Neonatal isoimmune purpura: response to platelet infusions

A systematic review and survey of the management of unexpected neonatal alloimmune thrombocytopenia

BACKGROUND

Unexpected neonatal alloimmune thrombocytopenia (NAIT) may have devastating consequences and its management is challenging. To design future trials, evidence from the literature and existing best practice need review.

STUDY DESIGN AND METHODS

This study was a cross-sectional survey of neonatal units in Germany and Canada to determine management strategies of NAIT and a systematic search for randomized controlled trials (RCTs).

RESULTS

Management of NAIT differs substantially between countries with regard to platelet (PLT) thresholds for screening, initiation of therapy, and treatment. Seventy-seven percent of Canadian physicians versus 68 percent of German physicians screen preterm and term infants, at a PLT threshold of 30 x 10(9) to 100 x 10(9) per L. In preterm infants, 60 percent of Canadian neonatologists commence treatment at a PLT count of between 30 x 10(9) and 50 x 10(9) per L. In Germany 32 percent of the physicians start treatment at this level and 25 percent use a threshold of between 10 x 10(9) and 20 x 10(9) per L. In term infants, 6 percent of the Canadian physicians and 16 percent of the German physicians use even lower treatment triggers of between 5 x 10(9) and 10 x 10(9) per L. In the presence of bleeding, 61 percent of German physicians await the arrival of antigen-negative PLTs, while 64 percent of Canadian neonatologists prefer intravenous immunoglobulin or random-donor PLTs (81%). Maternal PLTs are utilized by 31 percent of physicians in both countries. No RCTs were identified.

CONCLUSION

In the absence of RCTs, management of unexpected NAIT differs between countries. Clinicians and transfusion services may use the results of our study to develop collaborative protocols, redefine preferred hospitalwide strategies, and design future controlled trials.

Survey of the use and clinical effectiveness of HPA-1a/5b-negative platelet concentrates in proven or suspected platelet alloimmunization

The optimal treatment of neonatal alloimmune thrombocytopenia (NAIT) is the transfusion of compatible donor platelets. The National Blood Service in England has established panels of "accredited" donors negative for human platelet antigens HPA-1a and HPA-5b, the most commonly implicated alloantigens. We have retrospectively surveyed the frequency of use and clinical effectiveness of donations collected over a 13-month period from the Oxford accredited panel. Ninety-five per cent of hyperconcentrated platelets (HPCs) collected were issued, all for intrauterine transfusion to fetuses at risk of NAIT due to the presence of maternal platelet alloantibodies and previously affected siblings. Thirty-one per cent of paediatric platelet concentrates (PPCs) collected were issued, of which 57% were used for cases of suspected NAIT. Fifty-four per cent of adult therapeutic doses collected were issued; 5% of these were used in cases of suspected NAIT or proven post-transfusion purpura (PTP). Good increments were seen in most NAIT cases transfused with HPCs or PPCs, and a moderate increment in the one PTP case. We conclude that the establishment of accredited panels is justified and enables delivery of a clinically effective treatment for NAIT. Increased use and cost-effectiveness could be achieved by the delivery of an educational programme to neonatal unit clinical staff to increase the awareness and appropriate treatment of NAIT.

Feto-maternal alloimmune thrombocytopenia: a literature review and statistical analysis

Exploring prognostic factors that determine outcomes in fetomaternal alloimmune thrombocytopenia (FMAIT), a search of Medline was performed covering the years 1966 to April 1998. 376 articles were collected and reviewed; 140 articles contained the case histories of 297 mothers and 433 pregnancies that fulfilled entry criteria. More than 30 data variables were sought from these cases. The data were analysed using SPSS and Arcus Quickstat Biomedical. Nineteen different antigen incompatibilities were documented, the majority being human platelet antigen (HPA)-1a (77.3%), HPA-3a (3.5%) and HPA-5b (3.5%). The relative risk reduction (RRR) in mortality with any intervention was 57% (0.19-0.77) p = 0.009. Treatment of HPA-1a (PlA1) pregnancies with intravenous immunoglobulin (IVIG) increased the likelihood of a neurologically normal outcome, relative risk (RR) 1.68, confidence interval (1.3-2.2) p = 0.0003. Treatment of HPA-1a (PlA1) pregnancies with only antenatal complementary platelet transfusions increased the likelihood of a neurologically normal outcome, RR 1.63 (1.1-2.1) p = 0.01. Despite reviews of more than 400 cases of FMAIT, few prognostic variables are identifiable. Although IVIG appears to reduce the risk of intracranial haemorrhage (ICH), the dosage and timing of IVIG treatment was varied. This study highlights the need for standardised and directed research.

The management of alloimmune neonatal thrombocytopenia

Neonatal alloimmune thrombocytopenia (NAITP), defined as thrombocytopenia (platelet count < 150 x 10(9)/l) due to transplacentally acquired maternal platelet alloantibodies, occurs in approximately 1 per 1200 live births in a Caucasian population. In such a population, the majority (> 75 percent) of cases are due to fetomaternal incompatibility for the platelet specific alloantigen, HPA-1a (P1A1, Zwa). Incompatibility for the HPA-5b (Bra) alloantigen is the next most frequent cause of NAITP in Caucasians; much less common is NAITP due to incompatibility for HLA, blood group ABO or other platelet-specific antigens. In non-Caucasian populations (e.g. Orientals) HPA-1a incompatibility is a rare cause of NAITP and other alloantigens e.g. HPA-4b (Penb, Yuka) are implicated. The greatest clinical challenge relates to the antenatal management of pregnant women alloimmunized to the HPA-1a (P1A1, Zwa) antigen, and particularly the subset of such women who have a history of a previously affected infant with severe thrombocytopenia and/or intracranial hemorrhage (ICH). The risk of antenatal ICH in the fetus of such women is high enough to merit intervention, either weekly infusion of high-dose intravenous immunoglobulin G (IVIG) with or without corticosteroids given to the mother (the preferred approach in North American centres), or repeated in-utero fetal platelet transfusions (the preferred treatment approach in some European centres). Post-natal management of severely affected infants centres on the rapid provision of compatible antigen-negative platelets harvested from the mother or a phenotyped donor. The value of antenatal screening programs to detect 'at risk' alloimmunized women during pregnancy continues to be debated.

The fetal and neonatal consequences of maternal alloimmune thrombocytopenia

Alloimmune thrombocytopenia is a relatively common and under-recognized entity. Prospective screening studies have suggested that at least 1 in every 1000 babies will be affected. While the severity of prospectively identified neonates is not as great as those 'routinely' identified as newborns, the incidence of intracranial haemorrhage in the fetus and neonate is the highest for any immune thrombocytopenia. Diagnosis is complex for the laboratory in view of the large number of platelet antigens and the importance of having sufficient numbers of typed controls. The importance of identifying the affected newborn extends to the likely need for antenatal management of the subsequent affected fetus. Studies to determine the optimal approach to this problem are ongoing. Ideally, prenatal screening of all pregnant women could be performed but this is not currently in practice.

Immune thrombocytopenia in pregnancy: autoimmune and alloimmune

Auto- and alloimmune thrombocytopenias in pregnancy may seriously impact on both mother and fetus. Autoimmune thrombocytopenia (ITP) affects both mothers and fetuses but is considered to be quite benign for both groups. The 'facts' are that: 1) ITP occurs commonly in pregnancy; 2) there has been no reported maternal mortality in more than 20 years; 3) management, except at delivery, is similar to management in the non-pregnant state; 4) splenectomy is virtually never required during pregnancy; 5) significant neonatal thrombocytopenia occurs in approximately 10% of cases and intra-cranial hemorrhage (ICH) 1%; 6) the course of the first sibling predicts that of the next sibling; and 7) the fetal platelet count can be successfully determined (if desired) by either fetal blood sampling (FBS) or by fetal scalp sampling. Many other important considerations remain undetermined: 1) non-invasive prediction of severe fetal thrombocytopenia; 2) the appropriate mode of delivery for a thrombocytopenic fetus; 3) the role of anti-platelet antibody testing; and 4) the effects on the fetal platelet count of maternal therapy. Alloimmune thrombocytopenia (AIT) is easier to outline because it is a far more serious fetal disorder: 1) neonatal platelet counts < 20,000/microliter are common in AIT; 2) there is a 10-30% ICH rate in first affected newborns, some of which occur antenatally; 3) there is no universal prenatal screening although this would be scientifically feasible; 4) testing is complex and requires an experienced laboratory that can test at least five platelet antigens and has sufficient typed controls to confirm the specificity of any anti-platelet antibodies detected; 5) the second affected sibling in a family is usually more severely affected than the first; 6) treatment of the thrombocytopenic neonate can be accomplished with intravenous (i.v.) gammaglobulin and/or platelet transfusions; and 7) treatment of the fetal platelet count can be accomplished in most instances by infusing the mother with i.v. gammaglobulin with or without steroids; platelet transfusions to the fetus is another option.

Platelet and hemorrhagic disorders associated with pregnancy: a review. Part I

Disorders of coagulation remain an important potential cause of maternal morbidity and mortality. Bleeding disorders in pregnancy, unlike disorders of hypercoagulability, most often can have little impact on the mother but devastating consequences for the fetus. Further complicating the issue is that not all disorders of coagulation are inherited. Some are due to maternal illnesses unique to pregnancy, others are due to drug ingestion, and yet others remain idiopathic. In still other instances, thrombocytopenia is a minor consequence of a more severe disorder and will resolve when the inciting agent is removed or treated. A basic understanding of the pathophysiology of various conditions that lead to bleeding diathesis in pregnancy is necessary in order to effectively manage these varied clinical disorders. In addition, knowledge of whether the major morbidity is fetal or maternal or both can impact management. This review is concerned with the etiology, pathophysiology, diagnosis, and general management of commonly encountered disorders in pregnancy that place the mother and fetus at increased morbidity and mortality because of the potential for hemorrhage. Acutely acquired disorders and the resultant maternal manifestation versus a chronic disease process that is altered by the state of pregnancy are distinguished. Where possible, the incidence and prognosis of the disorder are provided. Actual cases are included to illustrate how similar presentations of distinctly different disorders can confuse and complicate an accurate diagnosis that is essential for appropriate management.

Neonatal alloimmune thrombocytopenia associated with anti-human platelet antigen-3a antibody

A sister and brother with neonatal alloimmune thrombocytopenic purpura (NAITP) caused by maternal anti-human platelet antigen (HPA)-3a are reported. The children had transient severe thrombocytopenia in the newborn period, and were treated with intravenous gamma-globulin and platelet concentrates from random donors. Although the sister had intracranial hemorrhage on day 2 postnatally, the development of the child has been normal and no neurological sequelae have been observed. The brother only had bloody stool when the platelet count was low, and did not have severe hemorrhagic manifestations. The diagnosis of NAITP was made by the sera from the mother, which contained anti-HPA-3a antibody directed against platelets of the children. The rate of recurrence might be high in this family, because the father of the patients was found to be homozygous for the HPA-3a gene.

Platelet disorders in newborn infants: diagnosis and management

Platelets are small, disc-shaped, anucleated cells formed by fragmentation of megakaryocytes in the bone marrow. They circulate in blood with a lifespan of 7 to 10 days and, together with fibrin, form hemostatic plugs at sites of vessel injury. Abnormalities of platelets, either quantitative or qualitative, may cause clinically significant bleeding with resultant morbidity and, on occasion, mortality. This review will focus on platelet disorders in neonates, defined as infants of up to 4 months of age. Special emphasis will be given to the physiology of platelet function in healthy and sick newborn infants. The review will be divided into sections as follows: role of platelets in hemostasis, platelet function in newborn infants, quantitative platelet disorders, qualitative platelet disorders, and platelet transfusion therapy.

Management of fetal alloimmune thrombocytopenia by weekly in utero platelet transfusions

Alloimmune neonatal thrombocytopenia (ANT) may cause intracranial haemorrhage in utero as well as at delivery. Recent management has concentrated on attempts to minimise fetal thrombocytopenia and prevent its complications. This report describes further experience with the use of repeated intravascular transfusions of compatible platelets in utero. The patient studied had already had one infant with intracranial haemorrhage due to ANT. In her next pregnancy, weekly intra-uterine platelet transfusions were given from 26 weeks, but intra-uterine death occurred at 30 weeks after the mother had a heavy fall. In her most recent pregnancy, weekly intravascular transfusions of platelets were given by cordocentesis from 29 to 34 weeks. The fetal platelet count was maintained above 30 X 10(9)/l for almost all of the last 6 weeks of pregnancy before delivery of a normal infant by Caesarean section at 35 weeks' gestation. This approach is effective in preventing severe fetal thrombocytopenia in the last trimester of pregnancy and is contrasted with alternative treatments of ANT. Further data are required to determine the efficacy and risks of these treatments.

Management of alloimmune neonatal and antenatal thrombocytopenia

Neonatal and antenatal alloimmune thrombocytopenia is induced by maternal antibodies against platelet-specific fetal antigens. This disease is rare but potentially severe because of intracranial bleedings which may occur during pregnancy or around birth. In the last decade our knowledge of this disorder has markedly advanced. New techniques are used in platelet immunology. New platelet antigens involved in these perinatal thrombocytopenias have recently been discovered. A group of women likely to produce the responsible platelet antibodies has been genetically defined as regards the PLA1 antigen. The quality of the sonographies and the possibility of performing cord vein puncture in early pregnancy afford a new approach in the management of perinatal alloimmune thrombocytopenias. But more must be done to prevent the complications of this disease.

Coagulation disorders in the neonate

A review is presented of those aspects of hemostatic mechanisms that differ between neonates and older children. Practical laboratory tests to evaluate bleeding disorders in the newborn infant are discussed, as are the most frequently encountered primary and secondary conditions that affect hemostasis, both in the otherwise healthy newborn and in the sick low birthweight infant. Pathophysiology and treatment approaches will be briefly reviewed.

Intrauterine intracranial hemorrhage in an infant with isoimmune thrombocytopenia

An infant with isoimmune thrombocytopenia and presumed intrauterine intracranial hemorrhage is presented. This infant developed focal seizures on day 6 of life which were controlled with phenobarbital. Presently, the infant is growing and developing normally. Previous cases have been described and are summarized in this paper. There is no known method for predicting or preventing this particular complication.

Ultrasound in congenital intracranial haemorrhage secondary to isoimmune thrombocytopaenia

Three cases of congenital intracranial haemorrhage, diagnosed post-natally by ultrasound, are presented. The ultrasound findings of intracranial haemorrhage secondary to thrombocytopaenia are discussed. The haematomas were partly organised at birth and the ventricular systems were dilated in two cases. This type of haemorrhage is a different entity from post-natal intracranial haemorrhage in the preterm infant.

The treatment of neonatal isoimmune thrombocytopenia with intravenous immunoglobin (IgG i.v.)

We present a report of the use of IgG i.v. to treat clinically manifest neonatal immune thrombocytopenia. The IgG i.v. was administered at a daily dosage of 0.4 g/kg body weight for 5 days. Treatment was started when the child was 3 days old and had a platelet count of 2 X 10(9)/l. Four days later the platelet count had risen to 200 X 10(9)/l. The diagnosis of immune thrombocytopenia was confirmed by platelet typing of the mother's and child's platelets and identification of anti-platelet antibodies in maternal serum.

Neonatal isoimmune thrombocytopenia. The natural course and management and the detection of maternal antibody

Isoimmune neonatal thrombocytopenia purpura (INTP) is a disease caused by platelet destruction by maternally derived antibody. Six patients were investigated. Maternal derived platelet transfusions in two patients resulted in posttransfusion platelet counts of greater than 85,000/microliter, a normal platelet survival of 7 to 10 days, and an early discharge from the hospital. In four patients, random donor platelet transfusions administered in the first week caused a transient rise of between 7 and 30,000 platelets per microliter with a return to pretransfusion levels 24 to 48 hours later, confirming the ineffectiveness of this form of treatment. Corticosteroid therapy is not effective and because of its potential toxicity should not be recommended for use in INTP. Platelet antigen was identified in the parents, and maternal sensitization was proven by the presence of platelet specific antibody. Severity of the disorder was estimated by the initial platelet count and the duration in weeks for the platelet count to rise above 100,000/microliter. Lymphocytotoxicity and immune lysis, as measured by the 51Chromium release assay, did not show an obvious relationship with the severity of the disorder. In our patients, high scores in the platelet suspension immunofluorescence test related to a more severe disorder, but a larger series would be necessary to confirm this finding.

Placental transfer of maternal anti-rabbit IgG causing falsely elevated TSH levels in neonates

Two infants were found to have markedly increased TSH levels, 104 and 154 mU/l of plasma, respectively, in a routine screening programme for congenital hypothyroidism. THe recall limit used was 50 mU/l of plasma. On follow-up, both infants were clinically euthyroid and had normal serum T4 and T3. The elevated TSH levels were confirmed only with some commercial radioimmunoassay kits--but not with others. Similar results were obtained in TSH assays of samples from their mothers, who had no other biochemical or clinical evidence of thyroid dysfunction. Both mothers had intense contact with rabbits over long periods. The apparent TSH activity was found to be associated with the IgG fraction. It was neutralized by the addition of normal rabbit serum to the samples and was caused by antibodies to rabbit immunoglobulin. The activity was eliminated from the circulation of both infants with a half-life of approximately one month. Apparently, the heterophilic antibodies were of maternal origin and were transferred to the foetus via the placenta. Infants with so-called transient hyperthyrotropinaemia identified in screening programmes have to be reevaluated to exclude false TSH elevations of this type.

Isoimmune neonatal thrombocytopenic purpura

Ten cases of isoimmune neonatal thrombocytopenic purpura (4 pairs of siblings and 2 singletons) are described. The condition was diagnosed by excluding other causes of thrombocytopenia , and in 8 cases it was confirmed by detecting antiplatelet antibodies in the mother. Perhaps steroids should be given to affected infants as soon as the condition is diagnosed in order to stabilise the capillary membrane. Exchange transfusion, using platelet antigen-negative blood if available, helps to remove antibodies and should be followed by the infusion of antigen-negative platelets, easily obtained from the mother by platelet-phoresis. The use of random donor platelets (as in 3 of these cases) was ineffective because 98% of the population are platelet antigen-positive. Nine of the infants recovered completely. The exception was an infant who developed hydrocephalus, possibly as a result of intracranial haemorrhage.

Platelet transfusion therapy

Platelet transfusions are of unquestionably proven benefit for the correction of thrombocytopenia or functional platelet disorders, and they have allowed for more intensive antineoplastic therapy. With the advent of blood component therapy most modern blood banks now have the capabilities for supplying at least limited quantities of platelets. Refinements in procurement methods will inevitably lead to a greater supply of platelets and the establishment of larger transfusion programs. These programs will need to incorporate facilities for platelet storage, recruitment of suitable donors, selection of special donors for refractory patients, and methods for quality control. As antineoplastic therapy becomes more aggressive, such transfusion programs will become an integral part of the operation of cancer treatment centers.

Immune reactions of platelets and their clinical significance

It is recognized that blood platelets with a peculiar cell membrane which carries numerous genetically determined immunogens, but is also intricately associated with the hemostatic process not only participate in immune reactions as a passive target cell, but also as an active functional cell. Pertinent results of animal experiments and of in vivo and in vitro investigations with human blood platelets are reviewed in the attempt to elucidate the role of immune reactions of platelets in the pathogenesis of various human bleeding disorders.