In this prospective study, 26 consecutive patients being treated for haematological malignancies receiving standard (i.e. non-leucocyte-depleted) blood components were observed for the development of refractoriness to platelet transfusions. One hundred and sixteen of the 266 (44%) platelet transfusions failed to produce a satisfactory response. In 102/116 (88%), the poor response was in the presence of non-immune factors known to be associated with platelet refractoriness. Non-immune factors were present alone in 78/116 (67%), and in combination with immune factors in a further 24/116 (21%). Immune factors (HLA and platelet-specific antibodies) were present during 29/116 (25%) of unsuccessful platelet transfusions. Statistical analysis confirmed that platelet refractoriness was significantly associated with the presence of non-immune factors. The non-immune factors associated with refractoriness were often multiple, most frequently a combination of fever, infection and antibiotic therapy. This study provides evidence that immune mechanisms were not the predominant cause of platelet refractoriness in the patient population studied. It also suggests that measures for the prevention of HLA alloimmunisation, such as leucocyte depletion, may have a limited impact in reducing the incidence of refractoriness to platelet transfusions.

The recognition that spontaneous intracranial haemorrhage (ICH) may occur in utero in fetomaternal alloimmune thrombocytopenia (FMAIT) led us to attempt to prevent this in 15 pregnancies of 11 women who had previously affected infants with FMAIT due to anti-HPA-1a. The antenatal management included fetal platelet transfusions and maternal steroids and/or high-dose intravenous immunoglobulin (IVIgG). In the first pregnancy, ICH occurred between 32 and 35 weeks' gestation before any treatment had been given, emphasizing the need for earlier intervention. Five of the 14 subsequent pregnancies in this study were considered to be severely affected (severe haemorrhagic complications in a previous infant and initial fetal platelet count < 20 x 10(9)/L in this study); four were managed successfully with weekly fetal platelet transfusions started between 18 and 29 weeks and continued until delivery at 33-35 weeks, and one severely affected case who was referred at 36 weeks was managed successfully with a single platelet transfusion prior to delivery. Five pregnancies were considered to be mildly affected (previous infants were unaffected by severe bleeding and initial fetal platelet count > 50 x 10(9)/L in this study). The platelet counts were maintained in one case with steroids and in three with IVIgG without the need for repeated platelet transfusions, but in the fifth the fetal platelet count fell despite steroids and IVIgG and serial platelet transfusions were required. Four pregnancies were unsuccessful; two pregnancies were terminated after severe ICH occurred at an early stage before fetal blood sampling had been carried out, one fetus died after the mother had a severe fall despite the successful initiation of fetal platelet transfusions and one died due to a cord haematoma which occurred at the time of the initial fetal blood sampling. The optimal management of FMAIT to reduce the risk of antenatal ICH remains uncertain. Steroids and IVIgG may be effective in some mildly affected cases but serial fetal platelet transfusions are the preferred therapy for those who are severely affected.

Feto-maternal incompatibility for the human platelet antigen HPA-1a is an important cause of severe fetal thrombocytopenia. The incidence is 1 in 1000-2000 pregnancies, which is more common than other conditions for which screening is presently carried out. Antenatal diagnosis and management are now available, but only for subsequent siblings following diagnosis of a previously affected infant. This study describes a pilot prospective screening programme for the antenatal detection of fetomaternal alloimmune thrombocytopenia (FMAIT) due to HPA-1a incompatibility. 3473 women were typed for HPA-1a using a method designed for large-scale typing. 71 women found to be HPA-1a negative were further tested for HLA-DR52a as a risk factor for alloimmunization. All women were monitored for the development of anti-HPA-1a throughout pregnancy and a cord full blood count was taken at delivery. Two affected pregnancies were found and treated: a singleton pregnancy was treated antenatally and a twin pregnancy after delivery. The study showed that screening for FMAIT could be established within the pre-existing antenatal red cell serology programme. It was concluded that screening should be based on platelet typing and offered regardless of parity. Further stratification, combining DR52a typing and HPA-1a antibody screening, although focusing on the group of women at greater risk, may not identify all affected pregnancies. Confirmation of the diagnosis and severity of FMAIT continues to depend on fetal blood sampling during pregnancy or cord blood samples after birth.

Risk of type 1 diabetes at 3 years is high for initially multiple and single Ab+ IT and multiple Ab+ NT. Genetic predisposition, age, and male sex are significant risk factors for development of Ab+ in twins.

A simplified method for large-scale HPA-1a phenotyping of platelets was developed for use in an antenatal screening programme for fetomaternal alloimmune thrombocytopenia (FMAIT). The test was based on the MAIPA assay, which was modified for antigen-typing with a well-characterized anti-HPA-1a reagent. The resulting assay gave reliable results, was inexpensive and allowed testing of large batches using semiautomated equipment.