Maternal Immunologic Diseases and Neonatal Disorders

After completing this article, readers should be able to: Autoimmune diseases are characterized by an immune attack on tissues of the body in the apparent absence of active infection. Symptoms arise from the resulting impairment of cell or organ function. Women are disproportionately affected by many autoimmune diseases, especially those conditions whose typical age of onset is in the early childbearing years. (1) For women who have autoimmune diseases, the physiologic changes of pregnancy often have consequences for the course of the autoimmune disease. (2) In turn, autoimmune disease or its treatment may affect the developing fetus or the newborn.Autoimmune diseases can be classified into two broad categories, based on the primary immunologic effector mechanism that causes disease. In one class of autoimmune disease, CD4+ T cells are believed to be the major effector cells, although they enlist other arms of the immune response, such as CD8+ T cells or macrophages, during the disease course. Examples of this class of autoimmune disease include insulin-dependent diabetes mellitus, rheumatoid arthritis, and multiple sclerosis. The other class of autoimmune diseases is caused by antibodies that are directed against self-antigens. (It should be noted that production of these antibodies is generally T-cell-dependent). Diseases in this class include myasthenia gravis, Goodpasture syndrome, Graves’ disease, antiphospholipid antibody syndrome (APS), immune thrombocytopenic purpura (ITP), and systemic lupus erythematosus (SLE). The autoantibody-mediated diseases can have direct consequences on the fetus and neonate because the antibodies are usually of the immunoglobulin G (IgG) type, which are transported across the placenta to the fetal circulation. In contrast, pathogenic T cells do not effectively transfer maternal disease transplacentally, although recent studies indicate that cells do traffic between the fetus and mother during pregnancy (see accompanying article in this issue). The latter observation has led to the hypothesis that microchimerism established after cellular exchange between healthy individuals may be involved in the development of some autoimmune diseases later in life (reviewed in this issue of NeoReviews and in Nelson (3)).This article addresses three antibody-mediated disorders that have significant consequences for the fetus or newborn: APS, ITP, and SLE. A fourth example, fetal and neonatal hyperthyroidism resulting from maternal autoimmune thyroid disease and transplacental passage of thyroid-stimulating immunoglobulins, has been reviewed recently in NeoReviews. (4) The fetal or neonatal manifestations of these maternal disorders are distinct from one another, reflecting both the different tissue targets of the antibodies and the apparently differing developmental “windows of susceptibility.” Finally, it should be emphasized that among mothers who have these disorders, the incidence of affected children is low. Thus, an emerging area of interest is the identification of other factors that contribute to these disorders beyond the presence of maternal autoantibodies.APS is an autoimmune disease associated with thrombophilia and recurrent pregnancy loss. (5) Although antiphospholipid antibodies and APS are seen in association with SLE and other rheumatic diseases, primary APS without other disease association occurs, and antiphospholipid antibodies can be found in 2% to 5% of healthy children and adults who have no pathologic associations. (6)(7) In addition to pregnancy loss, other pregnancy complications among women who have APS include fetal growth impairment, placental insufficiency, preeclampsia, and preterm birth. These obstetric complications apparently arise from the prothrombotic effects of maternal antiphospholipid antibodies on placental function. Vasculopathy, infarction, and thrombosis have been identified in placentas from women who had failed pregnancies and APS (reviewed in Esplin (8)). Pregnancy loss and other obstetric complications of APS are likely due to the procoagulant activity of antiphospholipid antibodies in inducing adhesion molecules, platelet activation, and aggregation factors and the inhibition of key anticoagulant factors including protein C and placental anticoagulant protein I (annexin V). (8)(9)Criteria for the diagnosis of APS were proposed at the International Antiphospholipid Symposium in Japan in 1999 (5) and include both obstetric and nonobstetric features and laboratory criteria. The obstetric criteria identify three types of pregnancy loss: 1) one or more unexplained deaths of a morphologically normal fetus at or beyond the 10th week of gestation; 2) one or more preterm births at or before the 34th gestational week as a consequence of severe preeclampsia or placental insufficiency; or 3) three or more consecutive spontaneous abortions before the 10th gestational week, with exclusion of other causes. Recurrent pregnancy loss occurs in 1% of women, and 55% of cases are associated with defects in hemostasis leading to thrombophilia. (10) APS was identified in 67% of women who had recurrent pregnancy loss due to a hemostasis defect. The rates of serious obstetric complications other than pregnancy loss, including preeclampsia, placental insufficiency, fetal growth retardation, and preterm birth, vary among studies, probably because of selection criteria (reviewed in Geis and Branch (11)). Preeclampsia was reported in 18% to 48% of patients who had APS and placental insufficiency in 30%. Prospective studies generally have supported these observations, although some have shown that no more than 10% of women who had APS developed preeclampsia, (12) and growth retardation ranged from 12% in APS pregnancies (normal without APS was 2%) (13) to no significant growth impairment. (14) These complications continue to occur even when pregnancy loss is prevented with low-dose heparin and aspirin. (15)The highest risk of APS with recurrent pregnancy loss has been associated with moderate-to-high levels of IgG anticardiolipin antibodies (>20 GPL) and lupus anticoagulant. (5) Moderate-to-high levels of IgM anticardiolipin antibodies also may confer risk, but the presence of anti-beta 2 glycoprotein 1 and other antiphospholipid antibodies (eg, antiphosphatidylserine, antiphosphatidylethanoloamine, and antiphosphatidylcholine) have not been shown to increase the predictive value above that seen with the presence of anticardiolipin antibodies or lupus anticoagulant.Treatment trials have focused on immunomodulation (eg, corticosteroids and intravenous immune globulin [IVIG]) and anticoagulation (eg, low-dose aspirin, heparin, low-molecular weight heparin). Although corticosteroids in conjunction with low-dose aspirin were 60% to 70% effective in preventing pregnancy loss, the side effects from the significant doses of prednisone required were prohibitive. Heparin with or without low-dose aspirin has been shown to be more effective than low-dose aspirin alone in prospective randomized studies. (12)(16) In the study by Kutteh, (16) the percentage of live births was significantly greater for patients treated with heparin (80%) compared with low-dose aspirin (44%). The doses of heparin ranged from thromboprophylactic doses (5,000 U twice daily) to higher doses (10,000 to 20,000 U twice daily). The use of low-molecular weight heparin is safe in pregnancy and appears to be promising in the treatment of pregnancy loss due to APS. (17) However, although live birth rates are improved with combination heparin and low-dose aspirin therapy, other obstetric complications continue to be a problem, including preeclampsia, preterm birth, and growth impairment. (15) For this reason, there has been an interest in adding monthly infusions of IVIG to heparin and low-dose aspirin therapy. Pilot studies suggested a tendency toward improved outcomes, although results were not statistically significant. (18) There have been rare cases of thrombosis in neonates, presumably due to maternal passage of antiphospholipid antibodies, but because this complication is so rare, postpartum anticoagulation of neonates is not recommended. In contrast, mothers are at increased risk for thrombosis postpartum, and maternal anticoagulation should be continued for up to 6 weeks after delivery. (8)The clinically important causes of maternal thrombocytopenia in pregnancy are gestational thrombocytopenia and autoimmune thrombocytopenia due to ITP. The healthy-appearing mother who has mild thrombocytopenia may have gestational benign thrombocytopenia, in which the platelet count usually does not decrease below 50 × 103/mcL (50 × 109/L) and does not produce a decrease in the platelet count of the fetus. In contrast, immune-mediated thrombocytopenia can result in significant fetal thrombocytopenia. Gestational benign thrombocytopenia is present in 70% to 80% of pregnancy-related thrombocytopenias. It is discovered initially during pregnancy, and in these patients, results of a reliable test for antiplatelet antibody are usually negative. Conversely, patients who have immune-mediated thrombocytopenia may have a history of thrombocytopenia before the pregnancy, and they usually have detectable antiplatelet antibody.ITP is a relatively common autoimmune disorder among women of childbearing age. Its frequency is approximately 1 to 2 per 1,000 live births, accounting for about 3% of all cases of maternal thrombocytopenia at delivery. ITP in pregnancy requires close management of both the mother and the infant. Obstetric management is aimed at reducing the risks of life-threatening maternal hemorrhage at the time of delivery, and fetal management is directed toward an appropriate strategy for the least traumatic delivery. A mother whose platelet count is less than 50 × 103/mcL (50×109/L) generally is delivered by cesarean section; those whose counts are greater than 50 × 103/mcL (50×109/L) are allowed to proceed with vaginal delivery if no obstetric contraindications exist. The use of IVIG therapy during pregnancy may improve platelet counts in the mother who is experiencing severe hemorrhage. However, IVIG is not appropriate treatment for preventing fetal thrombocytopenia because exogenous transport of IVIG across the placenta is inconsistent and unpredictable. (19) Recent studies have demonstrated that the lack of effect of maternally administered IVIG on fetal platelet counts may be attributed to the insufficient therapeutic level of transferred IgG in the cord blood, suggesting that higher doses in the mother may be effective for the fetus. (20)Autoimmune thrombocytopenic purpura in pregnant women can induce moderate or severe thrombocytopenia in the fetus or the newborn, and thrombocytopenia can occur as early as 20 weeks of gestation. Fortunately, the risk of complications arising from neonatal thrombocytopenia is low, but careful observation is required for the thrombocytopenic newborn of mothers who have ITP, even when the infant has no bleeding complications at delivery. (21)The frequency of intracranial hemorrhage has been estimated to be about 1% and is less frequent in autoimmune thrombocytopenia than in neonatal alloimmune thrombocytopenia (∼10%). The degree of maternal thrombocytopenia does not correlate well with the fetal or newborn platelet counts, and treatment of mothers with high-dose intravenous gamma globulin and prednisolone does not prevent neonatal thrombocytopenia. Indeed, some neonates may have significant thrombocytopenia even though their mothers apparently respond well to steroids or IVIG during the last trimester.No significant correlation has been observed between neonatal thrombocytopenia and maternal platelet autoantibodies. The history of a previous infant who had thrombocytopenia is the only important factor in estimating the risk of fetal thrombocytopenia. After birth, thrombocytopenia in the neonate usually worsens during the first days of life. Postnatal management typically involves observation when the platelet count is greater than 20 × 103/mcL (20×109/L) in a child who exhibits no clinical bleeding. For infants who have evidence of hemorrhage, single-donor irradiated platelets may be administered to control bleeding, even though the platelet count may not show a sustained increase. In addition, the infant may benefit from an infusion of IVIG. (22)(23)(24) The neonatal thrombocytopenia usually resolves within 4 to 6 weeks, after the passively acquired maternal autoantibodies dissipate. Maternal thrombocytopenia may persist, and the mother should be monitored closely until collagen vascular disease is excluded and the platelet count has returned to normal.NLE is a model of passively acquired autoimmunity in which immune abnormalities in the mother lead to the production of anti-SSA/Ro-SSB/La antibodies that cross the placenta and presumably injure fetal tissue. (25)(26) The most serious manifestation is damage to the cardiac conducting system that results in congenital heart block (CHB), which is usually third degree, although less advanced blocks have been observed. CHB typically is identified between 16 and 24 weeks of gestation. The mortality rate is approximately 20%, and the majority of surviving children require pacing. Cutaneous involvement (erythematous and often annular lesions with a predilection for the eyes, face, and scalp, frequently photosensitive) and, to a lesser extent, hepatic and hematologic involvement are also associated with maternal anti-SSA/Ro-SSB/La antibodies and are grouped under the heading of neonatal lupus syndromes. NLE—so termed because the dermatologic lesions of the neonate resemble those seen in SLE—is a misnomer because fewer than one third of mothers of affected children have SLE (many are asymptomatic), and the neonatal disease frequently is manifested only as heart block, a problem rarely reported in adults who have SLE. To date, complete block is irreversible. In contrast, the noncardiac manifestations are transient, resolving at about 6 months of life coincident with the disappearance of maternal autoantibodies from the neonatal circulation.A molecular definition for the target autoantigens, 52kDa and 60kDa SSA/Ro and 48kDa SSB/La, has been provided by the isolation of cDNA clones. (27)(28)(29)(30)(31) However, the pathogenesis of antibody-mediated insult remains largely unknown. The problem lies in finding a reasonable explanation for accessibility of these intracellular antigens to extracellular maternal antibodies. Evidence is emerging to support a role for physiologic apoptosis as a mechanism of translocating the antigens to the cell surface where they can be bound by cognate maternal antibodies and inadvertently program an inflammatory response by the macrophages. (32)(33) Additionally, several laboratories have demonstrated that anti-SSA/Ro antibodies are arrhythmogenic and inhibit inward calcium fluxes across cell membranes. (34)(35) Mice actively immunized with SSA/Ro antigens have given birth to pups that have varying conduction disturbances. (36)The nearly universal finding of anti-SSA/Ro-SSB/La antibodies in the serum of mothers (who may be totally asymptomatic or have a defined rheumatic disease such as SLE or Sjögren syndrome [SS]) whose fetuses are diagnosed in utero as having CHB is astonishing. Viewed from the converse perspective, the incidence of CHB in mothers known to have the candidate antibodies is low. Based on a recently published prospective study by Brucato and colleagues, (37) the frequency of third-degree block in an offspring of a mother who has anti-SSA/Ro antibodies is estimated at 1% to 2%. Whether the presence of anti-SSB/La antibodies confers an increased risk is unknown. Additionally, less advanced degrees of block may be more prevalent than appreciated. The risk of cutaneous lesions is unknown.Given the rarity of NLE, The Research Registry for Neonatal Lupus was established by the National Institute for Arthritis, Musculoskeletal and Skin Diseases in September 1994 and has been extensively described. (26) In contrast to SLE, a disease for which criteria have been established by the American College of Rheumatology, (38) no formal classification criteria yet exist for NLE. However, for the purposes of the Registry, a fetus, neonate, or infant is considered to have NLE if the following two criteria are met: 1) heart block or characteristic rash is diagnosed, and 2) maternal antibodies to the 52kDa SSA/Ro, 60kDa SSA/Ro, or 48kDa SSB/La ribonucleoproteins (or U1 RNP in cases of rash only) are identified. The Registry functions as a resource for basic researchers focused on the pathogenesis of disease and clinicians caring for patients.No serologic profile is unique to mothers of affected children, but compared with mothers of healthy children, anti-SSA/Ro antibodies are usually of high titer (frequently anti-52kDa SSA/Ro-positive by immunoblot) and associated with anti-SSB/La antibodies. (25)(39) There are reports of discordant dizygotic and monozygotic twins and low recurrence rates of CHB and skin disease. In the Registry, 14 of 80 next pregnancies following the birth of a child who had CHB resulted in CHB, three of whom had both CHB and rash and seven of whom had rash alone; 7 of 31 next pregnancies following the birth of a child who had NLE rash resulted in CHB, four of whom had both CHB and rash, and 12 of whom had rash alone. These findings indicate that factors (likely fetal) in addition to anti-SSA/Ro and SSB/La antibodies contribute to the development of NLE.Despite the absence of formal proof, clinical anecdotes support the hypothesis that injury to the AV node progresses through stages, with the final outcome being fibrosis of the node and irreversible third-degree block. One review of medical charts of 187 children who had CHB revealed that 16 (8.5%) had conduction abnormalities other than third-degree heart block either in utero or at birth (this is not meant to represent a true frequency because these were referred cases or siblings of children who had third-degree block). Nine had a prolonged PR interval on electrocardiography at birth or in the perinatal period. The degree of block progressed in four children. Importantly, this progression occurred subsequent to the clearance of maternal antibodies from the infant’s circulation. Accordingly, electrocardiography should be performed on all neonates born to mothers who have anti-SSA/Ro-SSB/La antibodies.Prophylactic prednisone, dexamethasone, or plasmapheresis therapy for mothers at risk (highest among those who birth to a child who had manifestation of is not at Maternal prednisone least in low and moderate early in pregnancy does not prevent the development of is not because prednisone administered to the mother is not active in the fetus, and levels of antibodies relatively during therapy. Although be the for fewer than of mothers at highest risk have an affected the in utero of block was not However, the of the PR interval on the can be by A prospective National of PR and in CHB, is to the PR interval in pregnant women who have anti-SSA/Ro antibodies, anti-SSB/La antibodies, or One of the is to identify the of block and to it is a for more advanced of the conduction the for For patients to in a the suggested is of the fetus from 16 to the fetus in whom a block is there are several for but no the fetus has been in complete block for more than weeks and there is no of or it be to with frequent and not For recently identified or complete blocks associated with or 4 of maternal is a reasonable the is a randomized prospective treatment of the to the of maternal in diagnosed heart the on cutaneous manifestations of NLE in of the onset of disease, of and recurrence rates in subsequent medical from the Research Registry for Neonatal Lupus were The mothers whose or antibodies and their infants who were diagnosed with rash heart between and There were 20 and test not a significant in when compared with offspring who had CHB, of whom were and were In most the infant’s rash the age of was 6 weeks, and the of the rash was had involvement most by the scalp, and the children, therapy for the cutaneous only to children initially but because there was no response, they were treated with was treated with systemic The rash in all of the children. In cases for which reliable were without of which were treated and 16 no therapy of However, 14 children had skin of the affected or and after at least 2 of these 14 children, were treated and 4 were There was no significant in outcome between treated and children, as by on the outcome of children who have NLE and their siblings is the Registry was to children who had manifestations of NLE or their siblings later developed autoantibodies or rheumatic mothers in the Registry returned on children who had NLE and their children had rheumatoid and diabetes and and congenital and syndrome had manifestations of NLE, and their mothers had manifestations of autoimmune and autoimmune disease The antibody test was in 2 of affected children, serum antibodies with SSA/Ro or SSB/La These that children who have NLE require continued especially to and if the mother has an autoimmune disease. or whose mothers have anti-SSA/Ro or SSB/La antibodies do not to have an increased risk of developing systemic rheumatic diseases during and early