Clinical significance of fetal intracranial hemorrhage

We reviewed our experience with six consecutive cases of fetal intracranial hemorrhage and the cases published in the English literature in an attempt to devise an original prognostic scoring system for antenatal intracranial hemorrhage.

The series included the cases of fetal intracranial hemorrhage detected at our institution between 1992 and 1994 by transabdominal ultrasonography. In addition, we performed an English literature search (Medline computer search, National Library of Medicine) of all reported cases of a prenatal diagnosis of intracranial hemorrhage. The prenatal ultrasonographic findings were correlated with the clinical outcome, which was divided into (1) normal outcome or mild neurologic sequelae and (2) poor outcome (severe neurologic impairment and fetal or neonatal death).

Six cases of intracranial hemorrhage were detected in a population of 6641 pregnancies (0.9/1000) at our institution. Parenchymal involvement was present in three cases. Review of the English literature revealed 35 additional cases with prenatal ultrasonographic findings and postnatal follow-up. The total cases (n = 41) were divided into three groups: (1) isolated intraventricular hemorrhage (n = 20), (2) parenchymal hemorrhage (n = 13), and (3) subdural or subarachnoid hemorrhage (n = 8). Overall, poor outcome was present in 68% of cases, including 45% (9/20) of intraventricular hemorrhage, 92% (12/13) of parenchymal hemorrhage, and 88% (7/8) of subdural or subarachnoid hemorrhage. The heterogeneity of the intraventricular hemorrhage group in both severity of antenatal findings and outcome prompted us to devise a prognostic scoring system based on prenatal ultrasonographic lesions, grouping cohorts with similar outcomes. Outcome was favorable in 100% (5/5) of grade 1 intraventricular hemorrhage cases, in 50% (6/12) of grade 2 cases, and in 0% (0/3) of grade 3 cases.

Fetal intracranial hemorrhage can be classified on the basis of the anatomic location of the intracranial bleeding. The prognosis is poor in nearly 90% of parenchymal and subdural hemorrhages, whereas it is better in the subgroup with intraventricular hemorrhage. The prognostic scoring system we propose for intraventricular hemorrhage may assist the physician in providing patients with prognostic information.