Fetal intracranial haemorrhages caused by fetal and neonatal alloimmune thrombocytopenia: an observational cohort study of 43 cases from an international multicentre registry

Abstract

Objective: To characterise pregnancies where the fetus or neonate was diagnosed with fetal and neonatal alloimmune thrombocytopenia (FNAIT) and suffered from intracranial haemorrhage (ICH), with special focus on time of bleeding onset.

Design: Observational cohort study of all recorded cases of ICH caused by FNAIT from the international No IntraCranial Haemorrhage (NOICH) registry during the period 2001-2010.

Setting: 13 tertiary referral centres from nine countries across the world.

Participants: 37 mothers and 43 children of FNAIT pregnancies complicated by fetal or neonatal ICH identified from the NOICH registry was included if FNAIT diagnosis and ICH was confirmed.

Primary and secondary outcome measures: Gestational age at onset of ICH, type of ICH and clinical outcome of ICH were the primary outcome measures. General maternal and neonatal characteristics of pregnancies complicated by fetal/neonatal ICH were secondary outcome measures.

Results: From a total of 592 FNAIT cases in the registry, 43 confirmed cases of ICH due to FNAIT were included in the study. The majority of bleedings (23/43, 54%) occurred before 28 gestational weeks and often affected the first born child (27/43, 63%). One-third (35%) of the children died within 4 days after delivery. 23 (53%) children survived with severe neurological disabilities and only 5 (12%) were alive and well at time of discharge. Antenatal treatment was not given in most (91%) cases of fetal/neonatal ICH.

Conclusions: ICH caused by FNAIT often occurs during second trimester and the clinical outcome is poor. In order to prevent ICH caused by FNAIT, at-risk pregnancies must be identified and prevention and/or interventions should start early in the second trimester.

Figures

Figure 1

Flow diagram of the study population.

Figure 2

Estimated time period for the onset of intracranial haemorrhage (ICH) is shown for 41 cases of ICH caused by fetal and neonatal alloimmune thrombocytopenia. An arrowhead to the left indicates that the earliest time of onset cannot be estimated, only that the bleeding occurred before the gestational age indicated on the x-axis.

Figure 3

(A) CT scan shows a very large intraparenchymal haemorrhage with mass effect and occupying most of the right frontal lobe in a term baby. Note a second but smaller haemorrhage on the left side adjacent to the ventricle but separate from this. The attenuation of both haemorrhages indicates that the haemorrhage is several weeks old at the time of imaging. We estimated the right haemorrhage to be 4–6 weeks old and the left still a couple of weeks older. The pregnancy was considered normal, and no intravenous immunoglobulin treatment was given. A live boy was born vaginally at 40 gestational weeks with petecchiae. The platelet count was 6×109/l. (B) This CT shows marked ventriculomegaly, partly due to hydrocephalus, partly due to loss of brain tissue in the left hemisphere. Note the very large ventricle on the left side occupying most of the left hemicranium. The shape of the left lateral ventricle is irregular and that of residual from a previous intraventricular haemorrhage with a paraventricular atrophic defect caused by an old periventricular haemorrhagic infarction. This pattern is pathognomonic for this condition even though no actual blood can be detected. The interpretation of this image is that of an intraventricular haemorrhage associated with a periventricular haemorrhagic infarction timed at 28–30 gestational weeks or earlier. Hydrocephalus was diagnosed intrapartum because of breech presentation. A live boy was born at term with a platelet count of 4×109/l. The child had severe cerebral palsy and a hydrocephalus shunt. (C) The CT scan shows multiple focal intraparenchymal haemorrhages throughout both cerebral hemispheres. Most haemorrhages are quite small. Note also the extensive extracranial subgaleal haemorrhage overlying the right hemicranium. All bleedings are of same age and maximum 7 days old. It was noted that the fetus was small on ultrasound scan at 25 weeks, but otherwise the pregnancy was considered normal. At 42 weeks a live boy was born vaginally, with multiple petecchiaes and multiple retinal bleedings. The platelet count at delivery was 14×109/l, with nadir value 8×109/l. He received platelet transfusions. At time of discharge the boy was described as alive and well, but we do not have data on long-term clinical outcome.