Pregnancy outcomes and genetic analysis for fetal ventriculomegaly

Introduction: Fetal ventriculomegaly (VM) is associated with neurodevelopmental disorders, partly caused by genetic factor. Methods: To systematically investigate the genetic etiology of fetal VM and related pregnancy outcomes in different subgroups: IVM (isolated VM) and NIVM (non-isolated VM); unilateral and bilateral VM; mild, moderate, and severe VM, a retrospective study including 131 fetuses with VM was carried out from April 2017 to August 2022. Results: 82 cases underwent amniocentesis or cordocentesis, of whom 8 cases (9.8%) were found chromosomal abnormalities by karyotyping. Meanwhile, additional 8 cases (15.7%) with copy number variations (CNVs) were detected by copy number variation sequencing (CNV-seq). The detection rate (DR) of chromosomal abnormalities was higher in NIVM, bilateral VM and severe VM groups. And CNVs frequently occurred in NIVM, bilateral VM and moderate VM groups. In the NIVM group, the incidence of chromosomal aberrations and CNVs in multiple system anomalies (19.0%, 35.7%) was higher than that in single system anomalies (10.0%, 21.1%). After dynamic ultrasound follow-up, 124 cases were available in our cohort. 12 cases were further found other structural abnormalities, and lateral ventricular width was found increased in 8 cases and decreased in 15 cases. Meanwhile, 82 cases underwent fetal brain MRI, 10 cases of brain lesions and 11 cases of progression were additionally identified. With the involvement of a multidisciplinary team, 45 cases opted for termination of pregnancy (TOP) and 79 cases were delivered with live births. One infant death and one with developmental retardation were finally found during postnatal follow-ups. Discussion: CNV-seq combined with karyotyping could effectively improve the diagnostic rate in fetuses with VM. Meanwhile, dynamic ultrasound screening and multidisciplinary evaluation are also essential for assessing the possible outcomes of fetuses with VM.


Introduction
Fetal lateral ventricular width, usually 7.6 ± 0.6 mm from 15 to 40 weeks of gestation, is a routine examination parameter in prenatal ultrasound screening.Fetal ventriculomegaly (VM) is defined as the width of one or both sides of the lateral ventricles ≥10 mm according to ultrasound diagnostic criteria (Yagel and Valsky, 2021), with a prevalence of 0.3‰-1.5‰(Myrianthopoulos, 1977).Fetal VM can be classified into different subtypes: mild (10-12 mm), moderate (13-15 mm), severe (>15 mm) VM (Griffiths et al., 2010); unilateral and bilateral VM; isolated VM (IVM) and nonisolated VM (NIVM) indicating plus other multi-system abnormalities.It has been reported that 7.9% fetuses with mild IVM may have neurodevelopmental retardation, and the prevalence of adverse outcomes may increase to 38%-90% in fetuses with severe IVM or hydrocephalus (Pagani et al., 2014;Letouzey et al., 2017;Thorup et al., 2019).Therefore, it is pivotal for clinicians to predict the possible prognosis of the affected fetuses.
The outcomes of fetal VM can be influenced by a variety of factors such as chromosomal abnormality, viral infection and progression in utero.Hence, it makes prenatal counseling more challenging for clinicians.The Society for Maternal-Fetal Medicine (SMFM) recommends that clinicians should perform a comprehensive evaluation for fetuses with VM, including detailed fetal structures, invasive prenatal diagnosis for genetic etiology, and a workup for fetal infections (Fox et al., 2018).Compared with traditional karyotyping, copy number variation sequencing (CNVseq) based on next-generation sequencing has outstanding advantages of high resolution, high throughput and whole genome coverage.In recent years, it has been widely applied in the field of prenatal diagnosis (Zhao and Fu, 2019).In this study, we focused on analyzing the value of genetic analysis, dynamic ultrasound screening, and multidisciplinary evaluation for the purpose of improving perinatal management in fetuses with different types of VM.

Study subjects
In this retrospective study, 131 fetuses diagnosed with VM by prenatal ultrasound were enrolled at the prenatal diagnosis center of Xuzhou Central Hospital from April 2017 to August 2022.Patients with underlying diseases or high-risk factors were ruled out, such as hypertension, immune diseases, trauma, twin or multiple pregnancy, etc.All pregnant women received prenatal counseling and provided written informed consents for further examinations.82 cases underwent invasive prenatal diagnosis for karyotyping, and 51 cases simultaneously underwent CNV-seq (Figure 1).124 cases were further evaluated by dynamic ultrasound examination and 82 cases further underwent fetal brain MRI (Figure 2).Our study has been approved by the ethics committee of Xuzhou Central Hospital (XZXY-LJ-2017-0626-003).

Ultrasound examination and MRI
Prenatal ultrasound scanning was performed by ultrasound physicians using a 2.5~5 MHz transabdominal transducer and color doppler ultrasound instrument (GE Voluson E8, American) according to the guidelines of the International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) (Salomon et al., 2022).Fetal VM was diagnosed as the width of lateral ventricles ≥10 mm at the level of the choroid plexus perpendicular to the ventricular cavity (Yagel and Valsky, 2021).
Fetal MRI was performed by experienced operators using a 1.5T scanner (Infineon, Philips Medical Systems).The transverse diameter of the trigones of both lateral ventricles was measured and any additional central nervous system (CNS) abnormalities should also be carefully examined (Fox et al., 2018).The results were then compared with ultrasound findings.

Karyotyping and CNV-seq
51 amniotic fluid samples and 31 cord blood samples were collected by ultrasound-guided amniocentesis or cordocentesis.These samples were then subjected to the following procedures:  in situ cell culture, harvesting, G-banding, and result judgment.All karyotyping reports were interpreted in line with the International System for Human Cytogenetic Nomenclature at the level of 300-400 bands (ISCN 2016).
Fetal genomic DNA extraction and library construction were conducted following laboratory standard operating procedures.The following massive whole-genome sequencing (0.1x) in parallel was carried out using the NextSeqCN500 platform (Illumina, United States).50 ng fetal DNA from each prenatal sample was used for CNV-seq.The sequencing data were analyzed by GISTIC 2.0 with reference to the human reference genome sequence version GRCh37/hg19, the standards of the American College of Medical Genetics (ACMG) (Richards et al., 2015;Riggs et al., 2020), and related databases such as Database of Genomic Variants (DGV).For the cases with abnormal CNVs, parental peripheral blood samples were also suggested to further verify whether it was inherited or de novo.

Follow-up of pregnancy outcomes
An experienced multidisciplinary team (MDT) from many departments, including ultrasound, magnetic resonance imaging (MRI), prenatal diagnosis and neurology, worked together to assess the possible prognosis of fetuses with VM.Pregnancy outcomes were recorded by postpartum telephone and electronic medical record system.Neonates and infants were followed up from 6 months to 2 years after birth and comprehensively assessed in terms of neurodevelopmental outcomes by experienced pediatricians.

Statistical analyses
The data were analyzed by SPSS software (version 26.0, IBM, Armonk, NY, United States).Continuous variables were expressed as mean ± standard deviation (SD), and independent sample t-test was used for comparison between groups.Enumeration data were expressed as frequency, and chi-square test or Fisher's exact test was used to compare the rates between groups (p < 0.05 was regarded as statistically significant).

Correlation between chromosomal aberrations and indicators for NIVM
In the NIVM group, 51 cases underwent the molecular genetic test, including 30 single system anomalies and 21 multiple system anomalies (Table 5).Although the DR of both chromosomal abnormalities and CNVs were higher in multi-system anomalies (19.0%, 35.7%) than in single system anomalies (10.0%, 21.1%), no significant differences were observed between the two groups (p > 0.05).It suggested that chromosomal abnormalities are the ones that increase the risk of NIVM.What's more, cardiovascular system abnormalities were mostly found in multi-system anomalies.25.0% (3/12) abnormal karyotypes and 25.0% (2/8) CNVs were revealed when combined with cardiovascular abnormalities.

Follow-up outcomes
After dynamic ultrasound screening, 124 fetuses with VM were available in the cohort (Figure 2).12 cases were further found other structural abnormalities, and lateral ventricular width was also found increased in 8 cases and decreased in 15 cases.Meanwhile, 82 cases underwent fetal brain MRI, and 21 were inconsistent with ultrasound findings, including 10 brain lesions with poor prognosis (agenesis of the corpus callosum (ACC) (n = 8), Dandy-Walker malformation (DWM) (n = 1), cerebral cortical growth retardation (n = 1)) and 11 cases of lateral ventricular width progression.It suggested that fetal brain MRI can improve the diagnosis of additional CNS anomalies and monitor the degree of VM.

Discussion
Fetal VM may increase the risk of neurodevelopmental abnormalities after birth, such as autism, schizophrenia, and epilepsy (Chang et al., 2020;Lok et al., 2021;Chen et al., 2022).Our study highlighted the importance of genetic etiology in fetal VM, and indicated that karyotyping together with CNV-seq can effectively improve the detection rate of fetuses with VM.
Previous studies have shown that chromosomal abnormalities occurred in 5%-20% of fetuses with IVM, and pathogenic CNVs are also common causes accounting for about 4.20% (Pagani et al., 2014;Chang et al., 2020;Wang et al., 2020).In our study, 9.8% cases with VM were found chromosomal abnormalities by karyotyping, meanwhile, additional 15.7% CNVs were detected by CNV-seq (Table 2; Table 4).Some are recognized as microdeletion or microduplication syndromes, including BBSOAS, 1p36 microdeletion syndrome, and 16p11.2deletion syndrome.The DR of chromosomal aberrations in fetuses with VM was higher by CNV-seq than by traditional karyotyping, regardless of single or multi-system anomalies.It suggested that CNV-seq can effectively  improve the DR of fetuses with VM over karyotyping (Wang et al., 2021).It is not hard to see that fetuses with NIVM, bilateral VM or moderate-tosevere VM have a higher risk of chromosomal abnormalities or CNVs.Therefore, more attention should be paid to the above fetuses, to effectively avoid birth defects with genetic factors.In fetuses with NIVM, the DR of chromosomal aberrations in multi-system anomalies was higher than that in single system anomalies (Table 5), which is consistent with the findings of Chang (Chang et al., 2020).It suggested that chromosomal abnormalities are the ones that increase the risk of NIVM.Thus, when fetal VM combined with multiple malformations, especially cardiovascular system, we should be more aware of the possibility of chromosomal aberrations.From our point of view, karyotyping combined with CNV-seq should be recommended to rule out genetic abnormalities, especially for fetuses with NIVM, bilateral VM or moderate-to-severe VM.It is often difficult to reveal fetal phenotypes at a certain point in time by prenatal ultrasound due to the limitations of the fetal position, gestational age, and amniotic fluid volume.In our study, during dynamic ultrasound screening, additional structural abnormalities in 12 patients and an increase or regression of lateral ventricular width in 23 patients were revealed.Therefore, dynamic ultrasound follow-up is of great significance to avoid under/overdiagnosis and assess the possible prognosis to facilitate better management for the affected fetuses.In addition, 10 cases of fetal brain lesions with poor prognosis were detected by MRI but missed by ultrasound, and 11 cases with lateral ventricular width progression were also detected.Therefore, fetal brain MRI can act as an effective complementary tool to confirm the diagnosis and reveal additional CNS anomalies.
Based on the above strategies, multidisciplinary consultation should be recommended for couples on whether to continue or terminate the pregnancy, especially in the late second or third trimesters, which might comprehensively provide better perinatal management for fetuses with VM.Regarding pregnancy outcomes, 79 cases continued the pregnancy with live births including 5 cases diagnosed with genetic etiology (a low proportion of sex chromosomes mosaicism (20% 47, XXY and 80% 46,XY), inversion of chromosome 9, microdeletion in 1p36.33p36.22inherited from the mother, microduplications in Xp22.11 and 11q25).From our experience, the involvement of MDT, as well as parental persistence, contributed to the continuation of pregnancy.Postnatal follow-ups from 6 months to 2 years showed 1 neonatal death at 6 months, and 1 speech and motor retardation at 2 years old.A large population-based study explained that even in apparent IVM, there was a significant risk of postnatal additional structural anomalies and neonatal death (Hannon et al., 2012).Hence, even for mild, isolated VM, we cannot leave out the possibility of poor results.It is pivotal for MDT to carefully evaluate fetuses with different types of VM both prenatally and postpartum.In our data, TOP was chosen in all cases with severe VM.When VM is the only abnormal finding, the parents make decisions mostly based on the severity of VM, as the poor outcomes increase with the degree of VM aggravated.Whether the degree of VM can act as an independent risk factor to judge fetal prognosis needs to be further explored through larger sample sizes.It is also the responsibility of MDT to carefully make a recommendation for TOP after comprehensive consideration.
There are still some limitations in our study that need to be improved.First, fetal brain MRI has been proven as a useful tool to identify additional CNS anomalies, such as ACC associated with poor outcomes (Wang et al., 2020), but not all cases in our study received the examination.Secondly, tests for pathogenic microorganisms such as cytomegalovirus and toxoplasmosis should be performed to rule out infectious factors (SMFM et al., 2020).Last but not least, the follow-up time and sample size were not sufficient to find neurodevelopmental abnormalities in the children.Longer follow-ups and more cases will be enrolled in the future to explore the outcomes of fetal VM.
In short, CNV-seq could effectively improve the diagnostic rate of fetal VM, over karyotyping.Meanwhile, dynamic ultrasound screening and multidisciplinary evaluation are also essential for better management of fetuses with VM.

FIGURE 1
FIGURE 1Flow chart of clinical management procedures.

FIGURE 2
FIGURE 2Flow chart of imaging diagnosis pathway.

FIGURE 3
FIGURE 3Flow chart of follow-up outcomes.

TABLE 1
Basic characteristics of study subjects.

TABLE 2
Chromosomal abnormalities in fetal VM detected by karyotyping.

TABLE 3
Abnormal chromosomes, CNVs and pregnancy outcomes in 16 fetuses with VM.

TABLE 4
CNVs in fetal VM detected by CNV-seq.