When a Mother's Immune System Mistakes Her Baby's Platelets for Something Foreign

A plain-language guide to fetal and neonatal alloimmune thrombocytopenia, drawn from a 2025 expert review of what doctors now know and still cannot predict

Sometimes a mother's immune system makes antibodies against her baby's platelets — the cells that stop bleeding — because the baby inherited a marker from the father that she does not carry. The antibodies cross the placenta and destroy the baby's platelets, a condition called fetal and neonatal alloimmune thrombocytopenia. Most affected babies do well, but a small number bleed into the brain before birth, and doctors still cannot reliably tell in advance which pregnancies are at risk.

The problem this condition creates for families

Almost every family who meets this diagnosis meets it by surprise. The pregnancy was normal, the scans were normal, the birth was normal — and then a healthy-looking newborn turns out to have pinpoint red spots on the skin and a blood test showing almost no platelets. Because there is no routine test in pregnancy that looks for this, the first affected baby in a family is nearly always found by accident. That surprise is the defining feature of the condition, and it shapes everything that follows: how it is diagnosed, how it is treated, and how much doctors can tell parents about a future pregnancy. A 2025 review by specialists in Giessen and Ulm, Germany, brings together what is known [1].

The condition itself is easier to grasp than its name. Platelets carry surface markers, rather like blood groups. If a mother lacks the common marker — the one called HPA-1a — and her baby inherits it from the father, her immune system may treat those platelets as foreign and make antibodies against them. Antibodies are designed to cross the placenta so that a newborn arrives with some of its mother's immunity, and these ones cross too. Once in the baby's circulation they attach to the baby's platelets and cause them to be cleared away. The result is a low platelet count and, in some babies, bleeding.

How common is it? Roughly 1 in 500 pregnancies produces these antibodies. Around 1 in 2,500 newborns has the condition seriously enough to cause a very low platelet count, which makes it the commonest cause of severe low platelets in newborns. The most feared complication, bleeding into the brain, occurs in about 1 in 10,000 newborns overall [2]. Put another way: it is uncommon, but it is not rare, and every large maternity service will see it.

How families used to face this, and what changed

For most of the twentieth century there was no name for it and no way to act. Babies were born with unexplained bruising and low platelets, some recovered, some did not, and nobody could tell a family whether it would happen again. The condition was properly described only in 1989, when a large European series of 348 suspected cases established the pattern — an otherwise healthy baby with bruising or bleeding and low platelets — and confirmed that maternal antibodies against the father's platelet markers were the cause [3].

Treatment, unusually, came before understanding. In 1984 doctors in Paris tried transfusing a mother's own platelets directly into her unborn baby shortly before delivery [4]. It worked, and for the next fifteen years the standard approach was to pass a needle into the umbilical cord repeatedly through pregnancy to check the baby's platelet count and top it up. This was frightening for families and genuinely dangerous: the cumulative risk of losing the pregnancy was around 6 in 100 [1].

The change that made the biggest difference to families came in 1988, when American doctors showed that giving the mother an infusion of immunoglobulin — antibodies pooled from donated blood — protected the baby without anyone touching the baby at all [5]. When the two approaches were later compared across 26 studies, the results were striking. Mothers treated with weekly infusions alone had good outcomes 98.7% of the time, with brain bleeding in only 4 of 315 pregnancies, while the invasive approach carried an 11% complication rate, a third of those complications ending in the loss of the baby [6]. Needles into the umbilical cord have essentially been abandoned. A weekly drip for the mother replaced them.

Why there is still no test in pregnancy

The obvious next step would be to test every pregnant woman, find the ones with these antibodies, and treat them. Norway actually tried this on a national scale, testing pregnant women and following those who turned out to have antibodies [7]. The programme showed that screening is possible. It did not show that screening is worthwhile, and the reason is uncomfortable.

Having the antibodies is common; having a serious problem because of them is not. In a large Dutch study that followed pregnancies where the mother had antibodies and the baby carried the marker, 81 such pregnancies produced exactly one baby with bleeding into the brain — about 1 in 80 [8]. If a screening programme treated everyone with antibodies, roughly 79 out of every 80 women would receive twenty or more weekly infusions, with the inconvenience, cost and side effects that involves, without their baby ever having been at real risk. And there is no blood test that separates the one from the seventy-nine. Antibody levels have been studied hard, but a high level predicts a seriously affected baby only about half the time, and — a problem that receives less attention than it should — different laboratories measuring the same sample can produce wildly different numbers [1].

There is one genetic test that helps, though only in a reassuring direction. Women who carry a particular immune gene called HLA-DRB3\*01:01 are the ones who make these antibodies; women without it almost never do. That makes the test very good at ruling risk out and poor at ruling it in.

What can and cannot be predicted for a next pregnancy

Once a family has had one affected baby, the picture becomes clearer, and this is where most of the practical value of the diagnosis lies. The crucial question is whether the affected child had bleeding into the brain.

The timing matters here, and it surprises many parents: when brain bleeding happens in this condition it usually happens before birth, often well before. In an international registry of 43 such cases, 23 — more than half — had occurred before 28 weeks of pregnancy, and about 60% affected the family's first baby [9]. This is why a carefully managed delivery, while important, cannot be the main protection, and why treatment during pregnancy starts early.

If the older brother or sister did not have brain bleeding, the outlook for a next pregnancy is genuinely reassuring. In one series, none of 64 untreated subsequent pregnancies in this situation resulted in brain bleeding — a result close to what was seen in women who were treated [6]. Doctors now debate openly whether these mothers need treatment at all, and a good conversation with a specialist should feel like a real weighing-up rather than an automatic prescription. If the older child did have brain bleeding, the risk of it happening again is high, and treatment is strongly recommended.

Low platelets, as distinct from bleeding, tend to repeat quite faithfully. Among 45 women followed into a later pregnancy, the platelet count was about the same in half, worse in about a third, and better in under a fifth. One finding is worth knowing: a first baby who escaped unharmed does not guarantee a second one will — 5 of 15 such women had an affected second baby [10]. If you have had antibodies detected, a later pregnancy is worth discussing with a specialist team even if your first child was fine.

What treatment looks like, and what happens to the baby

Where treatment is given, it is a weekly infusion of immunoglobulin for the mother, usually starting between 12 and 16 weeks if a previous child had brain bleeding, or between 20 and 24 weeks if not, and continuing until delivery. The commonest side effects are headache, nausea and a skin rash. If the father carries only one copy of the marker, a blood test on the mother from 12 weeks can often determine whether this baby has inherited it — and if not, no treatment or monitoring is needed at all.

After birth, the baby's platelet count is checked immediately and the head is imaged to look for bleeding. If the count is very low, or if bleeding is suspected, the baby is given a platelet transfusion, ideally from a donor whose platelets lack the marker the mother's antibodies attack. The count often keeps falling for a few days before it recovers, so repeated checks are normal and not a sign that something is going wrong; occasionally it takes eight to twelve weeks to return to normal. Deciding when a newborn needs a platelet transfusion is an area of active research more broadly, and In[Neo]Sight has covered it separately (platelet-thresholds2yr).

Most babies do well. For the small number who had bleeding into the brain, the outlook is serious, and follow-up is essential [11]. Something that is often missed deserves saying plainly: mothers carry this too. In one study, 54% of mothers reported anxiety, low mood or sleep problems lasting more than six weeks after the diagnosis [12]. If that is you, it is a recognised part of this condition, not a personal failing, and it is worth raising with your team.

What researchers are working on next

Three things are moving. The first is measurement: laboratories around the world are trying to agree on a single standard way of measuring these antibodies, because until they do, no study can convincingly link antibody level to risk of bleeding — and without that link, national screening cannot be justified. The second is a new class of drug. Medicines that block the placental transport of antibodies have already been tested successfully in a related condition affecting red blood cells [13], and trials are underway in this one; if they succeed, they may replace weekly infusions with something simpler. The third is a widening of what counts as a good outcome. Follow-up studies have found that children diagnosed with this condition have somewhat more learning and developmental difficulties by around age twelve than expected, including some who never had recognised brain bleeding [14] — which suggests that "no bleed" may be too narrow a definition of success.

The honest summary is that this is a condition doctors treat well and predict badly. If your family has been affected, the most useful thing you can do is make sure the diagnosis is properly confirmed in a specialist laboratory, and that any future pregnancy is planned with a team that knows the condition — because for a family already identified, the tools available today work very well indeed.

References

  1. Sachs UJ, Bedei I, Wienzek-Lischka S, et al. Diagnosis and Management of Fetal and Neonatal Alloimmune Thrombocytopenia: An Update 2025. Transfus Med Hemother. 2025;52(5):318–333. doi:10.1159/000547985 ↩
  2. Kamphuis MM, Paridaans NP, Porcelijn L, Lopriore E, Oepkes D. Incidence and consequences of neonatal alloimmune thrombocytopenia: a systematic review. Pediatrics. 2014;133(4):715–721. doi:10.1542/peds.2013-3320 ↩
  3. Mueller-Eckhardt C, Kiefel V, Grubert A, et al. 348 cases of suspected neonatal alloimmune thrombocytopenia. Lancet. 1989;1(8634):363–366. doi:10.1016/s0140-6736(89)91733-991733-9) ↩
  4. Daffos F, Forestier F, Muller JY, et al. Prenatal treatment of alloimmune thrombocytopenia. Lancet. 1984;2(8403):632. doi:10.1016/s0140-6736(84)90614-790614-7) ↩
  5. Bussel JB, Berkowitz RL, McFarland JG, Lynch L, Chitkara U. Antenatal treatment of neonatal alloimmune thrombocytopenia. N Engl J Med. 1988;319(21):1374–1378. doi:10.1056/NEJM198811243192103 ↩
  6. Winkelhorst D, Murphy MF, Greinacher A, et al. Antenatal management in fetal and neonatal alloimmune thrombocytopenia: a systematic review. Blood. 2017;129(11):1538–1547. doi:10.1182/blood-2016-10-739656 ↩
  7. Kjeldsen-Kragh J, Killie MK, Tomter G, et al. A screening and intervention program aimed to reduce mortality and serious morbidity associated with severe neonatal alloimmune thrombocytopenia. Blood. 2007;110(3):833–839. doi:10.1182/blood-2006-08-040121 ↩
  8. de Vos TW, Winkelhorst D, Porcelijn L, et al. Natural history of human platelet antigen 1a-alloimmunised pregnancies: a prospective observational cohort study. Lancet Haematol. 2023;10(12):e985–e993. doi:10.1016/S2352-3026(23)00271-500271-5) ↩
  9. Tiller H, Kamphuis MM, Flodmark O, et al. Fetal intracranial haemorrhages caused by fetal and neonatal alloimmune thrombocytopenia: an observational cohort study of 43 cases from an international multicentre registry. BMJ Open. 2013;3:e002490. doi:10.1136/bmjopen-2012-002490 ↩
  10. Tiller H, Husebekk A, Skogen B, Kjeldsen-Kragh J, Kjaer M. True risk of fetal/neonatal alloimmune thrombocytopenia in subsequent pregnancies: a prospective observational follow-up study. BJOG. 2016;123(5):738–744. doi:10.1111/1471-0528.13343 ↩
  11. Winkelhorst D, Kamphuis MM, Steggerda SJ, et al. Perinatal outcome and long-term neurodevelopment after intracranial haemorrhage due to fetal and neonatal alloimmune thrombocytopenia. Fetal Diagn Ther. 2019;45(3):184–191. doi:10.1159/000488280 ↩
  12. McFarland EA, Vander Haar EL, McKelvy MH, et al. Fetal-neonatal alloimmune thrombocytopenia: mothers are affected too. Br J Haematol. 2025;206(6):1754–1763. doi:10.1111/bjh.20112 ↩
  13. Moise KJ, Ling LE, Oepkes D, et al. Nipocalimab in early-onset severe hemolytic disease of the fetus and newborn. N Engl J Med. 2024;391(6):526–537. doi:10.1056/NEJMoa2314466 ↩
  14. de Vos TW, van Zagten M, de Haas M, et al. Children newly diagnosed with fetal and neonatal alloimmune thrombocytopenia: neurodevelopmental outcome at school age. J Pediatr. 2023;258:113385. doi:10.1016/j.jpeds.2023.02.031 ↩