Summary. Anti-HPA-5b is the most commonly found platelet-specific antibody among pregnant women, but it does not cause severe fetal±neonatal alloimmune thrombocytopenia in the majority of affected infants. However, as the sequelae of the affected children may become severe, it is necessary to identify the risk factors for neonatal alloimmune thrombocytopenia. Of 21 354 consecutive pregnant women, 138 [0´65%; 95% confidence interval (CI) 0´54± 0´75%], corresponding to 13´2% of the 1049 HPA-5b 2 women caculated by the gene frequency, were positive for anti-HPA-5b at the first trimester. Anti-HPA-5b was titrated in specimens obtained at the third trimester and antibodypositive women and their neonates were HPA-5 genotyped. Platelet counts in cord blood and 3 d after birth were assessed in the infants born to these mothers. Chi-square analysis showed no significant relationship between the titres of maternal antibody to HPA-5b and the number of pregnancies. There was a significant difference in platelet counts at d 3 between neonates who were compatible (267 Â 10 9 /l) or incompatible (220 Â 10 9 /l, P , 0´05) with maternal anti-HPA-5b. HPA-5b antibody titres $ 64 were related to the development of thrombocytopenia (, 150 Â 10 9 /l) in neonates 1 d and 3 d after birth. A high titre ($ 64) had a positive predictive value of 50% for thrombocytopenia 3 d after birth when the infant was HPA5b 1 and a negative predictive value of 100%. These results indicate that a high titre ($ 64) of anti-HPA-5b is associated with a higher risk of neonatal thrombocytopenia, even if anti-HPA-5b-induced severe thrombocytopenia rarely develops.

These observations strongly suggest that the main factor for TTV acquisition in children involves their age-associated increase in environmental interactions with infectious materials. Genotype 1 might be involved in a weak or a limited pathologic role, which can possibly be diluted by other harmless genotypes.

Background ： It has been controversial whether HLA antibodies cause