Fetal Effects of Autoimmune Disease

Placental Transfer: General Remarks

The maternal-fetal interface is quite efficient in its selective exclusion of substances during the transport process from the maternal to the fetal circulation. At the same time, the placenta selectively transfers other substances, a process that is facilitated by the proximity of the respective maternal-fetal vascular systems within the placental cotyledons. Although there is no mixing of the maternal and fetal blood, the placental barrier is not impermeable, and small amounts of fetal blood, including fetal cells, gain access to the maternal circulation, in most pregnancies through breaks in the fetal-maternal interface. When fetal blood cells are recognized as antigens by the maternal immunologic system, they may provoke an immune response and the production of immunoglobulins. This mechanism occurs in only a few pregnancies and is the basis of incompatibility disorders (see Chapter 24), whereby exogenous antigens, such as fetal cells or incompatible blood, sensitize the maternal immune system. The maternal antibodies, which are produced as a response to sensitization, cross the placenta and may destroy fetal cells. Generally, the mother is disease free, and the diagnosis is reached after the delivery of an affected infant or by screening tests.

A second type of maternal antibody that crosses the placenta and affects the fetus may arise from sensitization of the mother’s immune system by her endogenous antigens, with the resultant production of autoantibodies. The mother with autoantibodies has an autoimmune disorder, and the diagnosis of the maternal disease usually precedes the diagnosis of the fetal or neonatal complication.

These generalizations describe immune processes that may affect the fetus or neonate. Although the maternal immune system may produce a wide range of immunoglobulins, only maternal antibodies of the IgG class (but not IgM or IgA) can cross the placental barrier. The common denominators of such disorders are the production of IgG in the maternal compartment, the transfer of IgG through the placenta, and the effects of these antibodies in the fetal compartment or neonate. This chapter discusses examples of such disorders.

Fetal Thrombocytopenia

Box 21-1 lists the immunologic etiologies of fetal, and consequently neonatal, thrombocytopenia. The most significant pathologies are neonatal alloimmune thrombocytopenia and immune (idiopathic) thrombocytopenic purpura (ITP). Although the two conditions have some similarities, they are distinct diseases, each with a different underlying pathogenesis (Table 21-1) (see Chapter 88).

Box 21-1 Immunologic Etiology of Fetal or Neonatal Thrombocytopenia

Maternal production of autoantibodies

• Immune thrombocytopenic purpura

• Systemic lupus erythematosus

• Drug-induced thrombocytopenia

Neonatal alloimmune (isoimmune) thrombocytopenia

ABO incompatibility

TABLE 21-1

Characteristics of Neonatal Thrombocytopenia Based on Etiology

	Alloimmune Thrombocytopenia	Maternal Immune Thrombocytopenic Purpura
Cause of sensitization	Antigen on fetal platelets	Autoantibodies
Maternal platelet count	Normal	Low
Fetal platelet count	Low	Variable
Fetal risk (pregnancy)	High	Low
Fetal risk (delivery)	High	Depends on platelet count
Maternal risk	None	Depends on platelet count

Immune Thrombocytopenic Purpura

Immune thrombocytopenic purpura in adults is often a chronic disease mediated by autoantibodies directed against cell surface components (glycoproteins) of platelets (IIb/IIIa or Ib/IX). Thrombocytopenia occurs when the platelet-antibody complexes are destroyed by the reticuloendothelial system. A low platelet count raises suspicion for ITP, but the diagnosis is reached after exclusion of other causes of thrombocytopenia by history, physical examination, blood count, peripheral blood smear, and autoimmune profile.⁹ A spuriously low platelet count should be evaluated by examining a blood smear to exclude pseudothrombocytopenia caused by ethylene diaminetetra-acetic acid–dependent platelet agglutination. The normal range of platelet counts in nonpregnant women and neonates is 150,000 to 400,000/µL; however, the mean counts tend to be lower during pregnancy. The prevalence of maternal ITP is one to two cases per 1000 deliveries. The potential risk of a low platelet count for the mother is bleeding; however, the risk becomes significant only when the platelet count becomes less than 20,000/µL. A maternal platelet count of greater than 50,000/µL is considered to be hemostatic during vaginal or cesarean birth.

Thrombocytopenia of the fetus or newborn is caused by active transplacental transport of the antiplatelet antibodies; however, no significant correlation has been observed between neonatal thrombocytopenia and maternal autoimmune antibodies. A low platelet count increases the risk for hemorrhage, but this seems to be more theoretical than real because intrauterine fetal hemorrhage has not been reported in patients with ITP. The concern is for the potential trauma at birth and the potential risk for cerebral hemorrhage in the neonate. This serious complication is rare because the prevalence of fetal or neonatal ITP is about 10% that of maternal ITP, and less than half of these infants have platelet counts less than 20,000/µL. In the past, obstetricians performed antepartum cordocentesis (percutaneous umbilical vein blood sampling [PUBS]) or fetal scalp blood sampling to identify a fetus with a platelet count less than 50,000/µL and to deliver the fetus by the abdominal route. The current view holds that PUBS and fetal scalp sampling are unnecessary in pregnant women without known ITP even with platelet counts of 40,000/µL. Generally, when the maternal platelet count is greater than 50,000/µL and the fetal platelet count (or the platelet count of previous offspring) is unknown, cesarean section is not indicated; a vaginal delivery is allowed, and the cesarean option is reserved for obstetric indications. If the fetal platelet count is known to be less than 20,000/µL, cesarean section is appropriate.

Treatment of ITP during pregnancy follows the guidelines published in 1996 9,20 and mainly reaffirmed more recently.³¹ Pregnant patients with ITP and platelet counts greater than 50,000/µL throughout gestation and patients with platelet counts of 30,000 to 50,000/µL in the first or second trimester do not routinely require treatment.²⁰ Treatment in the form of glucocorticoids or intravenous immune globulin (IVIG) is indicated in patients with platelet counts less than 10,000/µL and patients with platelet counts of 10,000 to 30,000/µL who are in their second or third trimester or are bleeding. Intravenous immune globulin is an appropriate initial treatment for patients with platelet counts less than 10,000/µL in the third trimester and for patients with platelet counts of 10,000 to 30,000/µL who are bleeding. When glucocorticoid and IVIG therapy have failed, splenectomy is appropriate in the second trimester in women with platelet counts less than 10,000/µL who are bleeding. Splenectomy should not be performed in asymptomatic pregnant women with platelet counts greater than 10,000/µL.²⁰ Platelet transfusion is indicated for patients with counts less than 10,000/µL before a planned cesarean section and for those who are bleeding and expected to deliver vaginally. Prophylactic transfusions are unnecessary when the platelet count is greater than 30,000/µL and there is no bleeding.

Immune thrombocytopenic purpura does not preclude breastfeeding. The use of IVIG during pregnancy may improve platelet counts in the mother, but the treatment may not prevent fetal thrombocytopenia because the placental transfer of IVIG is inconsistent and mostly insufficient to reach fetal therapeutic levels.³ A retrospective study examined the morbidity of 92 obstetric patients with ITP during 119 pregnancies over an 11-year period.³⁵ The authors found that most of these patients had thrombocytopenia during pregnancy. At delivery, 89% had platelet counts less than 150,000/µL. For many patients, the pregnancy was uneventful; however, 21.5% of the women had moderate to severe bleeding. In 31.1% of the pregnancies, treatment was required to increase the platelet counts. Most deliveries (82.4%) were vaginal. Platelet counts of less than 150,000/µL were found in 25.2% of the infants, including 9% with platelet counts less than 50,000/µL. Treatment for hemostatic impairment was necessary in 14.6% of the infants. During the study period, two fetal deaths occurred, including one caused by hemorrhage.³⁵

After birth, the platelet count of newborns whose mothers have ITP-mediated thrombocytopenia may continue to decrease, and careful follow-up of the thrombocytopenia should be performed during the first week of life. Ultrasound imaging of the brain seems to be indicated if the count is less than 50,000/µL, even in the absence of neurologic findings. Neonates who exhibit severe thrombocytopenia (<20,000/µL) should be treated with platelet transfusion or IVIG or both. Neonates with platelet counts of 20,000 to 50,000/µL do not require IVIG treatment; however, careful platelet count monitoring is needed. Neonates with intracranial hemorrhage or any other bleeding manifestation should be treated with combined platelet transfusion and IVIG or glucocorticoid therapy, especially if the platelet count is less than 20,000/µL. Neonatal thrombocytopenia usually resolves within 4 to 6 weeks.

Neonatal Alloimmune Thrombocytopenia

The pathogenesis of neonatal alloimmune (also known as isoimmune) thrombocytopenia is similar to Rh disease. A mother with antigen-negative platelets is sensitized by antigen-positive fetal platelets gaining access to the maternal circulation via breaches in the placental barrier. As a result, the mother produces antiplatelet antibodies, and these IgG antibodies cross the placenta and destroy the fetal platelets. In contrast to Rh disease, 50% of neonatal alloimmune thrombocytopenia cases occur during the first pregnancy of an at-risk couple. This difference is explained by the higher immunogenicity of the platelet antigen and the smaller size of the platelets, which may facilitate their fetomaternal transfusion.

Of the several types of platelet antigens, the human platelet antigen 1a (HPA-1a) is involved in 80% to 90% of neonatal alloimmune thrombocytopenia cases in whites, and HPA-5b is responsible for a further 5% to 15% of the cases.¹⁹ Among people of color (e.g., Asians), HPA-1a incompatibility is a rare cause of neonatal alloimmune thrombocytopenia, and other alloantigens (e.g., HPA-4b) are implicated.⁵ The fetus acquires the antigen from the father. When the father is heterozygous, 50% of the fetuses would be affected, whereas all fetuses of a homozygous father would be HPA positive.

The prevalence of neonatal alloimmune thrombocytopenia is 0.5 to 2 cases per 1000 deliveries. Fetomaternal platelet incompatibility is much more frequent. The discrepancy is explained by the facilitating role of certain human leukocyte antigen (HLA) types that are associated with the development of neonatal alloimmune thrombocytopenia. HLA-DR3 is associated with a 10- to 30-fold risk for HPA-1a antibody production.

In the usual scenario, an asymptomatic woman delivers an otherwise normal infant in an otherwise uncomplicated birth. Most neonates are asymptomatic, and the thrombocytopenia is detected by a blood count performed for other perinatal causes. In some cases, neonates present with generalized petechiae, hemorrhage into abdominal viscera, excessive bleeding after venipuncture or circumcision, or, in extreme cases, abnormal neurologic manifestations secondary to intracranial hemorrhage. The platelet count commonly decreases further during the first week after birth.

The diagnosis of neonatal alloimmune thrombocytopenia involves typing platelet antigens in the newborn and in the parents to show that the mother lacks a platelet antigen that is present on the platelets of the father and the neonate. A more sophisticated test is to establish the existence of the antiplatelet antibody in the mother’s serum that is directed against a platelet antigen in the father. Testing the infant is generally unnecessary if the father is available for testing.

Human platelet antigen genotyping is now available as a routine laboratory technique and can be used to identify HPA incompatibilities between mother and child. Several techniques are known, and the polymerase chain reaction with sequence-specific primers is used. New microarray technologies are expected to support routine genotyping of all known HPAs, which allows detection of incompatibilities for low-frequency antigens, increasing the sensitivity for the detection of antibodies against low-frequency antigens.

Older methods that measure the antibody associated with platelets lack adequate specificity, but newer enzyme-linked immunosorbent assays specifically detect the antiplatelet antibody. In antigen capture immunoassays, monoclonal antibodies directed against platelet antigens are used to identify various known platelet antigens individually, although these may be negative in maternal blood 2 to 4 weeks after delivery in 30% of the cases. Flow cytometry and polymerase chain reaction assays can also be used to identify the patient’s platelet antigens. Establishing the diagnosis of neonatal alloimmune thrombocytopenia has immediate importance and implications for future pregnancies.

Management of the Neonate

In suspected cases of neonatal alloimmune thrombocytopenia, treatment should be started on the basis of the clinical diagnosis without waiting for the results of the immunologic workup. Management depends on the gestational age of the infant, the severity of the thrombocytopenia, the presence of bleeding, and the presence of additional risk factors for bleeding. Treatment is based on transfusion of random-donor, ABO-compatible and Rh-compatible, and HPA-1a-negative platelets (preferably with HPA-5b–negative platelets as well) in neonates with severe thrombocytopenia. This transfusion is compatible in approximately 90% of cases of neonatal alloimmune thrombocytopenia.5,16,36 When random-donor platelets are unavailable, washed maternal platelets can be administered. Human platelet antigen–incompatible platelets should be used only if compatible ones are unavailable; they can be combined with IVIG treatment to achieve a transient increase in the platelet count until IVIG becomes effective.⁵

High-dose IVIG, 1 g/kg per day for 2 days or 0.5 g/kg per day for 4 days, is also effective in increasing the platelet count in most cases,⁵ although the increase may be delayed for 1 or 2 days.¹¹ Corticosteroids were used in the past, but have become less popular since the availability of IVIG. In any case, the neonatal platelet count should be closely monitored during the first days of life.

Management of a Subsequent Pregnancy

When the diagnosis of alloimmune thrombocytopenia is established, parents need to be counseled regarding risks and management of future pregnancies. The recurrence rate of neonatal alloimmune thrombocytopenia in a subsequent pregnancy is greater than 90%, and the risk for intracranial hemorrhage is the same or greater than in the previous pregnancy. The difference is, however, that in a subsequent pregnancy the patient and her caregivers are aware of the neonatal alloimmune thrombocytopenia affecting the first child. In the absence of screening (which has very low cost-effectiveness) for the presence of antiplatelet antibodies in maternal blood, the diagnosis is almost impossible without a history of neonatal alloimmune thrombocytopenia in a previous gestation. One exception is an incidental finding of intracranial hemorrhage during an ultrasound scan.

The risk for antenatal intracranial hemorrhage in the fetus with alloimmune thrombocytopenia is substantial enough to warrant intervention either by giving the mother weekly infusions of high-dose IVIG with or without corticosteroids (the preferred approach in North American centers) or by repeated in utero fetal platelet transfusions (the preferred approach in some European centers).⁵ There is no way to predict which infant is going to have intracranial hemorrhage.³³ Antenatal therapy is aimed at increasing the number of fetal platelets regardless of the presence of a risk factor.

Although screening procedures are not indicated to detect neonatal alloimmune thrombocytopenia, a high index of suspicion is needed in certain cases (Box 21-2). Typically, a woman presents in early pregnancy with a history of delivering an infant with neonatal alloimmune thrombocytopenia or presents with some clues to the diagnosis.²⁷ The first step should be assessment of the father. In the heterozygous case, the status of the fetus is unknown, and direct assessment of fetal platelets is via PUBS or via genotyping cells in the amniotic fluid. The timing of the procedure and the risk involved are matters of debate. Failure to treat carries the risk for intrauterine intracranial hemorrhage, which is expected to occur in 30% of cases, with 10% of affected newborns dying and 20% experiencing neurologic sequelae secondary to intracranial hemorrhage. Percutaneous umbilical vein blood sampling has a high risk for miscarriage or fetal death. The operator must be prepared to transfuse platelets if the results show a dangerously low platelet count. If the infant is found to be HPA positive or the father is homozygous for the allele, there is a choice between serial platelet transfusions and IVIG administered to the mother.³² Amniocentesis is performed mainly to exclude the presence of platelet antigens and thus to avoid unnecessary interventions.

Box 21-2 Maternal History That Merits Assessment of Alloimmune Thrombocytopenia