Prenatal diagnosis and postnatal outcome of fetal intracranial hemorrhage: a single-center experience

Article information

Obstet Gynecol Sci. 2024;67(4):393-403

Publication date (electronic) : 2024 June 20

doi : https://doi.org/10.5468/ogs.24097

Suhra Kim, MD, Yun Ji Jung, MD, PhD, Jiwon Baik, MD, Hayan Kwon, MD, PhD, JoonHo Lee, MD, PhD, Ja-Young Kwon, MD, PhD, Young-Han Kim, MD, PhD

Department of Obstetrics and Gynecology, Institute of Women’s Medical Life Science, Yonsei University College of Medicine, Seoul, Korea

Corresponding author: Young-Han Kim, MD, PhD, Department of Obstetrics and Gynecology, Institute of Women’s Medical Life Science, Yonsei University College of Medicine, 50-1 Yonseiro, Seodaemoon-gu, Seoul 03722, Korea, E-mail: YHKIM522@yuhs.ac

Received 2024 March 22; Revised 2024 May 26; Accepted 2024 June 11.

Abstract

Introduction

Fetal intracranial hemorrhage (ICH) refers to a hemorrhage that occurs antenatally in the ventricles, subdural space, subarachnoid space, or brain parenchyma. This entity has been reported to affect approximately 0.5-1.0 pregnancies per 1,000 [1,2]. Various maternal and fetal predisposing factors are associated with fetal ICH [3,4]. However, the etiology can be identified in only 20-45% of fetuses [5,6]. Fetal ICH most commonly occurs in the third trimester, mainly due to germinal matrix hemorrhage [7]. Fetuses with a prenatal diagnosis of ICH are at high risk of perinatal mortality and adverse neurodevelopmental outcomes [8]. Sileo et al. [9] reported that perinatal death occurred in 14.6% of fetuses with ICH and 32.0% developed cerebral palsy. Given its strong association with poor outcomes, prenatal detection of fetal ICH is essential for pregnancy management.

Due to the subtle and variable characteristics of hemorrhage on ultrasonography, it is challenging to distinguish fetal ICH from other intracranial lesions [8,10]. The detection rate of fetal ICH has increased in recent years, which can be attributed to advancements in prenatal ultrasonography techniques and the increased utilization of fetal magnetic resonance imaging (MRI) [3,9,11,12]. Nevertheless, several cases of fetal ICH are still being missed, and the incidence varies across regions and studies [8]. Furthermore, depending on the clinician’s understanding of fetal ICH, parents may be provided with incorrect or inappropriate clinical advice. Therefore, precise knowledge of the short- and long-term consequences of fetal ICH in conjunction with prenatal ultrasonography findings is required.

This study aimed to investigate the prenatal ultrasonography findings and postnatal outcomes of fetal ICH for pregnancy management and prenatal parental counseling in 27 fetuses with ICH.

Materials and methods

Results

Twenty-seven fetuses with ICH were identified in our tertiary hospital. Demographic characteristics and ultrasonographic findings of the fetuses are summarized in Table 1. The median age of the women at the time of diagnosis in all the 27 fetuses was 32 years (range, 19-44). Median gestational age of the women at the time of diagnosis was 29 weeks and 4 days (range, 18 weeks 4 days to 36 weeks 6 days). Nineteen fetuses (70.3%) were identified during the third trimester of screening. When comparing postnatal outcomes, there were 16 live births (59.3%), seven lost to follow-up (25.9%), and four intrauterine fetal deaths (14.8%). There were no postnatal deaths among the liveborn infants during postnatal treatment or follow-up. Fetal ICHs was graded as 3-4 in 24 fetuses, while only three fetuses were graded as 1-2. In our study, we observed that fetal ICH occurred unilaterally in 55.5% fetuses (n=15) and bilaterally in 44.4% (n=12). Bilateral ventriculomegaly was present in 75% of fetuses, especially in those with grade 3 or higher ICH. Four fetuses had both subdural and intraventricular involvement (Fig. 1).

Table 1.

Prenatal characteristics and ultrasonography findings of fetal ICH

Fig. 1.

Image showing subdural hemorrhage. Patient was referred to our hospital due to a history of previous intrauterine fetal death (IUFD). First visit at 26 weeks showed IUFD, ventriculomegaly, left subdural space hemorrhage (arrow), right temporoparietal area hemorrhage (arrowhead), fetal ascites, pleural effusion, scalp edema, and intracardiac hematoma. After hysterotomy, cord next generation sequencing and parental gene screening was performed, but significant result were not found.

Representative ultrasonographic findings of fetal ICH are presented in Figs. 2-5. Fig. 2 shows that progressive ventriculomegaly accounted for most of the findings, as observed in 22 (81.5%) fetuses. Fig. 3 shows increased periventricular echogenicity, which was evident in 21 fetuses (77.8%). Fig. 4 shows the echogenic intraventricular clots elicited in 11 fetuses (40.7%). Fig. 5 illustrates the irregular choroid plexus observed in eight fetuses (29.6%). In most fetuses, the prenatal ultrasonography and postnatal MRI findings were consistent, except for one in which cerebral hemorrhage was too small to be detected.

Fig. 2.

Image demonstrating progressive ventriculomegaly. Ultrasound image (left) showing left ventriculomegaly with a hyperechogenic blood clot in the lateral ventricle. Magnetic resonance imaging (right) showing the same results, indicating ventriculomegaly (arrow) and a hyperechogenic blood clot (arrowhead).

Fig. 3.

Image showing increased periventricular echogenicity. Increased periventricular echogenicity accompanied by severe bilateral ventriculomegaly (both sides >30 mm).

Fig. 4.

Image demonstrating echogenic intraventricular clot. Ultrasound image showing an echogenic space-occupying lesion. A blood clot (1.2×0.6 cm) was noted in the ventricle with unilateral ventriculomegaly. Postnatal magnetic resonance imaging confirmed the presence of an old intraventricular hemorrhage.

Fig. 5.

Image showing an irregular shaped choroid plexus. The patient showed an irregular choroid plexus (arrowhead) with severe bilateral ventriculomegaly and increased periventricular echogenicity.

Cesarean delivery is generally indicated in cases of fetal ICH or IVH, depending on the progression of ultrasonography findings and etiology [7]. In this study, cesarean delivery was performed in 14 women (87.5%), and there were two (12.5%) vaginal deliveries. Among the women who underwent vaginal delivery, one fetus was delivered at 38 weeks of gestation with IVH grade 3, and the mother had a history of three previous vaginal deliveries. The other fetus was delivered at 39 weeks of gestation with IVH grade 4 and subdural hemorrhage, and the mother was nulliparous. Furthermore, the infant developed normally after birth due to the small size of the lesion.

Postnatal outcomes are summarized in Table 2. The median gestational age of the liveborn infants was 35 weeks and 2 days (range, 30 weeks 5 days to 39 weeks 5 days). Of the 16 liveborn infants, ten were delivered prematurely, and the average postnatal follow-up period was 40.1 months (range, 4-88). Ten infants (62.5%), including nine with ventriculoperitoneal shunt placement and one with intracranial surgery, were identified. The liveborn infant who underwent surgery underwent grid insertion and left occipital lobectomy. Motor impairment, including cerebral palsy, was present in 13 infants (81.2%), while six showed neurologic impairment. Developmental delay, assessed using the Bayley’s test, was observed in nine infants. Eleven infants (68.7%) had epilepsy. Characteristics of the individual data are presented in Table 3.

Table 2.

Summary of postnatal outcomes of fetal ICH

Table 3.

Summary of clinical data, ultrasonography findings, and MRI features of the 27 cases of fetal ICH

Discussion

Determining the etiology of fetal ICH has been reported to be an arduous task [6,9,11,16-18]. Maternal and fetal risk factors are also associated with fetal ICH. Maternal risk factors include vitamin K deficiency [19-21], history of trauma during pregnancy, drug exposure, especially aspirin and anticoagulation, infection, placental abruption, preeclampsia, and immune thrombocytopenia [7,14]. Fetal risk factors include fetal alloimmune thrombocytopenia, fetal thrombophilia, twin-to-twin transfusion syndrome, demise of a cotwin in monochorionic twins, and single gene mutations associated with specific genes. Mutation of the collagen type IV alpha 1 chain (COL4A1) gene is reportedly the most common cause of single gene mutation in fetal ICH or IVH [22,23]. In the present study, a case of COL4A1 gene mutation was identified. When a specific genetic factor is identified, clinicians must provide appropriate genetic counseling to parents for subsequent pregnancies through a trio test. If fetal and neonatal alloimmune thrombocytopenia is suspected, intravenous immunoglobulins can be considered [24]. However, some studies have suggested that intrauterine growth restriction may play a role in fetal ICH or IVH [1]; this was confirmed in only one fetus in our study. Specific factors could not be identified in 70% of the fetuses in this study. However, they were identified in four fetuses, which included fetal protein C deficiency, COL4A1, pyruvate dehydrogenase alpha 1, and centromere protein J single gene mutations, and single fetal death in monochorionic diamniotic twins.

Prenatal ultrasonography findings of fetal ICH can be diverse [5,7], and we systematically analyzed them across four categories: progressive ventriculomegaly, increased periventricular echogenicity, echogenic intraventricular clot, and irregular choroid plexus. Initially, fetal ICH manifests as a homogeneous echogenic zone without posterior shadowing in the ventricles or brain parenchyma, which is distinct from the choroid plexus. As the blood clot lyses, the ultrasonographic appearance becomes more heterogeneous, with the mass developing into an internal sonolucent core and an external echogenic rim. Blood clots often obstruct the cerebral aqueduct, which is a narrow channel that allows cerebrospinal fluid to flow between the third and fourth ventricles, resulting in bilateral ventriculomegaly [7,10]. Consequently, our study demonstrated that ventriculomegaly was the most common finding, with bilateral ventriculomegaly present in 75% of fetuses with high-grade (grade 3 and 4) ICH. Over time, blood within the ventricles increases echogenicity along the ventricular borders, facilitating the identification of progressive ventriculomegaly [7]. However, if the hemorrhagic lesion is too small or too recent, identifying fetal ICH or IVH in <4 hours may be challenging. This difficulty was evident in the 13th case in our study, which was the only mismatched case between prenatal ultrasonography and postnatal MRI because of its small size.

Fetal ICH was predominantly diagnosed in the third trimester in our study, which is consistent with the findings of previous studies [1,10,11,16,25]. Consequently, even if detailed sonography performed during the second trimester (between 18-22 weeks of gestation) is normal, clinicians should carefully monitor ultrasonography findings in the third trimester to avoid potential misdiagnosis of fetal IVH. Given the critical nature and need for long-term treatment of the neurological sequelae of fetal IVH, it is crucial to ensure that no neurological impairment occurs in the newborn during the delivery procedure, thereby preventing unwarranted blame on clinicians by parents [1].

Many researchers have attempted to introduce additional diagnostic modalities to detect fetal ICH at an early stage and more accurately. Magnetic resonance imaging exhibits high accuracy in detecting minor foci of germinal matrix hemorrhage [3,26-28] and could be beneficial in workups and prenatal counseling [7,10,11,14,18]. The European Neurosonography Working Group demonstrated that fetal MRI exhibited 5.4% more power than multiplanar neurosonography in identifying structural anomalies in the fetal brain [27]. However, MRI cannot serve as the primary imaging method for all pregnant women during first-line screening radiography, replacing ultrasonography to detect brain abnormalities, owing to cost factors. In our study, only four mothers underwent prenatal MRI, and the results did not differ from those obtained using ultrasonography. Additionally, it is questionable whether the outcomes would be affected, even if they were performed to influence the decisions of pregnant women and clinicians, given that abnormalities are often found at an advanced stage during ultrasonography.

With regard to postnatal outcomes (Table 2), all cases of neurosurgical intervention, motor impairment, neurological impairment, developmental delay, and epilepsy were observed in high-grade fetal ICH. Previous studies have reported a significant association between the grade and location of the hemorrhage and occurrence of severe neurologic complications [8,9,11,16,29,30]. Gupta et al. [25] showed that normal neurodevelopmental outcomes decreased in infants with higher grades of both unilateral and bilateral IVH, particularly in those with parenchymal involvement. Moreover, they demonstrated a better postnatal prognosis than our results, as their cases were evenly distributed across grades 1-4 in 37 liveborn infants [25]. Eldad et al. [31] revealed that groups with parenchymal hemorrhages exhibited a significantly higher incidence of severe neurological outcomes or neonatal death than those with non-parenchymal hemorrhages. However, our study did not observe a significant difference, owing to its small sample size.

Regarding the timing of diagnosis, the likelihood of requiring a ventriculoperitoneal shunt increased after 32 weeks of gestation, coinciding with the disappearance of the germinal matrix, as post-hemorrhagic dilatation persisted even after delivery [16]. Our study showed similar results: six of nine patients (66.7%) who required a ventriculoperitoneal shunt were diagnosed after 32 weeks of gestation. To improve our ability to prenatally predict postnatal neurological outcomes, we need to study various clinical prognostic factors and ultrasound findings and develop predictive markers.

Hemorrhagic lesions can undergo intrauterine progression or regression. Ghi et al. [8] reported that 72% of grade 1 and 2 hemorrhages were neurologically normal, whereas only 41% of grade 3 and 4 hemorrhages were considered normal. They asserted that complete regression of grade 1 and 2 hemorrhages in fetuses is possible, citing two cases of grade 2, where normal neonatal neurosonography was observed in 11 liveborn infants [8]. Conversely, previous studies have shown that intrauterine progression of hemorrhage from a lower to a higher grade is associated with worse neurodevelopmental outcomes [2,8,32]. Our findings revealed that only one of the 16 infants showed normal development (6.2%). Most liveborn infants exhibit high-grade IVH with intrauterine progression. We did not observe intrauterine regression in high-grade ICH cases; however, it was more common in low-grade ICH. Therefore, small transient hemorrhage cases in which regression might not be detected during antenatal ultrasonography are difficult to identify in our data.

Currently, in the context of low-grade fetal IVH, it has been reported that Cesarean section does not improve prognosis, and vaginal delivery does not worsen hemorrhage [6]. However, in high-grade cases, most pregnant women prefer Cesarean delivery because of the fear of worsening fetal brain damage during labor. Furthermore, in cases where delivery is necessitated by findings, such as progressive ventriculomegaly, a Cesarean section can be safely performed with the involvement of a pediatric neurosurgeon and neonatologist.

This study has several strengths. First, we compared prenatal and postnatal images and analyzed long-term followup outcomes (an average of 40.1 months) at a large single center in South Korea. Other studies reported an average infant/child developmental assessment age of 11.6 months [8] and 21.6 months [25]. Second, we used an extensive clinical database of all patients, including detailed genetic and hematological evaluations. Third, neurodevelopmental delay was quantitatively assessed using the Bayley scale along with clinical follow-ups at pediatric neurology and rehabilitation clinics. However, this study had a few limitations that should be considered. Low-grade cases have a low detection rate on ultrasonography; therefore, recruiting more low-grade samples is necessary. There may have been data collection bias, as only high grades were included. As this was a single-center study with a limited sample size of 27 participants, it is necessary to conduct further investigations to determine whether the results are significant after collecting data from additional patient groups. Furthermore, among the live births, preterm births accounted for more than half of the infants (10 [62.5%]), and preterm birth itself was a confounding factor for poor prognosis. However, most of the fetuses were delivered at late preterm, and given that brain maturity was almost complete at this time, the effect was considered small. Additionally, the rate of induced abortions could not be accurately estimated; therefore, the results may differ from those of other countries. Therefore, multicenter studies with larger sample sizes are warranted.

In conclusion, fetal ICH is a rare complication diagnosed during pregnancy, with subsequent fetal neurological sequelae or perinatal death. Fetal ICH is mainly diagnosed based on the findings of progressive ventriculomegaly on prenatal ultrasonography, and it tends to be identified as a higher grade. In most fetal ICH cases, the underlying etiology is unknown and the prognosis is usually poor, especially for fetuses with higher grades. Our study showed that fetuses diagnosed with prenatal ICH, especially those affected by a higher grade, may be at an increased risk of long-term neurodevelopmental sequelae. Comprehensive counseling should be provided to parents with fetal ICH to ensure that they receive appropriate antenatal counseling to make informed decisions and optimal postnatal care.

Article information Continued