evidence for anti-ideotypic antibodies induced by pregnancy. American Journal of Reproductive Immunology 6 : 145-151.

            102.    Bonaqura VR, Ma A, McDowell J, Lewison A, King DW, Suciu-Foca N 1987. Anti-clonotypic antoantibodies in pregnancy. Cellular Immunology 108 : 356-365.

            103.    Davies M, Browne CM 1985. Anti-trophoblast antibody responses during normal human pregnancy. Journal of Reproductive Immunology. 7 : 285-297.

            104.    Clark DA, Kiger N, Guenet J-L, Chaouat G 1987. Local active suppression and successxful vaccination against spontaneous abortion in CBA/J mice. Journal of Reproductive Immunology 10 : 79-85.

            105.    Chaouat G, Kolb JP, Kiger N, Stanis Lawski M, Wegmann TG 1985. Immunologic consequences of vaccination against abortion in mice. Journal of Immunology 134 : 1594-1598.

            106.    McIntyre JA, McComnnachie PR, Taylor C, Faulk WE 1984. Clinical, immunology and genetic definitions of primary and recondacy recurrent spontaneous abortion. Fertility and sterility 42 : 844-855.

            107.    McIntyre JA, Faulk WP, Nichols-Jolmson VR, Taylor CG 1985. Immunological testing and immunotherapy in recurrent spontaneous abortion. Obstetrics and Gynaecology 67 : 169-175.

            108.    McConnachie PR, McIntyre JA 1984. Maternal antipaterual immunity in couples predisposed to repeat pregnancy losses. Americal Journal of Reproductive Immunology 5 : 145-150.

            109.    Harris EW 1988. Clinical and immunological significance of anti-phospholipid antibodies. In : Beard RW, Sharp F (eds) Early pregnancy failure, mechanisms and treatment. Peacock Press, England, p 43-65.

            110.    Howard MA, Firkin BG, Healy DL, Choong S-CC 1987. Lupus anticoagulant in women with multiple spontaneous miscarriage. American Journal of Hematology 26 : 175-178.

            111.    Lepez M, Cartron J, Cartron JP et al 1983. Cytotoxicity of anti PP, PK antibodies and possible relationship with early abortions of P mothers. Clinical Immunology and Immunopathology 28 : 296-303.

            112.    Gleicher N 1987. A potential animal model for autoimmune reproductive failure. Americal Journal of Reproductive Immunology and Microbiology 14 : 122.

            113.    Mowbray JF 1988. Success and failures of immunisation for recurrent spontaneous abortion.In : Beard RW, Sharp F (eds) Early pregnancy failure, mechanisms and treatment. Peacock Press, England, p 325-336.     

            114.    Gill TJ 1987. Genetic factors in fetal loss. American Journal of Reproductive Immunology and Microbiology 15 : 133-137.

            115.    Johnson PM, Chia KV, Hart CA, Griffith HB, Frances W JA 1988. Trophoblast membrane infusion for unexplained recurrent miscarriage. British Journal of Obstetrics and Gynaecology 95 : 111-16.

            116.    Harrison RF 1988. Early recurrent pregnancy failure : treatment with human chorionic gonadotrophin. In : Beard RW, Sharp F (eds) Early pregnancy failure, mechanisms and treatment. Peacock Press, England, p 421-431.

            117.    Potter MF 1938. Foetal mortality in the offsring of apparently healthy women. Journal of Obstetrics and Gynaecology of the British Empire 45 : 233-249.

            118.    Shearman RP, Garrett WJ 1963. Double blind study of the effect of 17-hydraoxy-progesterone caproate on the abortion rate. British Medical Journal i : 292-295.

            119.    Stray-Pedersen B, Stray-Pedersen S 1988. Recurrent abortion : the role of psychotherapy. In : Beard RW, Sharp F (eds) Early pregnancy failure, mechanisms and treatment. Peacock Press, England, p 433-442.

            120.    Edmonds DK 1988. Use of cervical cerclage in patients with recurrent first trimester abortion. In : Beard RW, Sharp F (eds) Early pregnancy failure, mechanisms and treatment. Peacock Press, England, p 411-419.

            121.    Harris EN Asherson RA, Hughes GRV 1998 Antiphospholipid antibodies : autoantibodies with a difference. Annu Rev Med 39 : 261-271.

            122.    Branch WD, Scott JR, Kochenour NK et al 1985 Obstetric complications associated with the lupus anticoagulant N. Engl J Med 313 : 1322-1326.

            123.    Masala C, Sarice M Di prima MA et al 1989 Evidence for shared epitopes between cardiolipin and Pneumocystis carinii. Infect Dis 160 : 736-737.

            124.    Schulman LE 1987 Systemic Lupus erythematosus and the chronic biologic false positive test of syphilis. In : Wallace DJ, Dubois EL (eds) Dubois’ Lupus erythematosus, 3rd Edn. Lea & Febiger, Philadelphia.

            125.    Pengo V, Thiagarajan P Shapiro SS, Heine MJ 1987 Immunological specificity and mechanism of action of IgG lupus anticoagulants. Blood 70 : 69-76.

            126.    Triplett DA, Brandt JT 1989. Laboratory identification of the lupus anticoagulant, Br J Haematol 73 : 139-142.

            127.    Harris EN, Gharavi AE, Boey ML et al 1983 Anticardiolipin antibodies : detection by radioimmuno assay and association with thrombosis in systemic lupus erythematosus. Lancet ii : 1211-1214.

            128.    Pattison NS, McKay EJ Liggins GC, Lubbe WE 1987 Anticardiolipin antibodies : their presence as a marker for lupus anticoagulant in pregnancy. NZ Med J 100 : 61-64.

            129.    Hecht ER 1965 Lipids in blood clotting. Charles C Thomas, Springfield, IL.

            130.    Galli M, Comfurius P, Maassen C et al 1990a Anti cardio lipin antibodies (aCL) directed not to cardio lipin but to a plasma protein cofactor. Lancet 335 : 1544-1547.

            131.    McNeil HP, Simpson RJ, Chesterman CN, Krilis SA 1990 Antiphospholipid antibodies are directed against a complex antigen that includes a lipid-binding inhibition of coagulation : B2 glycoprotein (apolipoprotein H). Proc Natl Acad Sci USA 87 : 4120-4124.

            132.    Matsuura E, Igarashi Y, Fujimoto M et al 1990 Anticardiolipin and differential diagnosis of autoimmune disease. Lancet 336 : 177-178.

            133.    Chamely LW, McKay EJ, Pattison NS 1991a. Cofactor dependent and cofactor independent anticardiolipin antibodies. Thromb Res 61 : 291-299.

            134.    Wurm H 1984, B2 glycoprotein 1 (apolipoprotein H) interactions with phospholipid vesicles-Int J Biochem 16 : 511-515.

            135.    SK Housboe I 1985 B2 glycoprotein 1 : a plasma inhibitor of the contact activation of the intrinsic blood coagulation pathway. Blood 66 : 1086-1091.

            136.    Nimpf J, Wurm H, Kostner GM 1987 B2 glycoprotein 1 (apo-H) inhibits the release reaction of human platelets during ADP-induced aggregation. Atherosclerosis 63 : 109-114.

            137.    Barbui T, Cortelazzo S, Galli M et al 1998 Antiphospholipid antibodies in early repeated abortions : a case controlled study. Fertil steril 50 : 589-597.

            138.    Unander AM, Norbert R, Hahn L, Arfors L 1987 Anticardiolipin antibodies and complement in niety-ninety women with habitual abortion. Am J. Obstet Gynecol. 156 : 114-119.

            139.    Parke AL Wilson D, Bayes-Maierd 1991. The prevalence of antiphospholipid antibodies in women with recurrent spontaneous abortion, women with successful pregnancies and women who have never been pregnant. Arthritis Rheum 34 : 1231-1235.

            140.    Parazzini F, Ackia D, Faden D et al 1991. Antiphospholipid antibodies and reccurent abortion. Obstet Gynecol. 77 : 854.

            141.    Lockshin MD, Quamar T, Druzin M, Goei S 1987. Antibody to cardiolipin, cupus anticoagulant and fetal death. J Rheumatol 14 : 259-262.

            142.    Harris EN, Gharavi AE, Hedge U et al 1985. Anticardiolipin antibodies in antoimmune thrombocytopenic purpura. Br J Haematol 59 : 231-234.

            143.    Gerrard JM, Friesen LL, 1985 Platelets. In Poller L(ed) Recent advances in blood coagulation vol 4. Churchill Livingstone, Edinburgh, pp 139-168.

            144.    Carreras LO, Machin SJ, Deman R. et al 1981 Arterial thrombosis, intrauterine death and lupus anticoagulant : detection of immunoglobulin interfering with prostacyclin formation. Lancet i : 224-226.

            145.    Kwaan HC 1989 Protein C and protein S Semin Thromb Hemost 15 : 353-355.

            146.    Freyssinet J-M, Wiesel ML, Gauchy J et at 1986 An IgM Lupus anticoagulant that neutraulises the enhancing effect of phospholipid on purified endothelial thrombomodulin activity : A mechanism for thrombosis. Thromb Haemost 55 : 309-313.

            147.    Tsakiris DA, Settas L, Makris PE, Marbet GA 1990 Lupus anticoagulant, antiphospholipid antibodies and thrombophilia : relation to protein c-protein S-thrombomodulin J. Rheumatol 17 : 785-789.

            148.    Comp PC 1990 Laboratory evaluation of protein S status. Semin Thromb Hemos 16 : 177-181.

            149.    Andersson LO (1979) Physiochemical properties of antithrombin III, In : Collen D, Wiman B, Verstraete M. (eds) The physiological inhibitors of coagulation and fibrinolysis. Elsevier/North Holland Biomedical Press, Amsterdam, pp 39-42.

            150.    Marcum JA, Reilly CF, Rosenberg RD 1986. The role of specific forms of heparan sulphate in regulating blood vessel wall function. Prog. Hemost Throm 8 : 185-215.

            151.    Schulze-Lohoft E, Kraft F, Bleil L et al 1989 IgM-containing immune complexes and antiphospholipid antibodies in patients with Snedden’s syndrome. Rheum Int 9 : 43-48.

            152.    Birdsall MA, Pattison NS, Chamley LW 1992 Antiphospholipid antibodies in pregnancy. Aust N 2 J obstel Gynaecol 32 : 328-330.

            153.    Garcia-de la Torre, Hernandez-Vasquez L, Angulo-Vasquez J, et al : Prevalence of antinuclear antibodies in patients with habitual abortion and in normal and toxemic pregnancies. Rheumatol Int 1984; 4 : 87-89.

            154.    Xu L, Chang V, Murphy A, et al : Antinuclears antibodies in sera of patients with recurrent pregnancy wastage. Am J Obstet Gynaecol 1990; 163 : 1493-1497.

            155.    Cowchock S, Smith B, Gociat B : Antibodies to phospholipids and nuclear antigens in patients with repeated abortions. Am J Obstet Gynaecol 1986; 155 : 1002-1010.

            156.    Simpson JL, Mitts JL, Holmes LB, et al : Low fetal loss rates after ultrasound-proved viability in early pregnancy. JAMA 1987;258 : 2555-2557.

            157.    Siddiqui TA, Caligaris JT. Miodovnik M, et al : Rate of spontaneous abortion after first trimester somographic demonstration of fetal cardiac activity. Am J Perinatol 1988; 5 : 1-4.

            158.    Howert-de Jong MH, Termijtelen A, Eskes TKAB, et al : The natural course of habitual abortion. Eur J Obstet Gynaecol 1989; 33 : 221-228.

(A)      INTRODUCTION

(B)      DEFINITIONS

(C)     PHYSIOLOGY OF

            IMPLANTATION—

            IMMUNOLOGICAL ASPECTS

(D)     ETIOLOGY

(E)      PATHOGENESIS

(F)      CLINICAL FEATURES

(G)     CONCLUSION

(H)     REFERENCES
(A)      INTRODUCTION

         A spontaneous abortion is defined as any recognized, involuntary pregnancy loss occurring before fetal viability, 24 weeks. Approximately 80% of all spontaneous abortions occur before 12 weeks and are called early abortions. The rest occur between the thirteenth and the twenty-fourth week and are called late abortions. Majority of early abortions correspond to anembryonic pregnancies or blighted ova, whereas most of the abortions with a fetus present occur in the second trimester.

         It has been seen that only 57% of all conceptions advance beyond 20 weeks. Of those lost, 75% occur before implantation and only 25% are clinically recognizable¹. Recent investigations have found an overall pregnancy loss of 31% with 22% occurring before implantation².

(B)      DEFINITIONS

            Blighted ova—Early pregnancy losses in which fetal development is not observed with ultrasound and fetal tissue is absent on the histologic examination of the products of conception.

            Early fetal demise—Early pregnancy losses in which fetal development is clearly observed by ultrasound and fetal tissue is found on the histologic examination.

            Complete abortion—Cases in which ovum along with chorion/placenta is completely expelled.

            Incomplete abortion—Cases in which placenta or chorion is wholly or partly retained within the uterus.

            Primary aborters—Those without any previous live births. Primary miscarriage tends to be partner specific.

         Secondary aborters—Those who have had one or more children.

         Approximately 25% of pregnant women will threaten miscarriage but a smaller percentage actually complete it. Most spontaneous abortions occur in the first trimester, with the median between 8 and 9 weeks gestation3,4. If a fetal heart is detected in the first trimester (5-14 weeks), 95-98% gestate successfully^3,4,5. Most of those who abort show either absent fetal heart or an empty gestational sac. The frequency of chromosomal anomalies in material from recurrent aborters is approximately half that reported for sporadic aborters⁶.

(C)     PHYSIOLOGY OF IMPLANTATION—IMMUNOLOGICAL ASPECTS

         Many Pathological conditions of pregnancy encountered in clinical obstetrical practice e.g. miscarriage. IUGR & pre-eclampsia, are all due to failure of the normal controlling mechanisms on trophoblast migration, leading to insufficient invasion and vascular adaptation with the final result of poor perfusion of the feto-placental unit.

            Allorecognition

         NK cells tend to be cylolytic to largest cells which are deficient in MHC class I molecules. Thus, T cells recognise ‘non-self’, NK cells recognise ‘missing self’ of the absence of self. Neoplastic transformation or viral infection can lead to downregulation of class I molecules and it is possible that NK cells will provide the defence in these situations.

         Therefore, the phylogenetically more primitive NK cells have been designed for internal surveillance and to monitor the integrity of self while T cells have been developed to guard against non-self external aggressors.

            Immunology of the maternal fetal interaction

         The defensive cells present at the feto-maternal interface in the human uterus are macrophages and NK cells which are cells of innate system. B cells are absent and T cells sparce. Unlike the artificial situation of an organ transplant, the intermingling of cells from genetically different individuals occurs naturally during placentation. Also unlike a transplant, the mammalian conceptus is not rejected.

         Despite the original definition of NK recognition as ‘Non MHC restricted’ it's now generally accepted that NK cells are capable of allorecognition and that human NK cells can broadly discriminate between some MHC class I antigens. Human extravillous trophoblast expresses class I HLA-G and may be HLA-C also, but not the usual class I HLA-A, B like other somatic cells. Perhaps these class I antigens provide the recognition molecules by which decidual NK cells detect the allogeneic nature of trophoblast. A recent discovery is that NK cells express type II membrane glycoproteins which contain a C-type lectin-like domain.⁷

         Some of these NKreceptors have been found to recognise oligosaccharide ligands which deliver stimulatory or inhibitory signals to the NK cell⁸. A search for trophoblast-specific carbohydrate antigens might prove very fruitful. There is already evidence that sialylated type I blood gp related carbohydrates are preferentially expressed on certain human extravillous trophoblast sub-populations⁹. Thus, the concept of ‘self’ and ‘non-self’ in terms of NK cells may be very different to how it is generally perceived in classical T cell immunology. Decidual NK cells also produce colony stimulating factors, among other cytokines, so the possibility arises that during evolution trophoblast has managed to utilise certain aspects of the maternal cytokine response to sustain its own development. These cytokines will provide either positive or negative signals to ensure the correct degree of trophoblast migration within the uterine environment.

         Thus, although the human placenta is genetically foreign to the mother, the characteristics of its interaction with the uterus are more akin to those observed between unrelated primitive invertebrates than between mammalian allograft & host as seen in contemporary transplantation immunology.

            Development of trophoblast

         The formation of the human placenta begins at the time the blastocyst penetrates the maternal endometrium to become rapidly embedded in the uterine stroma.

            Early trophoblast development

         The first element to differentiate in the morula is the trophoectoderm. As the morula is converted into a blastocyst, trophoectoderm cells form a layer encircling blastocyst with the inner cell mass at one pole. The blastocyst inserts itself between the epithelial cells of the uterine mucosa on the 6th-7th day after ovulation.

         By the 8th day, trophoblast has differented into an outer multinucleated syncytiotrophoblast (primitive syncytium) and an inner layer of primitive mononuclear cytotrophoblast.

         By the 9th day, vacuoles or lacunae appear in the syncytium and these rapidly enlarge and fuse to communicate with each other. The establishment of a potential uteroplacental circulation occurs when maternal venous capiliaries are eroded by the syncytium so that blood can seep into the lacuna system. The lacunae will eventually become the intervillous space of the definitive placenta.

         By 12 to 13 days after fertilization, the blastocyst is completely embedded in the decidual stroma and the uterine surface epithelium has grown over the overlying defect.

         After this initial phase of nidation, trophoblast differentiation may be considered to occur along two main pathways-villous & extravilllous. Villous trophoblast ultimately covers all the chorionic villi of the definitive placenta and is concerned with transport of oxygen and nutrients from mother to child. Extravillous trophoblast migrates deep into the uterine mucosa as far as the myometrium.

            Villous trophoblast

         The earliest sign of villous development is the appearance of buds of cytotrophoblast protruding into the primitive syncytium in the second week of gestation. Penetration of solid cytotrophoblast buds by mesenchymal cells convert them into secondary villi. Tertiary villi are formed when the mesenchymal core is supplied by embryonic blood vessels.

         The embryonic vessels link with the vessels in the mesoderm of the chorionic plate, and ultimately with vessels in the fetus so that by the fourth week the feto-placental circulation is established. The mesodermal core of the chorionic villi is derived from cells of the inner cell mass which originate from the caudal margin of the embryonic disc.¹⁰

         During the second and third months of pregnancy, rapid extension and branching of the villi result in the villous tree characteristic of the mature placenta. In the first 3 months there is an obvious two-layered trophoblast covering of the villi. The inner cytotrophoblast cells sit on a basement membrane covered by the outer syncytiotrophast layer. Syncytiotrophoblast lines the entire intervillous space and thus comes into contact with maternal blood in a manner similar to endothelial cells. In the first trimester, the core of the villus contains loose mesenchymal cells, embryonic blood vessels and fetal macrophages (Hofbauer cells) Hofbauer cells appear very early in gestation (by the third week) and increase in number throughout the first trimester. As gestation proceeds, the villi become-smaller, the cytotrophoblast layer appears less prominent as the cells become widely separated and the syncytiotrophoblast becomes more irregular and narrow with clustering of the nuclei. The stroma contains the same elements, but the fetal capillaries become much more prominent and come to lie directly under the basement membrane at the periphery of the villus. By the last few weeks of gestation, thinned areas of syncytiotrophoblast form only a narrow rim over the fetal capillaries to become vasculosyncytial membranes, which are specialized zones for the facilitation of gas transfer across the placenta.

            Extravillous trophoblast

         (a) Interstitial trophoblast—After the second week, the budding cytotrophoblast which will ultimately form the villous tree breaks through the primitive syncytium. At the tips of these early villi, cytotrophoblast cells remain as a solid core known as the cytotrophoblast cell columns which fix the villi to the decidua. These are known as anchoring villi. The cytotrophoblast cell columns spread laterally and fuse with neighbouring columns to form a cytotrophoblast shell which encircles the entire embryonic sac. Once this shell has come into physical contact with decidua, isolated interstitial trophoblast cells stream into the uterine mucosa from the shell. After invasion into decidua has occurred the interstitial trophoblast cells become isolated fusiform pleomorphic cells from round uniform cohesive cells.

         On the maternal side of the cytotrophoblast shell in the region of the original primitive syncytium, two irregular narrow strips of fibrinoid necrosis appear known as Rohr’s layer and Nitabuch’s layer. Apart from these small foci of necrosis at the placental-decidual junction, the migration of trophoblast into maternal tissues is remarkable for the lack of tissue destruction or inflammatory response. The invading interstitial trophoblast cells apparently move towards the decidual spiral arteries as trophoblast is always concentrated around the arteries with a much sparser number in the intervening areas. Arteries in the decidua basalis which are surrounded by trophoblast show endothelial swelling and a characteristic destruction of the muscular media which is replaced with ‘fibriroid’ material. This disruption of vessel walls is not found in decidual vessels away from the implantation site. By 8 weeks of pregnancy interstitial trophoblast has extensively colonised the full thickness of the uterine mucosa to reach the decidual-myometrial junction. As the cells move deeper into the decidua, the trophoblast cells become rounder and multinucleated - these are known as placental bed giant cells.

         During the second trimester diffuse infiltration of the inner myometrium by placental bed giant cells like trophoblast is seen causing destruction of the musculo-elastic tissue of the intramyometrial segments of the spiral arteries.

            (b) Endovascular trophoblast—Cytotrophoblast shell provides the origin not only for the interstitial trophoblast but also for endovascular trophoblast. Where the shell comes to lie over the openings of the maternal spiral arteries, the trophoblast cells form a plug seemingly occluding the lumen. From these plug-like areas of the shell, trophoblast moves in a retrograde manner down the spiral arteries ‘like wax dripping down a candle.’¹¹

         Endovascular trophoblast is only seen in vessels which have already been surrounded by interstitial trophoblast and undergone medial disruption. Eventually, trophoblast cells become incorporated into the fibrinoid material in the wall, although the relative contribution of endovascular trophoblast and interstitial trophoblast to this process is uncertain, and may differ in the different segments of the vessel. Interestingly, the walls of the decidual veins are not similarly invaded by trophoblast. Syncytial knots, detached from chorionic villi into the intervillous space, are deported into the maternal circulation via these veins to be trapped in the blood vessels of the lungs where they are eventually lysed.

            Immunohistology of trophoblast

         Many phenotypic differences are evident among the trophoblast sub populations, particularly between villous and extravillous trophoblast. Multinuclear cell formation is the terminal state of both villous and extravillous developmental pathways-but these two end products are not identical. While many placental proteins are detected in syncytiotrophoblast, placental bed giant cells are largely unproductive. In addition, syncytiotrophoblast is HLA class I-(W6/32-) and placental bed giant cells are class I+ (W6/32+), indicating that the immunology of the trophoblast-maternal blood interface is likely to be different from the trophoblast-uterine interface.

            Second and third trimesters

         As pregnancy proceeds, there is a shift in the distribution of changes in the placental bed. In the second half of pregnancy, there is no further trophoblast penetration of the myometrium although vascular transformation by trophoblast appears to continue throughout gestation.¹² For the first 2 months of gestation, villi cover the entire surface of the chorionic sac. As the conceptus expands, the villi beneath the decidua capsularis degenerate, leaving a layer of trophoblast and mesenchyme known as chorion leave. The chorion leave fuses with the decidua parietalis with obliteration of the uterine lumen and regression of the uterine surface epithelium-transport of substances and even maternal cells in the event of infection may occur across this chorio-decidual junction.

            Genomic imprinting and trophoblast development

         Genomic imprinting is the phenomenon whereby genes have differential expression depending on the parental source from which they are inherited. The parental gene that is silent is said to be imprinted. GYNOGENOMES (both sets of haploid chromosomes are derived from the mother) created by replacement of the male pro-nuclei with female pronuclei after fertilization leads to the development of relatively normal sized embryos with poor development of extraembryonic membranes and the placenta. These conceptuses remain viable until the early somite stage, and then undergo involution.

         In contrast ANDROGENOMES (paternally derived chromosomes) created by transplantation of male pronuclei into ova in which the female pronuclei have been removed result in conceptuses with severely stunted embryos, but with well developed extra embryonic trophoblast tissues. Thus, the paternally transmitted genome appears to provide the necessary genetic information for trophoblast growth whereas the maternally transmitted genome is associated with the development of the embryo itself. The classic human example of genome’s imprinting is complete hydatidiform mole. These tumors contain two paternally derived sets of haploid chromosomes (diploid androgenomes).¹³ There is no associated fetus, grape-like distension of chorionic villi together with excessive placental tissue and trophoblast proliferation.

         Genomic imprinting has been viewed as a mechanism where the paternal genes, by having a positive effect on the growth of the placenta, give the fetus a greater chance of survival while the maternal genes, by limiting the growth of the placenta, will reduce the nutritional burden of each pregnancy and thus promote the mother’s chances of a successive pregnancy.^14,15

         In humans four genes are known to be imprinted, H19, IGF-II, WT1 and SNRPN¹⁶. At present it is not known what imprinted genes are important for trophoblast development or for implantation.¹⁷

         The gene encoding insulin-like growth factor II (IGF-II) is maternally imprinted.¹⁸

         Immunohistological studies on human placentae have shown that those trophoblast cells with mitotic activity express receptors for IGF, so the paternally expressed gene encoding this growth factor could be one important element in trophoblast development.

         H19 gene is paternally imprinted.¹⁹ It is expressed by normal trophoblast²⁰ as well as by trophoblast derived from complete moles,²¹ placental site trophoblast tumours and choriocarcinomas.²² Since complete moles are androgenetic, it would appear that the parental source of certain imprinted genes can sometimes be reversed. Loss of imprinting of IGF-II and H19 is also commonly associated with choriocarcinomas.²³

         Wilms’ tumour suppressor gene WT1 has no defined role in placental development.

         SNRPN (small nuclear ribonucleoprotein polypeptide N) is the fourth imprinted human gene.²⁴ To date, there is no evidence that imprinting of this gene affects the reproductive process.

         No data are available for MHC imprinting in the human placenta. Normal trophoblast development is likely to be subjected to the opposing influence of maternally expressed genes. The X chromosome may be especially important in this regard. The paternal X chromosome is preferentially inactivated in human trophoblast²⁵whereas X inactivation is random in somatic tissues, affecting the paternal and maternal X chromosome equally.²⁶ Since the male trophoblast has a single maternal X and female trophoblast has the paternal X preferentially inactivated, this could be a device whereby some X-linked genes important for trophoblast development could behave as though they were paternally imprinted.

         The gene encoding tissue inhibitor of metalloproteinases (TIMP) which may control trophoblast degradation of extra-cellular matrix and hence trophoblast invasion is X linked. The expression of this gene is a further example of how the mother can exert a balancing influence on trophoblast invasiveness.

         Thus the concept of imprinting as a mechanism designed to control maternal fetal interaction during the evolution of the placental form of reproduction is an attractive one.

            Role of retroviruses in trophoblast development

         Retroviruses are RNA enveloped animal viruses which contain a single-stranded RNA genome and an RNA dependent DNA polymerase to transcribe the viral RNA into a double-stranded insertion element for integration into host DNA.^27,28 Genetic mapping studies have identified three genes in the retroviral genome. In order from the 5’ to 3’ end, these are :

         1.   a gag gene which codes for viral structural core proteins.

         2.   a pol gene which codes for reverse transcriptise.

         3.   an env gene which codes for the major envelope proteins. Flanking these are the 5’ and 3’ long terminal repeat (LTR) region which contain sequences essential for viral expression. The inserted retroviral genome is known as the provirus, which can be expressed with the production of viral particles. These viral particles can be classified morphologically into four types (A, B, C & D) that can be transmitted horizontally like other viruses (exogenous transmission). However, an important characteristic of retroviruses is that they can remain unexpressed in the host cell, and be transmitted vertically in the genome though the genuline (endogeneous transmission).

         Many of these viruses appear to be selectively expressed in the placenta.²⁹

         Kato et al (1987)³⁰ have demonstrated that three polyadenylated RNAs of 9, 7.3 and 3.5 Kb long of HERV-R are abundant in first trimester and their human chorionic villi, representing 0.03-0.05% of the total mRNA for this tissue.

         Recent in situ hybridisation studies by Boylet al (1993),³¹ who probed with ERV-3 (HERV-R) envelope DNA generated by PCR amplification of genomic ERV-3 sequences, confirm that the syncytiotrophoblast layer of term placentae is the major site of expression. Presence of retroviral-encoded proteins in the placenta has been demonstrated-which include reverse transcriptase, 75 K Da protein and others. In addition to proteins, the presence of actual retroviral particles in the placenta has long been documented.^32,33 The earliest reports are electron microscopic demonstration of C-type viral particles.^34-42 These viruses are normally located on the convoluted plasma membrane at the basal border of the syncytiotrophoblast in apposition with either the underlying villous cytotrophoblast or basement membrane. Viral particles have never been observed at the brush border of the syncytiotrophoblast in contact with maternal blood in the intervillous space. Placenta is especially permissive for retroviral replication⁴³-incidence being 35-80%.

         It can be concluded that :

         1.   Retroviral transcripts are more abundant in chorionic villous tissue than in associated embryos.

         2.   Placental extracts contain retroviral reverse transcriptase.

         3.   Antisera raised against various retroviral-related proteins react against placental proteins.

         4.   The reverse is also the case in that antisera raised against trophoblast membrane antigens recognise shared determinants on retroviral infected cells.

         5.   Pregnant women show cell mediated & humoral immunity against retroviral proteins, indicating that these proteins are immunogenic in vivo.

         6.   Even complete viral particles may be expressed since C-type particles are frequently observed in human placenta.

         Further data is required to determine how retroviruses are potentially capable of playing an important role in placenta development.

         The factors why retroviruses are especially transcriptionally active in the placenta include hormonal and immunological.²⁷ Estrogen (17B-estradiol) has been reported to induce the production of retroviral proteins and RNA-directed situations of immunological activity. The implantation site represents their mutual recognition of each others could, in same way, trigger retroviral expression in the placenta as well as the uterus.

         It has been suggested that the fusion of villous cytotrophoblast to form the overlying layer of syncytiotrophoblast could be mediated by some retrovirual fusion protein since retroviruses are frequent inducers of syncytium formation.⁴⁴

         Besides affecting the host cell’s own expression of class I antigen, endogenous retroviruses may also encode their own transplantation antigen. In conclusion, it will be appreciated that endogenous retroviruses are potentially capable of playing an important role in placental development.

(D)     ETIOLOGY

         Causes of spontaneous abortions include :

         1.   Genetic abnormalities 50-60%

         2.   Endocrine abnormalities 10-15%

         3.   Chorioamniotic separations 5-10%

         4.   Incompetent cervix 8-15%

         5.   Infections 3-5%

         6.   Abnormal placentation 5-15%

         7.   Immunologic abnormalities 3-5%

         8.   Uterine anatomic abnormalities 1-3%

         9.   Unknown reasons < 5%

            Genetic causes

         Genetic anomalies are the most frequent and important causes of early pregnancy losses. The majority of genetically originated abortions occur before 8 weeks and are blighted ova. Chromosomal abnormalities are found in approximately 80% of blighted ova and 5% to 10% of the abortions in which a fetus is present. The abnormal zygote found in a blighted ovum results from an error in maternal or paternal meiosis I or II, from super fecundation of an egg by two spermatozoids, or from a chromosomal division in the absence of cytoplasmic division.

         The chromosomal abnormalities most frequently found in abortions are⁴⁵—

            I. Autosomal trisomy

            Trisomy 16 is most frequently found in abortus material (@ 60%), and in these cases the sac is completely empty. The trisomy predominantly affects chromosomes 16, 21 and 22, although it may affect any other of the autosomes.

            II. Triploidy

            This anomaly consists of a mean chromosomal count of 69 and occurs in 15% to 20% of all abortions with abnormal chromosomes. In many cases the sac is empty, but if a fetus is present, it has obvious abnormalities (omphalocele syndactyly, cleft lip and palate, etc.). In about 50% of the cases hydropic degeneration of the placenta is present. In triploidy without molar degeneration, the extra chromosomal set is probably maternal.

            III. Monosomy X

            It may be the result of the loss of an X chromosome at the time of fertilization, or it may be the result of nondisjunction during either male or female meiosis. It consists of @ 25% of all abortions with chromosomal abnormalities. About one of every 15 fetuses with 45, X Karyotype will not be aborted and will be identified at birth as an individual with Turner’s syndrome. Some of the abortion specimens with 45, X Karyotypes are sacs containing a small umbilical cord that inserts into an amorphous mass of embryonic tissue.

            IV. Tetraploidy

            This condition probably results from a failure of cytoplasmic division after a chromosome division in the germinal cells. The mean chromosome count is 92, found in 3-6% of blighted ova with abnormal chromosomes. In these cases abortions usually occur very early in pregnancy, and the embryo can not be recognized in the specimen.

            V. Structural rearrangement of the chromosomes

         This group consists of unbalanced translocations and inversions and accounts for 3% to 5% of abortions with abnormal chromosomes. Couples that have abortions with structural rearrangement of the chromosomes should have karyotype analysis because 5% will be carriers of the rearranged chromosomes.

         There are cases of early fetal death in which extensive investigations, including karyotype, give negative results. In these cases the lack of fetal development may be the result of a lethal molecular mutation that can not be identified with the present methods of analysis.

         Animal experiments using insertional mutagenesis (insertion and integration within the genome of a DNA fragment, which causes disruption of the function of one or several genes) have provided evidence indicating that inactivation of the genes containing the information for collagen synthesis causes an arrest in development and embryonic death⁴⁶. Fetal death may also be the result of mutations affecting genes that control the expression of other genes at the transcriptional level (Homeo box genes) or mutations that cause excessive concentration of products and embryonic cell toxicity.

            Endocrine Abnormalities

            I. Progesterone deficiency

         Most of the evidence incriminating progesterone deficiency in early pregnancy loss comes from studies demonstrating that luleal phase deficiency occurs more frequently in patients with recurrent abortions than in control patients^47-49. A rigorous diagnosis of corpus luteum defect requires histologic confirmation of endometrium out of phase by 2 or more days during the secretory period of the menstrual cycle. Because endometrial biopsies are not obtained during pregnancy, the only possible method of documenting a corpus luteum deficiency during gestation is by measuring the serum progesterone concentration⁵⁰. How ever, progesterone production by the corpus luteum is pulsatile and is characterized by marked fluctuations in serum levels⁵¹. As a result, many studies have shown a poor correlation between serum progesterone levels and endometrial biopsy findings.

         Low serum progesterone concentration is seen in patients with blighted ova⁵². Posing a problem in diagnosing progesterone deficiency. Because blighted ova are so frequent, patients with low progesterone concentrations in early pregnancy must be examined for blighted ova rather than treated for corpus luteum deficiency.

         Measurement of 17 hydroxy progesterone a substance produced exclusively by the corpus luteum is a better marker than progesterone. In a recent study, measurements of serum 17-hydroxy progesterone in patients conceiving after the use of ovulation-inducing agents could discriminate patients that miscarried from those that did not. Unfortunately measurement of this substance did not differentiate aborters from non aborters in patients that did not receive ovulation induction.

         Commonest indications for progesterone supplementation are in patients (i) with symptoms of threatened abortion, (ii) conceiving after ovulation induction, (iii) with a serum level < 15 ng/ml, (iv) with a hisory of early pregnancy losses.

            II. Thyroid deficiency

         Usually causes preterm labor after 24 wks. rarely causes early pregnancy loss.

            III. Diabetes

         Diabetes with both an elevated blood glucose and hemoglobin Aic in the first trimester have a significantly increased risk of spontaneous abortion, whereas those with good metabolic control had a risk similar to that of control subjects⁵³.

            IV. Increased Androgen

         A rare endocrine etiology for early pregnancy losses is maternal hyperandrogenicity⁵⁴ with elevated serum levels of testosterone and dehydroepiandrosterone sulfate (DHEAS). Pregnancy loss usually occurs as an early fetal demise at 14 weeks gestational age apparently because of corpus luteum dysfunction.

            V. Polycystic ovary syndrome

         Spontaneous abortions occur more frequently in patients with polycystic ovary syndrome (PCO) than in normal controls⁵⁵. The elevated serum luteinizing hormone concentration seems to have a deleterious effect on the corpus luteum⁵⁶. Pituitary suppression with GnRH agonists followed by hCG administration has been found to be useful in the prevention of this type of miscarriage⁵⁷.

            Infections

         Account for approx 3-5% of cases with early pregnancy loss. They can be :

         1.   Ascending infections facilitated by some degree of cervical incompetence. These infections are usually caused by a mixture of anaerobic and aerobic organisms with predominance of group B streptococci and E. coli. Ascending infections usually recur in subsequent pregnancies, suggesting the presence of a defect in the cervix facilitating uterine invasion by the vaginal flora. Histologic examination of the placenta shows severe chorioamnionitis.

         2.   Hematogenous infections in minority of cases with varicella, parvovirus, rubella, toxoplasmosis, herpes simplex, treponema, listeria, chlamydia, and mycoplasma detected mostly on fetal autopsy or in histologic or bacteriologic examination. Mycoplasma hominis and Ureaplasma urealyticum are occassionally found in cultures of aborted products of conception⁵⁸ which is in contrast with the high frequency of positive cervical and endometrial ureaplasma cultures found in patients with a history of repeated spontaneous abortions or infertility⁵⁹. Diagnosis and treatment of mycoplasma infection before pregnancy prevents recurrent abortion⁶⁰.

            Anatomic abnormalities of the uterus

         Cause approximately 10% to 15% of all abortions with adequate fetal development. Can be divided into

                        Acquired                                                                            Developmental

                  1.   Uterine synechiae                                        1.   Mullerian abnormalities

                  2.   Incompetent cervix                                            a. Septate uterus

                                                                                                b. Bicornuate uterus

                                                                                          2.   DES exposure in utero

                                                                                          3.   Incompetent cervix.

            I. Uterine synechiae

         The association between uterine synechiae and early pregnancy losses has been known since the original work of Asherman in 1947. Uterine synechiae are bandlike structures between the walls of the uterus causing minimal to almost complete obliteration of the uterine cavity. Histologically, these bands are made of fibrous tissue, myometrium, and endometrium. The endometrium around these adhesions is usually atrophic with distorted gland openings.

         In the majority of cases, synechiae are the result of intrauterine infection combined with surgical trauma after the retention of products of conception following abortion or delivery. Schenker and Margalioth⁶¹ found that in 66% and 22% patients, intrauterine adhesions were associated with postabortal and postpartum curettage, respectively. Also 14% of patients with synechiae were found to have a history of multiple pregnancy losses.

         Diagnosis is made by hysterosalpingogram or by direct hysteroscopic vision. Treatment requires surgical division of the fibrous bands, placement of an intrauterine device to avoid contact between the ends of the adhesions and estrogens to stimulate endometrial growth

            II. Mullerian abnormalities

         Septate and bicornuate uteri are associated with early pregnancy losses because of inadequate blood supply to the conceptus when the pregnancy is implanted in the relatively avascular septum. Another mechanism of pregnancy loss in these patients is incompetent cervix, an abnormality that is frequently present in patients with abnormal uterine anatomy. Hence it has been suggested that a prophylactic cervical cerclage should be used in all pregnant patients with congenital uterine anomalies⁶². Other abnormalities such as double uterus and unicornuate uterus are manifested more commonly by preterm labor, characteristically occurring later with each successive pregnancy.

            III. DES exposure in utero

         Approximately 70% of women exposed in utero to DES (diethyl stilboestrol) have a small, T-shaped uterus and an abnormally high frequency of poor pregnancy outcomes⁶³. These patients tend to lose their pregnancies between 20 to 28 weeks because of incompetent cervix. Patients exposed in utero to DES should have a cervical cerclage between 12 and 16 weeks to improve the chances of a normal pregnancy outcome.

            IV. Incompetent cervix

         It is usually the result of cervical trauma, most commonly overzealous mechanical dilation during a pregnancy termination or a diagnostic curettage. Deep cervical lacerations during vaginal delivery and extensive conization for the treatment of cervical dysplasia also cause incompetence. It is also seen in patients with mullerian fusion abnormalities and in patients exposed to DES in utero.

         Diagnosis is made during pregnancy by seeing bulging membranes and painless cervical dilation on perspeculum examination. Diagnosis is suspected in patients with a history of second trimester pregnancy losses and an increased amount of vaginal discharge—USG shows funnelling of membranes into the endocervical canal or diameter of the internal os > 23 mm. In nonpregnant patients, increased compliance of the cervix is demonstrated by passing a No. 8 Hegar or a No. 15 Pratt dilater or a Foley’s catheter filled with 1 ml of water through the internal cervical os.

         Cervical cerclage is placed in such patients at between 12 and 16 weeks transvaginally by Shirodkar or McDonald technique and transabdominally in patients with little cervix left after extensive conization.

            Subchorionic hematomas and

            chorioamniotic separations

         Bleeding occurs between the amnion and the chorion (chorioamniotic separation) or between the chorioamnion and the decidua (Subchorionic hematoma).

         The cause of such bleeding and membrane separation is unknown. Most patients with a severe initial episode do well, and the pregnancy continues without problems after a few days of spotting, but persistent bleeding has a guarded prognosis. Diagnosis is usually made by ultrasound and the estimation of the hematoma size by ultrasound seems to be of prognostic value⁶⁴.

            Defective placentation

         During normal placentation the spiral arteries undergo adaptive changes characterized by loss of the normal musculoelastic arterial wall and replacement by fibrinoid material containing trophoblastic cells. Thence, the narrow, thick-walled arteries are transformed into wide-open, tortuous vascular channels that provide the necessary blood flow for the developing conceptus. Abnormal placentation is synonymous with the lack of these changes. Patients with recurrent abortions caused by abnormal placentation that are able to prolong a pregnancy beyond the second trimester remain at high risk for preeclampsia, preterm labor, and fetal growth retardation. Abnormal placentation occurs with similar frequency in patients with chromosomally normal and abnormal fetuses but is never found in patients with blighted ova.

            Immunological causes

         Alloimmune Etiology—The conceptus acquires genetic material from both its parents. Thus the conceptus inherits and express on the surface of its cells its father’s histocompatibility antigens that are foreign (i.e., allogeneic & alloantigenic) to the mother.

            Defence mechanisms

         The pre-implantation embryo expresses only low levels of paternal histocompatibility antigens which include products of major histocompatibility gene compel such as HLA and non-MHC or minor histocompatibility (H) transplantation antigens⁶⁵.

         This pre-implantation embryo is protected by its small size which reduces the probability of encounter with any maternal effecter cells that have strayed into tubal or peritoneal fluid.

         i.    by lack of direct contact with lymphatic-containing, maternal tissues

         ii.    by its zona pellucida which is shed just before implantation⁶⁶.

         However if cells such as macrophages in tubal or peritoneal fluid become activated-secretion of cytokines such as interleukin-1, tumor necrosis factor alpha and granulocyte macrophage colony stimulating factor (GM-CSF) can lead to embryo damage and... failure may occur e.g. in endometrosis^67-69.

         Radiographic contrast dye used for HSGs may paralyse activated macrophages and this might explain why some couples unexpectedly achieve pregnancy after this type of investigation.

         At the time of implantation expression of H antigens is shut off so that only embryonic antigens are expressed.

         During first 5 post implantation days, there is re-expression of minor paternal H antigens followed on the fifth post-implantation day by MHC expression^65,70. There is paternal MHC expression on the human embryo from 6-20 days post implantation⁷¹ and this is confirmed by immuno-histological study of first trimester embryos⁷².

         The conceptus is in fact 2 grafts in one. The fetal tissue (graft # 1) is enclosed within a sac of membranes lined by fetal trophoblast cells that form the feto-maternal interface and placenta. Conceptus behaves in a manner different from conventional allgrafts and this is primarily due to its trophoblast (graft # 2).

         Fetal trophoblast is a unique tissue the development of which is dependent upon presence of paternal genetic material. Where the genes of the trophoblast are entirely paternal, hydatidiform moles and choricarcinoma can develop⁷³. These neoplastic growths are not rejected by the female even when high levels of paternal HLA antigen are expressed by the tumor⁷⁴.

         Trophoblast may be divided into :

         1.   Non MHC expressing cells founds primarily at the interface between the fetal placental capillaries and maternal blood^75-76.

         2.   MHC expressing interstitial and chorionic membrane trophoblast in contact with deciduas expresses a modified from of class I MHC antigen that lacks individual HLA determinant.

            Defence mechanisms protecting the fetal

            trophoblast and blood supply

         An important feature of most trophoblast tissue is its extreme resistance to rejection by antigen specific immune effector mechanisms⁷⁶ Trophoblast cells are highly resistant to killing by antigen specific cytotoxic76cells and by antibody + cytolytic null type cells (antibody dependent cell-mediated killing).

         Null type cells include :

         i.    Natural killer cells (NK) that lyse transformed and virus-infected cells selectively and without antigen specificity.

         ii.    Natural cytotoxic (NC) cells that lyse cells of solid tumors resistant to NK cells.

         iii.   Macrophages

         iv.   Lymphokine activated Killer cells (LAKs).

         The spontaneous activity of NC and NK cells may be boosted by cytokines released by T cells that mediate delayed type hypersensitivity and generation of LAKS is dependent upon such factors.

·         Trophoblast, while resistant to killing by NK & NC cells, is quite sensitive to killing by LAKS^77-79.

·         NK, NC and activated macrophages produce a cytotoxin called tumor necrosis factor-alpha (TNF-a) that damages a trophoblast cell line (Clark et al, unpublished data).

         Stimulation of TNF-a release can also occur via a non T cell dependent mechanism through exposure of macrophages to bacterial products such as endotoxin⁸⁰. TNF-a acts on vascular endothelium to promote clotting (the thromobosis produces nutritional death of the conceptus).

            Role of decidua

            (a) Suppression

         Prior to implantation, a novel population of hormone induced suppressor cells develops^81,82, in the uterine lining. There is no antigenic specificity and these cells can not release any soluble suppressor factors unlike classic suppressor T cells⁸¹. The relevance of these cells to spontaneous abortion is three fold viz.

         i.    They persist during first 4 days after implantation.

         ii.    During this period of time, the decidua has the ability to exclude macrophages and prevent the expression of delayed type hypersensitivity. Whereas after formation of the placenta, these suppressor T-like cells disappear and expression of DTH becomes possible⁸³.

         iii.   Rate of loss has been found to be increased after injection of anti-body to CD8, presumably by interfering with the hormone induced suppressor T-like cell population⁸⁴.

         By the fifth post-implantation day, a population of small sized suppressor cells replaces the CD8 + suppressor population. These cells are small cells with cytoplasmic granules, lack conventional T cell markers but possess surface receptor for the Fc end of IgG⁸⁵ which are recruited/activated by soluble products from trophoblast cells. These cells release a potent immunosuppresive molecule that is homologous with transforming growth factor beta (TGF-b)⁸⁶ -which in turn inhibits NK cell activation, LAK generation and the response of NC cells to activation by interleukin 3 and inhibits the cytotoxic respiratory burst of monocyte macrophage cells^86,87. TGF-b has the potential to block all of the natural effector mechanisms capable of attacking the trophoblast. TGF-b and TNF-a antigonize each other with respect to effects on the immune response⁸⁸.

            (b) Stimulation

         Presence of growth factors can stimulate placental trophoblast growth. These are cytokines-viz interleukin 3 (a T cell derived molecule) and colony stimulating factors CSF-1⁸⁹ and CSF-GM⁹⁰ produced by T cells and a variety of stromal cells participating in the decidual response.

            Role of HLA sharing in recurrent abortion

         The major histocompatibility complex (MHC) of which HLA is the human variety consists of several linked loci that code for cell surface proteins important in immune function. Class I antigens (HLA loci A, B & C) serve as targets for recognition by cytotoxic T cells and as co-elements in recognition of viral antigens.

         Class II antigens (HLA D Loci) code for surface proteins restricted to a small number of cell-types involved in immune recognition and response. These immune response associated antigens (Ia antigens) associate with a variety of foreign molecules and allow their recognition by T helper/DTH cells.

         In allogeneic recognition reactions, of which the fetomaternal relationship may be one example, Ia differences cause the strongest proliferation of responding T cells in the host, Assumably, Ia is not expressed on the trophoblast due to this reason⁷⁶.

         Since HLA antigens represent strong antigens leading to graft rejection-it was rather surprising to find an increased frequency of HLA antigen sharing among couples with reduced pregnancy success (recurrent abortion)⁹³. It was proposed that the defect in abortion was lack of adequate stimulation of a helpful immune response in the mother due to a hypoantigenic conceptus⁹⁴. 25% of habitually aborting couples showed abnormal reactivity of the wife’s lymphocytes against the husband’s cells in vitro.

         It has been suggested that HLA-D sharing is a harbinger of sharing at a locus called TLX where TLX represents an antigen shared between lymphocytes and trophoblast. If husband and wife share the same TLX, the conceptus will not be antigenic and will not stimulate a protective immune response in the mother. It is proposed that deliberate immunization with husband or third-party lymphocytes will stimulate the protective blocking antibody response that is missing in aborting patients^95-100. This antibody inhibits an MLC reaction by recognizing the antigen as the receptor in the responding T cell^101,102; while it is called a blocking antibody.

         It seems likely that TLX sharing, if it does in fact occur, includes all of the important antigenic determinants (epitopes) on the molecule so that the father and fetus are still foreign but not sufficiently different to elicit an intrauterine immune response. Alternately, what may be shared is an antigen (TLX) which performs a helper function in ensuring an adequate induction of immunity to a minor H antigen on the embryo, possibly an embryonic antigen. Unless the husband is TLX homozygous and shares an antigen in common with his partner in which case a 100% failure rate would be predicted, even after 7-10 recurrent abortions there would still be a 50% or greater chance of success based on simple Mendelian genetics.

         Primary Aborters—No live births, abort only with one mate, have significant HLA sharing, and abort due to deficient immune response to antigens on the fetal trophoblast (possibly TLX). The response is humoral¹⁰³ and is proposed to stimulate local suppressor cell activity in decidua^104,105. These antibodies tend to block allorecognition by maternal T cells in vitro.

         Secondary Aborters —One or more live births, no HLA sharing, may abort with several partners and may possess antibody to paternal cells as detected by a variety of methods^106-108. Abortion is attributed to a toxic effect of the antibody. These patients may have second rather than first-trimester losses, possess anti-phospholipid antibodies that interfere with coagulation (so called lupus anticoagulant) and occasionally have clinically diagnosable autoimmune disease^109,110. It has been suggested that some first trimester aborters and primary aborters may also have such antibodies, and if blocking activity is already present, treatment with aspirin and prednisone (as is done for patients with lupus anticoagulant) is more effective than immunization¹⁰⁰. Antibody to blood group antigen p has also been linked to abortion of p+ fetuses¹¹¹.

         Revised classification may be :

         1.   Primary aborters without evidence of autoantibodies or blocking factors.

         2.   Secondary aborters without evidence of autoantibodies or blocking factors, but possibly possessing antibodies against antigens of the husband.

         3.   Primary and secondary aborters with evidence of autoantibodies or blocking factors^113,114.

         4.   Genetic aborters¹¹³.

         First two groups may be suitable for immunotherapy.

            Immunotherapy

         Success rates as high as 95% has been noted with immunotherapy when patients with pre-existing blocking antibodies¹⁰⁰ were excluded.

         Atternative forms of treatment for recurrent abortion

            Treatment                                                                                                                               Success Rate

         1.   Human Placental trophoblast¹¹⁵                                                                    75%

         2.   hCG¹¹⁶                                                                                                        94%

         3.   Corpus lutum extract¹¹⁷                                                                                84%

         4.   Progesterone¹¹⁸                                                                                           84%

         5.   Tender loving care¹¹⁹                                                                                    72%

         6.   Cerclage¹²⁰                                                                                                  85%

            Antiphospholipid antibodies

         Antiphospholipid antibodies comprise a family of autoandibodies which have a well-established association with fetal loss^121,122. They are most commonly associated with fetal loss, thrombocytopenia or thrombotic events without evidence of auto immune disease viz. the antiphospholipid syndrome.

         Primary Antiphospholipid Syndrome—must include one clinical and one serological feature.

            Clinical features

         1.   Recurrent venous or arterial thrombosis

         2.   Recurrent fetal loss

         3.   Thrombocytopenia

            Serological features

         1.   IgG aCL>20 GPL

         2.   LA

         3.   IgM aCL>20 MPL + LA

         I MPL or GPL is equivalent to 1 ug of affinity purified aCL IgM or IgG.

         Diagnostic Tests—The antiphospholipid antibodies are a diverse family of autoantibodies which share in common a reactivity with negatively charged phospholipids.

         There are three clinically significant members—viz. biological false positive test for syphilis.

·         Lupus anticoagulant

·         Anticardiolipin antibodies.

            I. Biological False Positive Test For Syphilis

         A biological false positive test can result from :

         1.   Antibodies produced in response to infection by a number of non treponemal pathogens¹²³. In this situation the BFP is likely to be transient, reflecting recent activity of the pathogen. These antibodies are not associated with thromobosis or fetal loss.

         2.   Autoantibodies produced by patients with autoimmune discuses, particularly patients with APS and/or SLE. These autoantibodies can persist for many years and are associated with both thrombosis and fetal loss.

            The BFP reaction requires

         1.   that an phospholipid based screening assay such as the VDRL is persistently positive for 6 months.

         2.   that the absence of treponemal infection is confirmed by a non-phospholipid-based assay such as the TPHA¹²⁴.

            II. Lupus Anticoagulant

         It comprises autoantibodies of either IgG or IgM class which prolong phospholipid-dependent coagulation assays by reacting with negatively charged phospholipids¹²⁵. Coagulation assays used to screen for LA include the activated partial thromboplastin time (APTT), Kaolin clotting time (KCT), dilute Russell Viper venom test (dRVVT) and dilule tissue thromboplastin assay (dTta)¹²⁶.

         The minimum criteria for the detection of LA proposed by Triplett & Brandt (1989) are :

         1.   A prolongation of a phospholipid-dependent screening test such as the APTT.

         2.   Demonstration that the abnormality is due to an inhibitor rather than a factor deficiency.

         3.   Proof that the inhibitor is directed against phospholipids¹²⁶.

         In order to confirm the presence of an inhibitor the screening test is repeated using a mixture (1 : 1) of the patient’s plasma with normal plasma. If the abnormality is due to an inhibitor the test will remain prolonged. If the abnormality is due to a factor deficiency the normal plasma will act as a source of the factor and the test result will correct to normal.

         Similarly a platelet neutralization procedure (PNP) is used to confirm the antiphospholipid nature of an inhibitor, Lysed platelets are added to the abnormal plasma, and the screening test (APTT or dRVVT) is repeated. An abnormality caused by LA will correct in a PNP whilst an abnormality due to a factor inhibitor will not.

            III. Anticardiolipin Antibodies

         It is detected by solid phase immune assay (enzyme-linked immuno sorbent assay (ELISA) or radio immuno assay (RIA)^127,128. The aCL assay has proved to be 200-400 times more sensitive than the VDRL test and detected 90% of LA in a population¹²⁷.

            IV. Antibodies to other phospholipids

         Phospholipid antigens other than cardiolipin have been used in solid-phase assays for aPL because cardiolipin being found exclusively in the mitochondria is unlikely to be the physiological antigen for aCL as it is not exposed to circulating aPL¹²⁹.

         Phosphatidyl serine has been used as an alternative antigen as it is located in membranes of endothelial cells and platelets. However, PS is found in the interior leaflet of non-activated cell membranes and is only exposed to circulating antibodies after cell activation causes its transfer to the exterior leaflet. Some use phosphatidyl ethanol amine which is in both the exterior and interior leaflets of cell membranes and thus is exposed to circulating aPL.

            Anticardiolipin antibody cofactor, B2 Glycoprotein

         The requirement of a cofactor to facilitate the binding of aCL to phospholipids has been described recently^130-132. The cofactor, which is present in normal serum, has been identified as B2-glycoprotein (B2-GP1 - also called apolipoprotein H)¹³¹.

         The invitro properties of B2 - GP1 are :

         1.   Inhibitor of intrinsic phase of coagulation.

         2.   Inhibitor of platelet activation.

         3.   Phospholipid binding protein^134-136.

         The formation of a trimolecular complex involving B2 - GP1, aCL and phospholipid prevents the normal function of B2 - GP. Also, Matsuura et al (1990)¹³² have demonstrated that, unlike aCL of autoimmune origin, aCL resulting from syphilitic or other infections bind to phospholipid in the absence of B2 - GP1. This difference in the cofactor requirements of aCL from various origins explains why only aCL of autoimmune origin are associated with thrombosis and recurrent fetal loss.

         Prevalence of aPL, both LA and aCL, in the general obstetric population is 2%.

         There is a definite correlation between the presence of aCL and LA. Among patients with SLE and LA, 59% will have aCL and in those with SLE and aCL, 45% will have LA. Incidence of aPL in women with three or more first trimester miscarriages varies between 14%¹³⁷ and 42%¹³⁸. Parke et al (1991)¹³⁹ found a positive frequency of aPL in women with recurrent miscarriage as 16%, women who had undergone normal pregnancy as 7% and women who had never been pregnant as 3%. Late fetal losses are also associated with aPL-incidence being 30-40%. Parazzini et al (1991)¹⁴⁰ concluded that there is no apparent justification for considering aPL to be a risk factor for fetal loss among women who present with spontaneous miscarriage or fetal death and have no previous spontaneous fetal loss. Lockshin (1987)¹⁴¹ noted that fetal death occurred in 77% women with SLE with aCL alone versus 50% patients with LA alone. He concluded that aCL was more sensitive and specific for predicting fetal death than LA.

(E)      PATHOGENESIS

            I. Effect on Platelets

         Early workers described thrombocytopenia as a frequent finding in patients with aPL¹⁴². Also thrombosis associated with APS is congruent with aPL binding to and disrupting the function of platelet membrane phospholipids. However, the negatively charged phospholipids, with which aPL react, are present only in the inner leaflet¹⁴³. Therefore, non activated platelets should not be antigenic for aPL. Upon activation the distribution of phospholipids in the platelet membrane is altered with the transfer of the negatively charged PS to the exterior leaflet of the membrane. This permits the platelet to participate in the processes of coagulation and may make them antigenic for aPL.

            II. Effect on the Vascular Endothelium

         (1) Inhibition of prostacyclin production—aPL which reacts with membrane phospholipid would inhibit the production of prostacyclin by the vascular endothelium and promote thrombosis¹⁴⁴. aPL inhibits the release of arachidonic acid from membrane phospholipid¹⁴⁴.

         (2)Inactivation of the protein C/protein S/Thrombomodulin pathway—Protein C acts as an anticoagulant by inhibiting the activated coagulation factors Va, VIIIa and platelet bound Va-Xa complex¹⁴⁵. Two steps in this pathway are phospholipid dependent and may be inhibited by aPL.

         a.   Protein C is activated by thrombo modulin - a protein present on the surface of vascular endothelial cells. Thrombomoduln must be bound to phospholipid to yield optimum protein C activation¹⁴⁶. It has been demonstrated that aPL can inhibit the invitro activation of protein C by thrombomodulin^146,147.

         b.   Once activated, protein C requires a co-factor, designated protein S, which facilitates the binding of activated protein C to the platelet membrane¹⁴⁸. Once bound to the platelet membrane, the protein C/protein S complex then inhibits coagulation factors Va and Xa. aPL can interact with phospholipids and inhibit the protein S dependent anticoagulant activity of activated protein C.

         (3) Inhibition of antithrombin III anticoagulant pathway—Endothelial cells express upon their surface heparin like molecules-glycosaminoglycans (GAGS)-which activate antithrombin III¹⁴⁹. The GAG heparan sulphate is believed to be the physiological activator of ATIII and is particularly important as an anticoagulant in the microcirculation¹⁵⁰. It has been proposed that aPL could cross-react with GAGS and inhibit the activation of ATIII.

            Clinical Associations with APS

         The hall mark of the presence of aPL is the triad of arterial and venous thrombocytopenia and fetal loss.

            Medical Disorders

         (I)      Neurological Disorders

                  1.   Transient ischaemic attacks

                  2.   Amaurosis fugax

                  3.   Migraine like headaches

                  4.   Acute ischaemic encephalopathy

                  5.   Multi-infarct dementia

                  6.   Degenerative myelopathy.

         (II)      Cardiac Disorders

                  1.   Valvular lesions

                  2.   Valvular or chamber thrombosis

                  3.   Coronary artery occlusion

         (III)     Adrenal gland involvement

                  1.   Addison’s disease

         (IV)    Skin

                  1.   Live do reticularis

         (V)     Lungs

                  1.   Pulmonary hypertension

         Snedden’s Syndrome comprises the clinical triad of live do reticularis, cerebrovascular occlusion and labile hypertension in the presence of aPL¹⁵¹. Evidence for an underlying collagen disorder can be found in 30-40% of individuals with aPL.

         Drugs such as procainamide, guanidine, phenytoin, chlorpromazine, valproic acid, amoxycillin, hydrallazine and propanolol have been reported to induce aPL but drug induced aPL is generally not associated with thrombotic complications or fetal loss.

            Obstetric disorders

         1.   First trimester miscarriage.

         2.   Later fetal loss with evidence of growth retardation.

         3.   Later fetal loss without evidence of growth retardation.

         4.   Placental abruption.

         5.   Pre-eclampsia-often severe and early onset.

         6.   Chorea gravidarum.

         Bird Sall et al (1992)¹⁵² found that 33% of women who had a shill birth due to placental abrupion had aPL.

            Indications for investigation of the presence of antiphospholipid

            antibodies in an obstetric population

         1.    All autoimmune diseases

         2.    Thrombocytopenia

         3.    Previous arterial or venous thrombotic event

         4.    BFP VDRL

         5.    Recurrent (>3) first trimester miscarriages

         6.    All fetal losses after 20 weeks of pregnancy

         7.    Placental abruption (previous or current pregnancy)

         8.    Fetal growth retardation (previous or current pregnancy)

         9.    Severe early onset pre-eclampsia (previous or current pregnancy)

         10. Chorea gravid arum.

            Management

            Specific therapy

         I.    Women with antiphospholipid antibodies without previous loss or significant medical disease.

               —Managed with careful monitoring alone

         II.    Antiphospholipid syndrome with fetal loss

               A.   Low level antibodies (aCL<60 GPL; KCT<250s)

                     Most patients of this group deliver a normally grown infant at term without intervention.

                     These women are treated with low dose aspirin—monitoring their LA level during pregnancy and observe closely.

               B.   High litre antibodies (aCL>60 GPL; KCT>250s).

                     These patients require active treatment with close monitoring of fetus and mother.

                     Treatment is with low dose aspirin + (i) Heparin 10,000 IU s/c BD in patients with previous thrombotic event.

         (ii)   Corticosteroids to reduce KCT<200 s in patients with no thrombotic event.

            Various therapies for APS

            (A) Single Agent Therapy

            1. Corticosteroids :

         Mechanism of action—Immune suppression by inhibiting the production of Interleukin 2 by T4 cells.

         Dosage—40-60 mg/day

         The steroid dose is titrated against the KCT value to maintain a KCT value below 200s, usually commencing with prednisone 40 mg and measuring KCT values monthly. For maintenance low dose prednisone 10-20 mg/day is used. Administration of corticosteroids will suppress the LA but not aCL.

         Side effects : Cushinoid features, acne, adrenal insufficiency, diabetes mellitus, oral candida, hypertension, osteoporosis.

         Overall live birth rate—44%.

            2. Heparin :

         Mechanism of action—Facilitates the action of ATIII.

         Dosage—Subcutaneous heparin sufficient to increase APTT to 1.5-2 times normal in those with normal APTT, or if APTT prolonged already heparin dose sufficient to achieve thrombin time of >100s.

         Usually a dosage of 10,000 IU twice daily is used in women with previous thromboembolis or those with aPL in high titre.

         Side effects : Bruising, thrombocytopenia, osteoporosis.

         Overall live birth rate—77%.

            3. Aspirin :

         Mechanism of action—Inhibits cyclooxygenase in the platelet which preferentially lowers platelet thromboxane, leaving endothelial prostacyclin synthesis relatively intact.

         Dosage—75-80 mg per day.

         Overall live birth rate—83%.

            (B) Combination therapy

         1.   Coricosteroid + low dose aspirin

               Over all live birth rate—68%

         2.   Heparin plus low-dose aspirin

               Overall live birth rate—88%

         3.   Azathioprine plus corticosteroid

               (75-100 mg)

         4.   Immunoglobulin prednisone plus low dose aspirin

               Overall live birth rate—71%.

            General Care

         Woman with APS is at significant risk of hypertension, venous thrombosis, pulmonary embolism or thrombocytopenia. Besides she may develop activation of her autoimmune disease.

            (C) Asymptomatic Antinuclear Antibodies

         Significantly greater prevalence of low-titer antinuclear antibodies (ANA) has been reported in patients with unexplained fetal losses before viability than in normal control subjects¹⁵³. High frequency of positive ANA titres also occurs in patients with fetal losses caused by nonimmunologic factors such as uterine anatomic malformations and luteal phase defect^154,155. ANA titre is between 1 : 20 and 1 : 160 and the fluorescent pattern is usually speckled or homogenous. Occasionally there will be a positive history of nasal or oral ulcerations, skin sensitivity, or unexplained recurrent musculo-skeletal pain. A positive ANA tite in a patient with prior early pregnancy losses indicates the need for further investigations of autoimmune factors viz. LAC; ACA and BFP-ST. If any of these tests is positive, autoimmunity is most probably responsible for pregnancy losses.

(F)      CLINICAL FEATURES

         The clinical types of abortions which obstetrician usually encounters regarding early pregnancy losses are

         1.   The patient with first trimester vaginal bleeding sub grouped into

               a.   Threatened abortion

               b.   Inevitable abortion :

                     Complete

                     Incomplete

         2.   The patient with fetal death or second trimester abortion—Missed Abortion

         3.   The patient that has a history of multiple early pregnancy losses—Recurrent Abortion.

            Threatened Abortion

         The most common symptom of a patient with impending abortion is vaginal bleeding. In most patients there is an interval of several days between the onset of symptoms and the actual miscarriage. However, in some the symptoms progress rapidly so these patients should be examined within a reasonable time.

         At the time the ovum is becoming embedded in the uterus, that is, at the time the first menstrual period is missed or soon after, a slight implantation hemorrhage is not uncommon. This diagnosis can only be made in retrospect and the safe rule is to regard any bleeding, no matter how little, which is not explained by a lesion on the vagina or cervix as evidence of threatened abortion. The bleeding is indicative of some degree of separation of the chorion from the decidua and it varies in amount, duration and type. At first the discharge is bright red. when it changes to dark brown it means that active bleeding has ceased and that old blood in the uterus is undergoing dissolution.

         Sometimes the patient also complains of backache and mild lower abdominal discomfort due to uterine contractions. However the pain is usually mild in these cases.

            Inevitable Abortion

         Abortion becomes inevitable if, in addition to the clinical features of threatened abortion, there are painful uterine contractions, dilatation of the cervix, or extrusion of some part of the conceptus through the os. Ballooning of upper vagina, tenderness of the uterus and pyrexia are other suggestive signs.

         Many abortions occur with a minimum of warning and upset, others are characterized by recurrent haemorrhage and this may produce serious shock. Retention of the products in the canal can itself produce shock.

         Incomplete Abortion—The bleeding does not get progressively less but varies from day to day, becoming heavy from time to time. It continues intermittently for weeks and months and may be accompanied by periodic uterine colic. A normal menstrual rhythm is not re-established.

         In long standing cases, the attachment of the chorion to the uterus becomes organized into fibrous tissue and becomes a placental polyp.

         On examination the uterus is slightly enlarged and often softened. Internal os is patulous—a very important sign indicating that there is something within the uterus.

            Complete Abortion

         The bloody discharge decreases progressively and usually ceases within 7-10 days. The first menstrual period occurs 4-6 weeks later.

            Clinical assessment of patients with

            first trimester vaginal bleeding

         Firstly the gestational age of the pregnancy should be estimated by clinical dating and by assessment of uterine size. A large for date uterus may indicate a Hydatidiform-mole whereas a small uterus suggests a blighted ovum. Presence of a tender adnexal mass suggests an ectopic pregnancy. Also, the pelvic examination may show cervical changes if the process is advanced.

         For pregnancy of less than 6 weeks GA, serum hCG is done to decide whether an ultrasound should be done and how to interpret its results. It is possible to see a gestational sac inside the uterus using TVS only when S.hCG is 1000 MIU/ml or more.

         Patients with first trimester bleeding and serum hCG levels below this critical value should have a repeated quantitative hCG evaluation 3 days later.

• If the hCG value doubles—pregnancy is likely to be intrauterine and there is a high probability of a normal outcome.
• If the hCG value does not double and initial progesterone and estradiol concentrations are low (progesterone < 15 ng/ml estradiol < 200 mg/ml), then the pregnancy is abnormal, either a blighted ovum or an ectopic.

         The predictive accuracy of low values for hCG, progesterone, and estradiol is 90-95% and most patients exhibiting this combination have spontaneous abortions or tubal pregnancies. The prediction of a normal outcome based on favorable hormone concentrations early in gestation is accurate in 80% of the cases. Also sonographic evidence of normal fetal cardiac activity in these patients indicates a low risk for abortion. Simpson et al¹⁵⁶ demonstrated that the fetal loss rate after documenting the presence of fetal heart motion by USG at 8 weeks is only 3.2%. This finding has been confirmed by other investigators¹⁵⁷.

         Patients with first trimester vaginal bleeding as well as reassuring ultrasound and hormonal findings should be told that the possibility of spontaneous abortion is small (2.5% to 3.2%). Bed rest + progesterone supplementation in cases with serum value < 15 ng/ml is recommended in these cases.

         Patients with blighted ova need karyotyping of the products of conception. If the karyotype of the blighted ovum reveals an autosomic trisomy, the patient will be at higher risk for a subsequent trisomy offspring and genetic amniocentesis should be recommended in subsequent pregnancies. If the karyotype shows structural rearrangement of the chromosomes, parents should go for karyotyping to rule out the possibility that one is a carrier of a translocation or an inversion.

         Cases with inevitable abortion require evaluation of the uterus under general anesthesia. However patients in shock should be resuscitated first.

            Missed Abortion

         Sometimes the fetus dies in utero but the uterus fails to respond normally by expelling it. The fetus then becomes macerated or mummified, the liquor amnii is absorbed and the placenta becomes pale and thin. Carneous (Blood) Mole is one variant caused by multiple hemorrhages in the choriodecidual space. A mass of partly organized blood clot and chorion is formed eventually after absorption of the dead fetus.

         Patient usually has complained of slight uterine bleeding following symptoms and signs of a normal pregnancy. The discharge clears up temporarily and this pregnancy is apparently progressing normally until it becomes clear after some weeks of observation that the uterus is not growing, indeed it becomes smaller and harder in consistency. Sooner or later hemorrhage recurs or there may be an intermittent brown discharge. Breast signs retrogress and symptoms such as nausea disappear. Diagnosis is usually made by ultrasound.

         Complications include disseminated intravascular coagulation caused by products of placental degeneration but is unlikely to occur for at least 3 weeks after the death of the pregnancy. Stimulation of the uterus can precipitate the trouble presumably by raising the intrauterine pressure to drive thromboplastin into the circulation. So blood tests for fibrinogen content and for clot stability before and during the induction of abortion are must. Normal level of fibrinogen is 350-450 mg%. The danger level is < 100 mg%.

         Treatment of afibrinogenaemia is only evacuation of uterus which causes correction in 12-14 hours.

         Histologic and microbiologic examination of the placenta is a fundamental part of the evaluation of these patients as the most common reasons for early fetal demise include chromosome abnormalities, antiphospholipid syndrome, ascending infection, subchorionic hematomas and abnormal placentation. Placenta will show extensive acute inflammatory changes in patients with ascending infection and typical lesions in patients with chronic villitis caused by cytomegalic virus infection. Thrombosis of fetal and maternal vessels will be seen in patients with protein C deficiency—possibility of fetal chromosome abnormalities can also be diagnosed.

         Fetal autopsy to rule out genetic syndrome, fetal blood for karytyping as well as for bactrial cultures for mycoplasma, ureaplasma, chlamydia and listeria may be done.

         ANA titer, anticardiolipin antibody, lupus anticoagulant and a TORCH titer may also help searching for the cause. HSG is recommended in unexplained cases a few weeks after their miscarriage to rule out a uterine anomaly. Induction with laminaria tents for cervical dilation followed by prostaglandin suppositories or high dose oxytocin are preferable over uterine evacuation and curettage—so as to get an intact fetus and placenta.

         Recurrent Abortions—Incidence is 1%. Three consecutive pregnancies ending in spontaneous abortion constitute the definition of recurrent abortion.

            Clinical Assessment and Diagnostic Work-up

         Majority of patients will have recurrent anembryonic abortions or blighted ova in this group. Following history and findings are suggestive of recurrent blighted ova :

         i.    Finding of molar degeneration in the histological examination of the products of conception of a prior abortion is suggestive of triploidy.

           ii.    Previous abortions showing empty sacs on ultrasound examination is suggestive of trisomy 16.

           iii.   History of a malformed infant in the family suggests balanced translocation in one of the parents.

           iv.   Interval infertility is again suggestive of balanced translocation in the parents.

           v.   History of repetitive abortions occurring before 12 weeks.

         (a) The probabily of a successful pregnancy in such cases is 62%¹⁵⁸ without any treatment.

         (b) Recurrent abortions with early fetal demise—Need evaluation of following factors.

               i.    Anatomic abnormalities of uterus and cervix best assessed by HSG (Hysterosalpingography) surgery, cervical cerclage can correct the defect.

               ii.    Corpus luteum deficiency :

                     Results in progesterone deficiency leading to pregnancy loss in first trimester especially between 8 and 12 weeks when the production of progesterone switches from the corpus luteum to the developing placenta. Typically, these patients have uterine contractions for several days preceding the onset of bleeding and the abortion. Diagnosis is made by endometrial biopsy or serum progesterone estimation during the secretory phase of the cycle.

               iii.   PCOD & Hyperandrogenism diagnosed by serum luteinizing hormone/follicle-stimulating hormone (LH/FSH) ratio and total testosterone concentrations. Other endocrinal problems like thyroid gland dysfunction should be treated. GnRH agonists for pituitary suppression followed by induction of ovulation with hCG is helpful in PCOD. For hyperandrogenism prednisone is used.

               iv.   Growth retarded but chromosomally normal fetuses. A biopsy of placental implantation area (obtained by curetting at the time of delivery) shows the lack of physiologic changes in the spiral arteries.

               v.   Autoimmune disorders diagnosed by ANA titer, lypus anticoagulant, VDRL, anticardiolipin and SS-A antibodies. Testing for HLA sharing or antipaternal antibodies has no role in establishing a prognosis. Low dose aspirin + prednisone 40-60 mg/day or low dose aspirin + subcutaneous heparin 5000 U 12 hrly is recommended.

               vi.   Repeated ascending infections due to group B streptococci, mycoplasma, ureaplasma and chlamydia can cause repetitive early pregnancy losses in patients who are carriers of the organism causing the infection and in patients with some degree of cervical in competence.

         Cerclage operation + antibiotic treatment is recommended in these cases.

(G)     CONCLUSION

         Differentiation of all abortion cases into embryonic and anembryonic pregnancies before abortion occurs with the advent of ultrasonography has simplified the counselling of patients with threatened abortions and of patients who had one or more spontaneous abortions. Couples with blighted ova do not require extensive workup, whereas patients who have aborted cytogenetically normal fetuses need as extensive search for nongenetic factors responsible for the pregnancy loss.

         Eventually it can be concluded that the obstetrician should investigate each pregnancy loss until the underlying cause is discovered. Otherwise, the diagnosis and management of patients with repetitive early pregnancy losses will be affected by inadequate information about the nature of their prior abortions.

(H)     REFERENCES

                 1.    Miller JF, Williamson E, Glue J, et al : Fetal lose after implantation : A prospective study. Lancet 1980; 2 : 554.

                 2.    Wilcox AJ, Weinberg CR, O’Connor JF, et al : Incidence of early loss of pregnancy. N Engl J Med 1988; 319 : 189-194.

                 3.    Regan L 1988 A prospective study of spontaneous abortion. In : Beard RW, Sharp F (eds) Early pregnancy failure, mechanims and treatment. Peacock Press, England, p 23-42.

                 4.    Grudzinskas JG, Stabile I, Campbell S 1988. Early pregnancy failure : biochemical and biophysical assessment. In : Beard RW, Sharp F (eds) Early pregnancy failure, mechanisms and treatment. Peacock Press, England, p 183-192.

                 5.    Simpson JL, Mills JL, Holmes LB et al 1987 Low fetal loss rates after ultrasound-proved viability in early pregnancy. Journal of the American Medical Association 258 : 2555-2557.

                 6.    Boue J, Boue A, Lazar P 1975. Retrospective and prospactive epidemiological studies of 1500 Karyotypes spontaneous human abortions. Teratoloty 12 : 11-26.

                 7.    Chambers, WH, Adamkiewicz, T and Houchins. JP (1993) Type-II integral membrane preteins with characteristics of C-type animal lectivs expressed by natural killer (NK) cells. Glycobiology 3 : 9-14.

                 8.    Bezouska, K., Yuen, CT, O’Brien, J, Childs, RA, Chai, W, Lawson, AM, Drbal, K, Fiserova, A, Pospisi, M and Feizi, T (1994). Oligosaccharide ligands for NKR-P1 protein activate NK cells and cytotoxicity. Nature 372 : 150-157l.

                 9.    King, A and Loke, YW (1988). Differential expression of blood group related carbobydrate antigens by trophoblast subpopulatins. Placenta 9 : 513-521.

              10.    Crane, JP and Cheung, SW (1998) An embryonic model to explain cytogenetic inconsistencies observed in chorionic villous versus fetal tissue. Prenatal Liganosis 8 : 119-129.

              11.    Ramsey, EM and Donner, MW (1980) Placental Vasculature and circulation. Georg Thieme, Stuttgart, pp. 12-13.

              12.    Pijenenbarg R (1994) Trophoblast invasion. Reproductive medicine Review 3 : 53-73.

              13.    Lawler, SD and Fisher, RA (1993) The contribution of the paternal genome : hydatidiform mole and choriocarscinoma. In : The Human Placenta, Eds. CWG, Sargent, IL and StarKey, PM, Blackwell Scientific, Oxford, pp. 82-112.

              14.    Moore, T and Haig, D (1991) Genomic imprinting is mammatian development : a parental tug-of-war. Trends in Genetics 7 : 45-49.

              15.    Haig, D (1993) Genetic conflicts in human pregnancy. Quarterly Review of Biology 68 : 495-532.

              16.    Jinno, Y, Yum, K, Nishikawa, K, Kubota, T Ogawa, O Reeve, AE and Niikawa, N (1994) Mosaic and polymorphic imprinting of the WTI gene in humans. Nature Genetics 6 : 305-309.

              17.    Goshen, R, Ben-Rafael, Z Gon. KB, Lustig, O, Tannos, V, de-Groot, N and Hochberg, AA (1994). The role of genomic imprinting in implantation. Fertility and Sterility 62 : 903-910.

              18.    GiannouKakis, N, Deal, C Paquette, J, Good year, CG and Polychromakos, C (1993) Parental genomic imprinting of the human 1GF2 gene. Nature Genetics 4 : 98-101.

              19.    Zhang, Y and Tycko, B (1992) Monoallelic expression of the human H19 gene. Nature Genetics 1 : 40-44.

              20.    Rachmilewitz, J, Gileadi, O Elder-Geva, T, Schneider, ., De Groot, N and Hochberg, A (1992) Transcription of the H19 gene in differentiating cytotrophoblast from human placenta. Molecular Reproduction and development 32 : 196-202.

              21.    Mutter, GL, Stewart CL, Chaponot ML and Pomponio RJ (1993) Oppositely imprinted genes H19 and insulin-like growth factor 2 are co-expressed in human androgenetic trophoblast. American Journal of Human Genetics 53 : 1096-1102.

              22.    Ariel I, Lustig O, Oyer CE, Elkin M, Gonik B, Rachmilewitz, J Biran, H Goshen R, De Groot N and Hochberg A (1994) Relaxation of imprinting in trophoblastic disease. Gynecology and Oncology 53 : 212-219.

              23.    Hashimoto K, Azuma C, Koyama M, Ohashi K, Kamiura S, Nobunaga T, Kimura T, Tokugawa Y, Kanai T and Safi F (1995). Loss of imprinting in choriocarcinoma. Nature Genetics 9 : 109-110.

              24.    Reed ML and Leff SE (1994) Maternal imprinting of human SNRPN, a gene deleted in Prader-Willi syndrome. Nature Genetics 6 : 163-167.

              25.    Harrison KB (1989) X-chromosome inactivation in the human cytotrophoblast, Cytogenetics and cell Genetics 52 : 37-41.

              26.    Lyon MF (1993) Epigenetic inheritance in mammals. Trends in Genetics 9 : 123-128.

              27.    Hellman A, Weislow OS, Twardzik DR and Fowler AK (1979) Type C retrovirus activation and possible functions in the normal and tumour-bearing host. cancer Research 39 : 2902-2907.

              28.    Coffin J. M. (1990) Retroviridae and replication. In : Virology, Eds. Fields B. N. and Knipe D. M., Raven Press, New York pp. 1437-1500.

              29.    Johnson P. M., Lyden T. W. and Mwenda J. M. (1990) Endogenous retroviral expression in the human placenta. American Journal of Reproductive Immunology 23 : 115-120.

              30.    Kato N, Pfeifer-Ohlsson S, Kato M, Larsson E, Rydnest J, Ohlsson R and Cohen M (1987) Tissue-specific expression of human provirus ERV3 mRNA in human placenta : two of the three ERV3 mRNA contain human cellular sequences. Journal of Virology 61 : 2182-2191.

              31.    Boyd, MT, Bax, CM, Bax, BE, Bloxam, DL and Weiss, RA (1993) The human endogenous retrovirus ERV-3 is upregualted in differentiating placenatal trophoblast cells virology 196 : 905-909.

              32.    Parem, S (1979) C-type virus expression in the placenta. Current Topics in Pathology 66 : 175-189.

              33.    Kalter, SS (1983) Viral expression in the trophoblast. In : Biology of Trobhoblast, Eds. Loke, YW and Whyte, A, Elsevier, Amsterdam, pp. 627-662.

              34.    Kalter, SS, Helmke, RJ, Heberling, RL, Panigel, M, Fowler, AK, Strickland, JE and Hellman, AJ (1973) C-type particles in normal human placentas. Journal of the National Cancer Institute 50 : 1081-1083.

              35.    Benveniste, RE, Lieber, MM, Livingston, D, Sherr, CJ and Todaro, G (1974) Infertious C-type virus isolated from a baboon placenta. Nature 248 : 17-20.

              36.    Dalton AJ, Hellman, A, Kalter, SS and Helmke, RJ (1974) Ultrastructural comparision of placental virus with several type-C oncogenic viruses. Journal of the National Cancer Institute 52 : 1379-1381.

              37.    Vernon, ML McMahon, JM and Hackett, JJ (1974) Additional evidence of C particlesx in human placentas. Journal of the National Cancer Institute 52 : 987-989.

              38.    Feldman, D (1975) An electron microscopic study of virus particles in rhesus monkey placenta. Proceedings of the National Academy of Sciences USA 72 : 118-121.

              39.    Kalter, SS, Heberling, RL, Smith, GC and Helmke, RJ (1975) C-type viruses in chimpanzee (Pan sp.) placentas. Journal of the National Cancer Institute 55 : 735-736.

              40.    Seman, G, Levy, BM, Panigel, M and Dmochowski, L (1975) Type-C virus particles in placenta of the cottontop marmoset (sanguinus oedipus). Journal of the National Cencer Institute 54 : 251-252.

              41.    Imamura, M, Phillips, PE and Mellors, RC (1976). The occurrence and frequency of type-C virus-like particles in placentas from patients with systemic lupus erythmatosus and from normal subjects. American Journal of Pathology 83 : 383-389.

              42.    Dirksen, ER and Levy, JA (1977) virus-like particles in placentas from normal individuals and patients with systemic lupus erythematosus. Journal of the National Cencer Institute 59 : 1187-1192.

              43.    Chandra, S, Liszczak, T, Korol, W and Jensen, EM (1970). Type-C particles in human tissues. I. Electron microsupic study of embryonic tissues in vivo and vitro. International Journal of Cancer 6 : 40-45.

              44.    Owens, R J, Tanner, CC, Mulligan, MJ, Srinivas, RV and Compans, RW (1990) oligopeptide inhibitors of HIV-induced syncytium formation. AIDS Research and Human Retroviruses 6 : 1289-1296.

              45.    Etiben B, Bartels I, Bahr-Porsch S, et al : Cytogenetic analysis of 750 spontaneous abortions with the direct preparation method of chorionic villi and its implications for studying genetic causes of pregnancy wstage. Am J Hum Genet 1990; 47 : 656-663.

              46.    Lohler J, Timpl R, Jaenisch R : Embryonic lethal mutation in mouse collagen alpha-1 (I) gene causes rupture of blood vessels and is associated with erythropoietic and mesochipnat cell death. Cell 1984; 38 : 597.

              47.    The PT, Byrd JR, McDonough PG : Etiology and subsequent reproductive performance of 100 couples with recurrent abortion. Fertil Steril 1979; 32 : 389.

              48.    Botella-Llusa J : The endometrium in repeatede abortion. Int J Fertil 1962; 7 : 147.

              49.    Balasch J, Creus M, Marquez M, et al : The significance of luteal phase deficiency on fertility : A diagnostic and therapeutic approach. Hum Reprod 1986; 1 : 145.

              50.    Hernandez Horta JL, Gordillo-Fernandez J, Soto de Leon B, et al : Direct evidence of luteal insufficiency in women with habitual abortion. Obstet Gynecol 1977; 49 : 705-708.

              51.    Filicori M, Butler JP, Crowley WF : Neuroendocrine regulation of the corpus luteum in the human. J Clin Invest 1984; 73 : 1638.

              52.    Yuen BH, Livinston JE, Poland BT, et al : Human chorionic gonadotropin, estradiol, progesterone, prolactin and B-Scan ultrasound monitoring of complications of early pregnancy. Obstet Gynecol 1981; 57 : 207-214.

              53.    Mills JL, Simpson JL, Driscoll SG, et al : Incidence of spontaneous abortion among normal women and insulin-dependent diabetic women whose pregnancies were identified within 21 days of conception. N Engl J Med 1988; 319 : 1617-1623.

              54.    Badarau L, Gavrilita L, Iuschevici C : A microscopic study of placenta and uteroplacental membrances in repeated abortion due to hyperandrogenism. Rev Fr Gynecol Obstel 1972; 67 : 685-697.

              55.    Sagle M, Bishop K, Ridley N, et al : Recurrent early miscarriage and polycystic ovary. Br Med J 1988; 297 : 1027-1028.

              56.    Regan L, Owen EJ, Jacobs HS : Hypersecretion of Luteinizing hormons, infertility, and miscarriage. Lancet 1990; 336 : 1141.

              57.    Johnson P, Pearce JM : Recurrent spontaneous abortion and polycystic ovary disease. Comparison of two regimens to induce ovulation, br Med J 1990; 300 : 154-156.

              58.    Sompolinsky D, Soomon F, Elkinal L, et al : Infections with mycoplasma and bacteria in induced midtrimester abortion and fetal loss. Am J Obstet Gynaecol 1975; 121 : 610-616.

              59.    Stray-Pedersen B, Eng J, Reikvan TM : Uterine T-mycoplasma colonization in reproductive failure. Am J Obstet Gynaecol 1978; 130 : 307.

              60.    Qinn PA, Chewchuck AB, Suber J, et al : Efficacy of antibiotic therapy in preventing spontaneous pregnancy loss among couples colonized with genital mycopleasma. Am J Obstet Gynaecol 1983; 145 : 239-244.

              61.    Schenker JG, Magalioth EJ : Intrauterine adhesions : An updated appraisal. Fertil steril 1982; 37 : 593-610.

              62.    Golan A, Langer RM, Wexler B, et al : Cervical cerclage : Its role in the pregnant anomalous uterus. Int J Fertil 1990; 35 : 164.

              63.    Kaufman RH, Noller K, Adam E, et al : Upper genital tact abnormalities and pregnancy outcome in diethylstilbestrol exposed progeny. Am J Obstet Gynaecol 1984; 148 : 973.

              64.    Pedersen JF, Mantoni M :Large intrauterine haematomata in threatened miscarriage : Frequency and clinical consequences. Br J Obstet Gynaecol 1990; 97 : 75.

              65.    Billington WD, Bell SC 1983. Immunology of mouse trophoblast. In : Loke YW, Whyte A (eds) Biology of trophoblast. Elsevier, Amsterdam, p 571-595.

              66.    Warner CM, Brownell MS, Ewoldsen MA 1988. Why aren’t embryos immunologically rejected by their mothers ? Biology of Reproduction 38 : 17-29.

              67.    Morcos RN, Gibbons WE, Findley WE 1985 Effect of peritoneal fluid on in vitro clearage of 2-cell mouse embryos : possible role in infertility associated with endometriosis. Fertility and Sterility 44 : 678-683.

              68.    Halme J, Becker S, Haskill S 1987. Altered maturation and function of peritoneal macrophages : possible role in pathogenesis of endometriosis. American Journal of Obstetrics and Gynaecology 156 : 783-788.

              69.    Hahn DW, Carraher RP, Folesy RG, McGuire JK 1986. Experimental evidence for failure to implant as a mechanism of infertility associated with endometriosis. American Joural of Obstetrics and Gynaecology 155 : 1109-1113.

              70.    Rhagupathy R, Singh B, Barrington-Leigh J, Wegmann TG 1981. Ontogeny and turnover Kinetics of paternal H-2K antigenic determinants on the allogeneic murine placenta. Journal of Immunology127 : 2074-2079.

              71.    Rugh R 1968 The mouse. Its reproduction and development. Burgess Publishing Co, Minneapolis, Minnesota, USA.

              72.    Butterworth BH, Khong TY, Loke YW, Robertson WB 1985. Human cytotrophoblast populations studied by monuclonal antibodies using single and double biotinavidin peroxidase immunocy to chemistry 33 : 977-983.

              73.    Szulman AE 1988. The biology of trophoblast disease : Complete and partial hydatidiform males. In : Beard RW, Sharp F (eds) Early pregnancy failure, mechanism and treatment. Peacock Press, England, p 309-316.

              74.    Sasagawa IM, Ohmomo V, Kanazawa K, Takeuchi S 1987. HLA expression by trophoblast of invasive moles. Placenta 8 : 111-118.

              75.    Bulmer JN, Jolmson PM 1985. Antigen expression by trophoblast populations in the human placenta and their possible immunological relevance. Placenta 6 : 127-140.

              76.    Head JR, Drake BL, Zuckermann FA 1987. Major histocompatibility antigens on trophoblast and their regulation : implications in the maternal-fetal relationship. American Journal of Reproductive Immunology and Microbiology 15 : 12-18.

              77.    Drake BL, Head JR 1988. Murine trophoblast cells are susceptible to lymphokine-activated Killer (LAK) cell lysis. American Journal of Reproductive Immunology and Microbiology 16 : 114.

              78.    Clark DA, Croy BA, Rossant J, Chaouat G 1986a Immune presensitization and local intrauterine defence as determinants of success or failure of murine interspecies pregnancy. Journal of Reproduction and Fertility 77 : 633-643.

              79.    Clark DA, Chaput A, Tutton D 1986b. Active suppression of host-versus-graft reaction in pregnant mice. VII. spontaneous abortion of allogeneic DBA/S X CBA/J fetu;ses in the uterus of CBA/J mice correlates with deficient non-T Suppressor cell activity. Journal of Immunology 136 : 1668-1675.

              80.    Kornbluth RS, Edginton TS 1986. Tumor necrosis faitor production by human monocytes is a regulated event : induction of TNF-alpha-mediated cellular cytotoxicity by endotoxin. Journal of Immunology 137 : 2585-2591.

              81.    Brieley J, Clark DA 1987. Characterization of hormone dependent suppressor cells in the uterus in pregnant and pseudopregnant mice. Journal of Reproductive Immunology 10 : 201-218.

              82.    Daya S, Clark DA, Devlin C, Jarrell J, Chaput A 1985. Preliminary characterization of two types of suppressor cells in the human uterus. Fertility and sterility 44 : 778-785.

              83.    Redline RW, Shea CM, Pappioannou VE, Lu CY 1998. Defective anti-listerial responses in deciduoma of pseudopregnant mice. American Journal of Pathology 133 : 485-497.

              84.    Clark DA, Banuratt DK, Manuel J, Fulop G, Croy BA 1989b SCID mice in reproductive biology. In : Phillips RA (ed) contemporary topics in immunobiology. Springer, Heidelber (inpress).

              85.    Slapsys RM, Younglai E, Clark DA 1988. A novel suppressor cell is recruited to decidua by fetal trophoblast-type cells. Regional Immunology I : 182-189.

              86.    Clark DA, Falbo M, Rowley RB, Banwatt D, Stedronska-clark J 1988. Active suppression of host-versus-graft reaction in pregnant mice. IX. Soluble suppressor activity obtained from allopregnant mouse decidua that blocks the response to interleukin 2 is related to transforming growth factor beta. Journal of Immunology 141 : 3833-3840.

              87.    Tsmunawaki S, Sporn M, Ding A, Nathan C 1988. Deactivation of macrophages by transforming growth factor-beta. Nature 34 : 260-262.

              88.    Ranges GE, Figari IS, Espevik T, Palladino MA Jr 1981. Inhibition of cytotoxic T cell development by transforming growth factor beta and reversal by recombinant tumor necrosis factor alpha. Journal of Experimental Medicine 166 : 991-998.

              89.    Bartocii A, Pollard JW, Stanley ER 1986. Regulation of colony stimulating factor 1 during pregnancy. Journal of Experimental Medicine 164 : 956-961.

              90.    Athnasakis I, Bleackley RC, Paetkau V, Guilbert L, Barr PJ, Wegmann TG 1987. The immunostimulatory effect of T cells and T cell lymphokines on murine fetally derived placental cells. Journal of Immunology138 : 37-44.

              91.    Thornes BJ, Mermod JJ, Vassalli P 1987. Phagocytosis and inflammatory stimuli induce GM-CSF MRNA in macrophages through post-transcriptional regulation. Cell 48 : 671.

              92.    Zucali JR, Dinarello CA, Oblon DJ, Gross MA, Andersson L, Weiner RW 1986. Interleukin 1 stimulates fibrobcasts to produce granulocyte-macrophage colony-stimulating activity and prostaglandin E2. Journal of Clinical Investigation 77 : 1857.

              93.    Thomas ML, Harger JH, Wagener DK, Robin BS, Gill TJ 1985. HLA sharing and spontaneous abortion in humens. American Journal of Obstetrics and Gynaecology 8 : 1053-1058.

              94.    Taylor C, Faulk WP 1981. Prevention of recurrent abortion with Leukocyte transfusions. Lancet ii : 68-70.

              95.    Rokline RE, Kitzmiller JL, Carpenter CB, Garvoy MR, David JR 1976. Maternal fetal relation. Absence of an immunological blocking factor from the sera of women with chronic abortions. New England Journal of Medicine 295 : 1209-1213.

              96.    Fizet D, Bousqauet J 1983. Absence of a factor blocking cellular cytotoxicity reaction in serum of women with recurrent abortion. British Journal of Obstetrics and Gynaecology 90 : 453-456.

              97.    P ower DA, Mather DA, MacLeod AM, Lind T, Catto GRD 1986 maternal antibodies to paternal B-lymphocytes in normal and abnormal pregnancy. Americal Journal of Reproductive Immunology and Microbiology 10 : 10-13.

              98.    Takukawa K, Kamazawa K, Takeuchi S 1986. Production of blocking antibodies by vaccination with husband’s lymphocytes in unexplained recurrent aborters : the role in successful pregnancy. American Journal of Reproductive Immunology and Microbiology 10: 1-9.

              99.    Beer AE 1988. Pregnancy outcome in couples with recurrent abortions following immunological evaluation and therapy. In : Beard RW, Sharp F (eds) Early Pregnancy loss, mechanisms and treatment. Peacock Press, England, p 227-354.

            100.    Unander M 1988 Discussion. In : Beard RW, Sharp F (eds) Early pregnancy failure, mechanisms and treatment. Peacock Press, England, p 400-403.

            101.    Singal DP, Butter L, Liao SK, Joseph S 1984. The fetus as an allograft evidence for anti-ideotypic antibodies induced by pregnancy. American Journal of Reproductive Immunology 6 : 145-151.

            102.    Bonaqura VR, Ma A, McDowell J, Lewison A, King DW, Suciu-Foca N 1987. Anti-clonotypic antoantibodies in pregnancy. Cellular Immunology 108 : 356-365.

            103.    Davies M, Browne CM 1985. Anti-trophoblast antibody responses during normal human pregnancy. Journal of Reproductive Immunology. 7 : 285-297.

            104.    Clark DA, Kiger N, Guenet J-L, Chaouat G 1987. Local active suppression and successxful vaccination against spontaneous abortion in CBA/J mice. Journal of Reproductive Immunology 10 : 79-85.

            105.    Chaouat G, Kolb JP, Kiger N, Stanis Lawski M, Wegmann TG 1985. Immunologic consequences of vaccination against abortion in mice. Journal of Immunology 134 : 1594-1598.

            106.    McIntyre JA, McComnnachie PR, Taylor C, Faulk WE 1984. Clinical, immunology and genetic definitions of primary and recondacy recurrent spontaneous abortion. Fertility and sterility 42 : 844-855.

            107.    McIntyre JA, Faulk WP, Nichols-Jolmson VR, Taylor CG 1985. Immunological testing and immunotherapy in recurrent spontaneous abortion. Obstetrics and Gynaecology 67 : 169-175.

            108.    McConnachie PR, McIntyre JA 1984. Maternal antipaterual immunity in couples predisposed to repeat pregnancy losses. Americal Journal of Reproductive Immunology 5 : 145-150.

            109.    Harris EW 1988. Clinical and immunological significance of anti-phospholipid antibodies. In : Beard RW, Sharp F (eds) Early pregnancy failure, mechanisms and treatment. Peacock Press, England, p 43-65.

            110.    Howard MA, Firkin BG, Healy DL, Choong S-CC 1987. Lupus anticoagulant in women with multiple spontaneous miscarriage. American Journal of Hematology 26 : 175-178.

            111.    Lepez M, Cartron J, Cartron JP et al 1983. Cytotoxicity of anti PP, PK antibodies and possible relationship with early abortions of P mothers. Clinical Immunology and Immunopathology 28 : 296-303.

            112.    Gleicher N 1987. A potential animal model for autoimmune reproductive failure. Americal Journal of Reproductive Immunology and Microbiology 14 : 122.

            113.    Mowbray JF 1988. Success and failures of immunisation for recurrent spontaneous abortion.In : Beard RW, Sharp F (eds) Early pregnancy failure, mechanisms and treatment. Peacock Press, England, p 325-336.     

            114.    Gill TJ 1987. Genetic factors in fetal loss. American Journal of Reproductive Immunology and Microbiology 15 : 133-137.

            115.    Johnson PM, Chia KV, Hart CA, Griffith HB, Frances W JA 1988. Trophoblast membrane infusion for unexplained recurrent miscarriage. British Journal of Obstetrics and Gynaecology 95 : 111-16.

            116.    Harrison RF 1988. Early recurrent pregnancy failure : treatment with human chorionic gonadotrophin. In : Beard RW, Sharp F (eds) Early pregnancy failure, mechanisms and treatment. Peacock Press, England, p 421-431.

            117.    Potter MF 1938. Foetal mortality in the offsring of apparently healthy women. Journal of Obstetrics and Gynaecology of the British Empire 45 : 233-249.

            118.    Shearman RP, Garrett WJ 1963. Double blind study of the effect of 17-hydraoxy-progesterone caproate on the abortion rate. British Medical Journal i : 292-295.

            119.    Stray-Pedersen B, Stray-Pedersen S 1988. Recurrent abortion : the role of psychotherapy. In : Beard RW, Sharp F (eds) Early pregnancy failure, mechanisms and treatment. Peacock Press, England, p 433-442.

            120.    Edmonds DK 1988. Use of cervical cerclage in patients with recurrent first trimester abortion. In : Beard RW, Sharp F (eds) Early pregnancy failure, mechanisms and treatment. Peacock Press, England, p 411-419.

            121.    Harris EN Asherson RA, Hughes GRV 1998 Antiphospholipid antibodies : autoantibodies with a difference. Annu Rev Med 39 : 261-271.

            122.    Branch WD, Scott JR, Kochenour NK et al 1985 Obstetric complications associated with the lupus anticoagulant N. Engl J Med 313 : 1322-1326.

            123.    Masala C, Sarice M Di prima MA et al 1989 Evidence for shared epitopes between cardiolipin and Pneumocystis carinii. Infect Dis 160 : 736-737.

            124.    Schulman LE 1987 Systemic Lupus erythematosus and the chronic biologic false positive test of syphilis. In : Wallace DJ, Dubois EL (eds) Dubois’ Lupus erythematosus, 3rd Edn. Lea & Febiger, Philadelphia.

            125.    Pengo V, Thiagarajan P Shapiro SS, Heine MJ 1987 Immunological specificity and mechanism of action of IgG lupus anticoagulants. Blood 70 : 69-76.

            126.    Triplett DA, Brandt JT 1989. Laboratory identification of the lupus anticoagulant, Br J Haematol 73 : 139-142.

            127.    Harris EN, Gharavi AE, Boey ML et al 1983 Anticardiolipin antibodies : detection by radioimmuno assay and association with thrombosis in systemic lupus erythematosus. Lancet ii : 1211-1214.

            128.    Pattison NS, McKay EJ Liggins GC, Lubbe WE 1987 Anticardiolipin antibodies : their presence as a marker for lupus anticoagulant in pregnancy. NZ Med J 100 : 61-64.

            129.    Hecht ER 1965 Lipids in blood clotting. Charles C Thomas, Springfield, IL.

            130.    Galli M, Comfurius P, Maassen C et al 1990a Anti cardio lipin antibodies (aCL) directed not to cardio lipin but to a plasma protein cofactor. Lancet 335 : 1544-1547.

            131.    McNeil HP, Simpson RJ, Chesterman CN, Krilis SA 1990 Antiphospholipid antibodies are directed against a complex antigen that includes a lipid-binding inhibition of coagulation : B2 glycoprotein (apolipoprotein H). Proc Natl Acad Sci USA 87 : 4120-4124.

            132.    Matsuura E, Igarashi Y, Fujimoto M et al 1990 Anticardiolipin and differential diagnosis of autoimmune disease. Lancet 336 : 177-178.

            133.    Chamely LW, McKay EJ, Pattison NS 1991a. Cofactor dependent and cofactor independent anticardiolipin antibodies. Thromb Res 61 : 291-299.

            134.    Wurm H 1984, B2 glycoprotein 1 (apolipoprotein H) interactions with phospholipid vesicles-Int J Biochem 16 : 511-515.

            135.    SK Housboe I 1985 B2 glycoprotein 1 : a plasma inhibitor of the contact activation of the intrinsic blood coagulation pathway. Blood 66 : 1086-1091.

            136.    Nimpf J, Wurm H, Kostner GM 1987 B2 glycoprotein 1 (apo-H) inhibits the release reaction of human platelets during ADP-induced aggregation. Atherosclerosis 63 : 109-114.

            137.    Barbui T, Cortelazzo S, Galli M et al 1998 Antiphospholipid antibodies in early repeated abortions : a case controlled study. Fertil steril 50 : 589-597.

            138.    Unander AM, Norbert R, Hahn L, Arfors L 1987 Anticardiolipin antibodies and complement in niety-ninety women with habitual abortion. Am J. Obstet Gynecol. 156 : 114-119.

            139.    Parke AL Wilson D, Bayes-Maierd 1991. The prevalence of antiphospholipid antibodies in women with recurrent spontaneous abortion, women with successful pregnancies and women who have never been pregnant. Arthritis Rheum 34 : 1231-1235.

            140.    Parazzini F, Ackia D, Faden D et al 1991. Antiphospholipid antibodies and reccurent abortion. Obstet Gynecol. 77 : 854.

            141.    Lockshin MD, Quamar T, Druzin M, Goei S 1987. Antibody to cardiolipin, cupus anticoagulant and fetal death. J Rheumatol 14 : 259-262.

            142.    Harris EN, Gharavi AE, Hedge U et al 1985. Anticardiolipin antibodies in antoimmune thrombocytopenic purpura. Br J Haematol 59 : 231-234.

            143.    Gerrard JM, Friesen LL, 1985 Platelets. In Poller L(ed) Recent advances in blood coagulation vol 4. Churchill Livingstone, Edinburgh, pp 139-168.

            144.    Carreras LO, Machin SJ, Deman R. et al 1981 Arterial thrombosis, intrauterine death and lupus anticoagulant : detection of immunoglobulin interfering with prostacyclin formation. Lancet i : 224-226.

            145.    Kwaan HC 1989 Protein C and protein S Semin Thromb Hemost 15 : 353-355.

            146.    Freyssinet J-M, Wiesel ML, Gauchy J et at 1986 An IgM Lupus anticoagulant that neutraulises the enhancing effect of phospholipid on purified endothelial thrombomodulin activity : A mechanism for thrombosis. Thromb Haemost 55 : 309-313.

            147.    Tsakiris DA, Settas L, Makris PE, Marbet GA 1990 Lupus anticoagulant, antiphospholipid antibodies and thrombophilia : relation to protein c-protein S-thrombomodulin J. Rheumatol 17 : 785-789.

            148.    Comp PC 1990 Laboratory evaluation of protein S status. Semin Thromb Hemos 16 : 177-181.

            149.    Andersson LO (1979) Physiochemical properties of antithrombin III, In : Collen D, Wiman B, Verstraete M. (eds) The physiological inhibitors of coagulation and fibrinolysis. Elsevier/North Holland Biomedical Press, Amsterdam, pp 39-42.

            150.    Marcum JA, Reilly CF, Rosenberg RD 1986. The role of specific forms of heparan sulphate in regulating blood vessel wall function. Prog. Hemost Throm 8 : 185-215.

            151.    Schulze-Lohoft E, Kraft F, Bleil L et al 1989 IgM-containing immune complexes and antiphospholipid antibodies in patients with Snedden’s syndrome. Rheum Int 9 : 43-48.

            152.    Birdsall MA, Pattison NS, Chamley LW 1992 Antiphospholipid antibodies in pregnancy. Aust N 2 J obstel Gynaecol 32 : 328-330.

            153.    Garcia-de la Torre, Hernandez-Vasquez L, Angulo-Vasquez J, et al : Prevalence of antinuclear antibodies in patients with habitual abortion and in normal and toxemic pregnancies. Rheumatol Int 1984; 4 : 87-89.

            154.    Xu L, Chang V, Murphy A, et al : Antinuclears antibodies in sera of patients with recurrent pregnancy wastage. Am J Obstet Gynaecol 1990; 163 : 1493-1497.

            155.    Cowchock S, Smith B, Gociat B : Antibodies to phospholipids and nuclear antigens in patients with repeated abortions. Am J Obstet Gynaecol 1986; 155 : 1002-1010.

            156.    Simpson JL, Mitts JL, Holmes LB, et al : Low fetal loss rates after ultrasound-proved viability in early pregnancy. JAMA 1987;258 : 2555-2557.

            157.    Siddiqui TA, Caligaris JT. Miodovnik M, et al : Rate of spontaneous abortion after first trimester somographic demonstration of fetal cardiac activity. Am J Perinatol 1988; 5 : 1-4.

            158.    Howert-de Jong MH, Termijtelen A, Eskes TKAB, et al : The natural course of habitual abortion. Eur J Obstet Gynaecol 1989; 33 : 221-228.

 

By Dr. Narendra Malhotra, Dr. Amita Singh, Dr. Jaideep Malhotra