SEARCH

SEARCH BY CITATION

Keywords:

  • alloimmune thrombocytopenia;
  • intracranial hemorrhage;
  • IVIG

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Diagnosis of AIT
  5. Approach to neonatal AIT
  6. Screening for alloimmune thrombocytopenia
  7. Summary
  8. Disclosure of Conflict of Interests
  9. References

Summary.  Fetal and neonatal alloimmune thrombocytopenia (AIT) is the commonest cause of severe thrombocytopenia in neonates, and of intracranial hemorrhage (ICH) in term neonates [1] (J Trop Pediatr, 1999; 45: 237). If a newborn is affected with AIT, the next child will likely be more severely affected, and therefore fetal thrombocytopenia will begin early in gestation [2, 3] (Arch Neurol, 1984; 41: 30; N Engl J Med 1997; 337: 22). This creates a risk of in utero ICH even if there was not one in the previous pregnancy. There are new developments in AIT in regard to diagnosis, treatment, and screening which will be the focus of this review.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Diagnosis of AIT
  5. Approach to neonatal AIT
  6. Screening for alloimmune thrombocytopenia
  7. Summary
  8. Disclosure of Conflict of Interests
  9. References

Fetal and neonatal alloimmune thrombocytopenia (AIT) is the platelet equivalent of Rh disease; the most commonly involved platelet antigen is HPA-1a or PlA1. The mother is HPA-1a-negative (1b1b) and the father is HPA-1a-positive, as is the fetus. When the mother is exposed to fetal platelets, anti-HPA-1a antibodies are generated. These antibodies traverse the placenta via FcRn, as does all maternal IgG, and create fetal thrombocytopenia. There are important differences between AIT [1–8] and Rh disease (Table 1).

Table 1.   Differences between Rh disease and AIT
 RhAIT
Incidence1/100 (25% severe)1/1000
First child affectedNoYes
Routine screening in placeYesNo
Testing readily availableYes (any blood bank)No (send out)
Prophylaxis availableYes (Rhogam)No
Severe clinical phenotypeHydrops, kernicterusICH
Management of next pregnancyIn utero RBC TxMaternal IVIG

Diagnosis of AIT

  1. Top of page
  2. Abstract
  3. Introduction
  4. Diagnosis of AIT
  5. Approach to neonatal AIT
  6. Screening for alloimmune thrombocytopenia
  7. Summary
  8. Disclosure of Conflict of Interests
  9. References

When to suspect AIT

AIT is typically suspected when a neonate with bleeding is found to have a very low platelet count. Occasionally, thrombocytopenia is discovered incidentally on a blood count obtained to r/o sepsis.

Diagnosis: neonatal

Pilot screening studies suggested that neonates with platelet counts <50 000 μL−1 were the ones most likely to have AIT [10–12]; larger screening studies confirmed this. Comparing more than 100 thrombocytopenic neonates with AIT (due to PLA1 incompatibility) to those with non-AIT [7], the most discriminative finding was a platelet count < 50 000 μL−1. If a platelet counts less than 20 000 μL−1 was chosen as the cutoff, >90% of these neonates were shown to have AIT, but only 50% of the AIT neonates were identified. A platelet count of less than 50 000 μL−1 identified 90% of those diagnosed with AIT, but also included some non-AIT cases. As an important part of the reason to make the diagnosis is for the next pregnancy, the cutoff of 50 000 μL−1 was chosen. ‘Unexplained’ thrombocytopenia was also significantly more common in AIT, but less differentiating from non-AIT because one-third of the AIT cases also had another reason to be thrombocytopenic, for example clinical illness. A family history of transient neonatal thrombocytopenia was only present in 18% of the AIT cases. These findings might have been skewed by the reason to do the testing: certain patients who were severely thrombocytopenic and clinically ‘ambiguous’ might have been more likely to be tested. In this regard, ICH was not more frequent in AIT cases, but was intraparenchymal rather than intraventricular, and affected AIT neonates were older, larger, and had higher Apgar scores than the non-AIT ICH cases. A criterion not useful for discriminating thrombocytopenic neonates with AIT is the response to platelet transfusions. Data from the same article [7] comparing AIT to non-AIT patients showed no difference in response to random platelet transfusions. A subsequent study confirmed that most neonates with AIT respond to incompatible platelet transfusions [13,15]; IVIG was useful in some of the cases that did not. These findings support random donor platelet transfusion as first line therapy for severe neonatal thrombocytopenia of any etiology. The duration of effect of the transfusion needs to be monitored to determine the need for repeated transfusion and/or additional therapy, such as IVIG or matched platelets.

What entities are in the differential diagnosis?

Any disease that makes a fetus sick, even thrombosis, can cause neonatal thrombocytopenia [16]. Few conditions, however, are likely to result in platelet counts <50 000 μL−1 [7,14]. These include ‘TORCH’ infections accompanied by a ‘blueberry muffin’ rash; trisomies 13, 18, and 21 with their characteristic features; and inherited thrombocytopenias, particularly the amegakaryocytic ones, CAMT and TAR, as well as Wiskott–Aldrich syndrome and the MYH9 related disorders [17–19].

Approach to neonatal AIT

  1. Top of page
  2. Abstract
  3. Introduction
  4. Diagnosis of AIT
  5. Approach to neonatal AIT
  6. Screening for alloimmune thrombocytopenia
  7. Summary
  8. Disclosure of Conflict of Interests
  9. References

When AIT is suspected, laboratory testing of the mother and father (not the neonate) needs to be arranged. In addition, clinical management needs to be initiated without awaiting test results.

Laboratory investigation

Current consensus is that HPA 1, 3, and 5 should be screened in all cases of potential AIT and also HPA 4 if the patient is of Asian descent. HPA 9 and HPA 15 are the next most common antigen incompatibilities. For testing to confirm AIT, testing must reveal both a platelet antigen incompatibility between the parents and maternal antibody directed against that antigen. If an antibody in the mother directed against the father’s platelets is identified but not an antigen incompatibility, the laboratory needs to distinguish between ‘unimportant’ antigens for example HLA and blood group A vs. an incompatibility causing AIT with an antigen that has not yet been tested. Conversely, identifying a platelet antigen incompatibility without a specific antibody is insufficient to make the diagnosis of AIT although if the mother is HPA-1b1b, one should be suspicious of AIT [7]. In our experience, when a maternal antiplatelet antibody cannot be identified on initial testing soon after delivery of the affected neonate, testing in the 10–20 week period of the next pregnancy is the optimal time to try again to identify it.

Testing should only be performed by a very experienced reference laboratory. Both a large number of available typed controls and DNA-based technology to type multiple antigens are required as well as the capability to pursue identification of rare antigens. If the clinical diagnosis is highly likely, that is, the platelet count is very low on day 1 and there is no other explanation for the neonatal thrombocytopenia, usually AIT will be confirmed serologically. If AIT is not confirmed in such a patient, it is important to ask the laboratory to pursue their highest level of workup. If further workup is unrevealing of AIT, during the subsequent pregnancy, screening for maternal antibody would be obtained every 10 weeks or so. It is also important to review the clinical features with more follow-up to see if another cause could be identified, for example inherited thrombocytopenia.

Management of the affected neonate

The mainstay of treatment of neonatal thrombocytopenia is random donor platelet transfusion [7,14,15]. Platelet transfusion is recommended in well term neonates if the count is less than 30 000 μL−1, unless an ICH is diagnosed, in which case 100 000 μL−1 is used. A higher count, for example 50 000 μL−1 may be selected for prematurity, birth asphyxia, or another predisposition to ICH. IVIG could also be infused [5] as matched (antigen negative) platelets. The latter are the most commonly concentrated (not washed) maternal platelets, but could also be random donor platelets which are HPA1b1b/5a5a, compatible in >90% of AIT cases [20].

For marked neonatal thrombocytopenia-presumed AIT, consider the following recommendations:

  • 1
    Ten–20 mL/kg of platelets may be given.
  • 2
    The IVIG dose is 1 g kg−1 day−1 for 1–3 days depending upon response.
  • 3
    We use 1 mg of IV methylprednisolone every 8 h (3 mg day−1) with IVIG without a taper; this lower dose lessens the risk of fungal infection.
  • 4
    A head ultrasound, CT or MRI is mandatory for a thrombocytopenic neonate [21]. If it is AIT complicated by even asymptomatic ICH, the target platelet count is higher, that is >100 000 μL−1, and management of the next affected pregnancy will be more intensive and starts earlier. Optimally, the head sono should be obtained as close to birth as feasible to facilitate, determining whether an ICH, if present, is antenatal or perinatal by its characteristics [22,23].
  • 5
    AIT often resolves within 2–4 weeks. To exclude inherited or other thrombocytopenias, follow the platelet count until a normal count is achieved off treatment.

If AIT is confirmed in the previous sibling, will the next fetus be affected?

If the previous child was affected, the likelihood of the next fetus being affected depends upon the father’s typing. If, for example, the father is a homozygote for HPA-1a (as 75% of HPA-1a men are) and the mother is homozygous for HPA-1b and has an anti-HPA-1a antibody, then all subsequent fetuses from that couple will be affected. If there is paternal heterozygosity, uncertainty, or unavailability, then amniocentesis or even fetal blood sampling would be required to determine the fetal platelet type. In cases in which the father is a known homozygote, treatment can be initiated without fetal blood sampling and without amniocentesis [24]. There have been extremely rare cases in which a subsequent platelet-antigen-incompatible fetus has not been affected, and therefore, one can generally assume that the current (second) fetus is affected. How affected? The second affected fetus will be at least as severely but usually more severely affected than the previously affected sibling [3].

Antenatal treatment of the subsequent affected fetus

  • 1
    To minimize ICH, the fetal platelet count is used as a surrogate marker of risk of fetal ICH.
  • 2
    Fetuses require different management depending upon the history of their previous sibling. On the one hand, 1 g kg−1 week−1 of IVIG alone is clearly insufficient in siblings of fetuses with a previous in utero ICH. On the other hand, 0.5 mg kg−1 day−1 of prednisone is almost as good as 1 g kg−1 week−1 IVIG in the least affected fetuses: those with a sibling without an ICH and with a pretreatment fetal platelet count >20 000 μL−1 [22].
  • 3
    If the initial fetal platelet count is <20 000 μL−1 at 20 weeks of gestation, IVIG alone at 1 g kg−1 week−1 has substantially lesser effect than IVIG + prednisone and a low response rate [22]. It is presumably in such fetuses with early severe fetal thrombocytopenia that cases of ICH will be seen in a second pregnancy even though they did not occur in the first sibling.
  • 4
    No non-invasive biologic marker of severity of fetal thrombocytopenia currently exists.
  • 5
    Fifty percentage of 98 affected fetuses already had platelet counts <20 000 μL−1 by 25 weeks indicating early severity. Forty percentage had lower fetal platelet counts at that time than their previously affected siblings had had at birth indicating worsening in subsequent pregnancies [3].
  • 6
    No spontaneous improvement of fetal thrombocytopenia occurs. It actually worsens as gestation progresses, presumably because of increasing transport of maternal IgG across the placenta.
  • 7
    For platelet antigen incompatibilities other than HPA-1a, much less data exist regarding antenatal management and clinical course. In limited study, incompatibility of HPA-3a, while infrequent, is as severe as that of HPA-1A [25], while incompatibilities of HPA-5b and HPA-9b are less severe [26]. HPA-4 incompatibility seems also to be severe [27], and most rare antigens are identified because of a severe case of neonatal AIT.
  • 8
    Fetal blood sampling could be performed as early as 18–20 weeks. However in severe fetal thrombocytopenia, bleeding occurs with fetal blood sampling. This complication rate can be reduced, but not entirely corrected, by administration of matched platelets [22,23,28–30]. Therefore, it should be avoided whenever possible.

Since the initial study demonstrating efficacy of maternally-administered IVIG [31], there have been several randomized trials that have helped to determine the optimal approach to management [22,24,32]. The first step is to assess the risk of ICH in the fetus: currently this is based almost entirely on the past fetus. Recent studies have stratified treatment based upon whether an ICH occurred in the previous sibling and, if so, whether it happened in the second trimester, in the third trimester, or perinatally.

Based on data accumulated from a number of trials [22–24,30–32], our current policy is to start treatment at 12 weeks of gestation in any mother whose previous baby had an intracranial hemorrhage. One would initiate treatment with IVIG 2 g kg−1 week−1 in (fortunately rare) circumstances in which the previous fetus suffered an ICH in the second trimester (prior to or at 28 weeks). Treatment is initiated with either 1 or 2 g kg−1 week−1 of IVIG if the previous sibling had an ICH in the third trimester or perinatally. If there was no ICH in the previous sibling, one can wait until 20–26 weeks to initiate treatment with IVIG 1 g kg−1 week−1 and prednisone 0.5 mg kg−1 day−1. It is our policy to give the higher level of therapy if there is uncertainty. For example, if ultrasound monitoring in the previous fetus demonstrated no hemorrhage at 24 weeks of gestation and an ICH at 32 weeks of gestation, we would treat on the highest risk arm, as we could not be certain that the ICH occurred after 28 weeks [23].

Key points from past studies

  • 1
    In the past, we advocated fetal blood sampling to determine response to therapy. All current treatment programs minimize or even eliminate the use of fetal sampling [20,33].
  • 2
    Our approach, if there were to be no sampling, would be to do a staged acceleration of therapy, that is, additional IVIG or steroids at a fixed time, as if sampling had been performed and confirmed a low count, and therefore treatment was intensified.
  • 3
    A consensus rational approach to the management of subsequent pregnancies when the first affected fetus did not have an ICH (most cases) is complicated. On the one hand, many patients would do well with almost any therapy (e.g., 0.5 mg kg−1 day−1 prednisone or IVIG 1 g kg−1 week−1). On the other hand, the approximately 25% of severely affected fetuses with fetal platelet counts <20 000 μL−1 by 24 weeks of gestation (had fetal sampling been performed) would not do well unless they received at least IVIG 1 g kg−1 week−1 + prednisone 0.5 mg kg−1 day−1. Distinguishing these fetuses intrapartum is currently not possible without fetal sampling. Pending the ability to make such a distinction non-invasively, we would recommend treatment for all fetuses that would be effective for the most severely affected.
  • 4
    There is no established usefulness of antibody testing in a second affected fetus to determine how affected the fetus is.

Screening for alloimmune thrombocytopenia

  1. Top of page
  2. Abstract
  3. Introduction
  4. Diagnosis of AIT
  5. Approach to neonatal AIT
  6. Screening for alloimmune thrombocytopenia
  7. Summary
  8. Disclosure of Conflict of Interests
  9. References

Neonatal screening

Neonatal screening studies have supported the value of the 50 000 μL−1 platelet cutoff for suspecting AIT [7,10]. However, this applies entirely to HPA-1a incompatibility. Other incompatibilities, for example HPA-5b, are milder [34], and therefore may occur at higher platelet counts [26].

Antenatal screening

Three large studies of antenatal screening for HPA-1a incompatibility have been performed [9,35,36]. Two, from East Anglia [35] and Scotland [36] in the UK, were performed in approximately 25 000 cases each. The largest study in Norway encompassed more than 100 000 pregnancies [9]. Key findings from these studies were:

  • 1
    It had been estimated that the incidence of AIT in the neonate was approximately 1:5000. However, in the screening studies using HPA-1a incompatibility only, the incidence was 1:1000.
  • 2
    This has confusingly created two parallel sets of statistics for key features of AIT, such as the rate of ICH. ICH occurs in 11–21% of clinically recognized cases [6,7], but much less commonly in cases detected by routine screening [9,35,36]. A complicating feature in the assessment of the largest study is that screened cases were delivered electively by cesarean section at 37–38 weeks, potentially reducing the incidence of ICH.
  • 3
    Antibody titers in the HPA-1a negative cases may be useful [37,38].
  • 4
    Antenatal management remains unclear in cases diagnosed by antenatal screening.

The bottom line is that it is currently unclear if routine antenatal screening should be instituted.

Summary

  1. Top of page
  2. Abstract
  3. Introduction
  4. Diagnosis of AIT
  5. Approach to neonatal AIT
  6. Screening for alloimmune thrombocytopenia
  7. Summary
  8. Disclosure of Conflict of Interests
  9. References
  • 1
    Use only a very experienced reference laboratory to obtain testing.
  • 2
    Manage neonates with platelet counts <50 000 μL−1 (especially on the first day of life) as if they have AIT. Send testing on the parents, begin with random platelet transfusion, and urgently obtain a head ultrasound to determine the appropriate platelet threshold for the treatment of the neonate.
  • 3
    If the previously affected sibling had an ICH, the next affected fetus will have early, severe thrombocytopenia and in utero ICH, without effective treatment.
  • 4
    Effective antenatal treatment exists for cases recognized in the previously affected sibling. Maternal-Fetal Medicine specialists are the involved practitioners.
  • 5
    Treatment intensity depends upon ICH in the previous sibling. Avoid fetal blood sampling whenever possible.

Disclosure of Conflict of Interests

  1. Top of page
  2. Abstract
  3. Introduction
  4. Diagnosis of AIT
  5. Approach to neonatal AIT
  6. Screening for alloimmune thrombocytopenia
  7. Summary
  8. Disclosure of Conflict of Interests
  9. References

The author received clinical research support from the following companies: Amgen, Biogen-IDEC, Cangene, Genentech, GlaxoSmithKline, Genzyme, Immunomedics, Ligand, MGI Pharma/Eisai, Inc, and Sysmex. He participated for Baxter, Amgen, and GlaxoSmithKline in their speaker’s bureau program. He owns stock in Amgen and GlaxoSmithKline. He has participated in Advisory Boards for Amgen, GlaxoSmithKline, Ligand, and Baxter.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Diagnosis of AIT
  5. Approach to neonatal AIT
  6. Screening for alloimmune thrombocytopenia
  7. Summary
  8. Disclosure of Conflict of Interests
  9. References