Clinical Outcomes of Fetal Ventriculomegaly: A Retrospective Analysis from a Tertiary Referral Center

Introduction

Fetal ventriculomegaly (VM) is defined on prenatal ultrasound of the cerebral ventricles dilation, with a ventricular diameter measuring ≥10 mm during the second and third trimesters of pregnancy.^1,2 The prevalence of VM ranges from 0.3 to 2.0 per 1000 pregnancies.^1,3 Based on the ventricular diameter, VM can be classified into three categories: mild (≥10 mm and <12 mm), moderate (≥12 mm and <15 mm), and severe (≥15 mm).^4,5 VM can also be categorized as unilateral or bilateral, depending on whether one or both sides are affected. Additionally, VM is classified as isolated VM (IVM) or non-isolated VM (NIVM), depending on the presence or absence of other ultrasound abnormalities.

When VM is observed on prenatal ultrasonography, fetal magnetic resonance imaging (MRI) is recommended to evaluate for additional central nervous system (CNS) abnormalities. Periodic ultrasound examinations should be performed to assess the progression of the ventricular dilation, facilitating both prognostic assessment and optimal pregnancy-management strategies.

Many studies explored the etiology of VM and its impact on the pregnancy outcomes, with most studies being limited by small sample size.^6–9 The prognosis of fetal VM, especially for those with mild and moderate IVM, remains controversial. Reported neurodevelopmental outcomes range from normal to severe impairment. Several studies have shown that IVM is associated with an increased risk of neurodevelopmental disorders.^10–12 However, other studies and a recent systematic review suggest that the incidence of neurodevelopmental delay in fetuses with mild and moderate VM is comparable to that in the general population.^13,14 Several potential objective factors contribute to the discrepancy: variability in follow-up durations, heterogeneity in inclusion criteria and technical differences in imaging protocols. In the present study, we conducted a detailed analysis of fetal VM based on a large cohort. Data involving the additional imaging findings, intrauterine history and long-term postnatal (minimum 2-year) outcomes were analyzed, in order to provide more evidences for a better prenatal counseling and perinatal management.

Materials and Methods

Study Population

We retrospectively reviewed singleton pregnancies complicated by fetal VM that referred to our center between January 2014 and December 2018. All fetuses underwent detailed anatomical assessment by a fetal medicine specialist- physicians who had completed standardized prenatal diagnostic training and obtained the requisite technical service certification. Examinations were performed using high-resolution color Doppler ultrasound systems, including ACUSON S2000, PHILIPS IU22, EPIQ7, GE Voluson E8, and E10 models. For transabdominal ultrasound, 2D convex array probes with frequencies of 1–6 MHz or 2–9 MHz and 3D volume probes with frequencies of 4–8 MHz were used. In cases requiring targeted evaluation of the fetal nervous system, transvaginal endocavitary probes with frequencies of 5–9 MHz were employed. Fetal neurosonography was performed according to the ISUOG guidelines on fetal CNS ultrasound examination,¹⁵ utilizing four standard coronal planes and three sagittal planes. The width of the fetal lateral ventricles was measured at the level of the glomus of the choroid plexus, with calipers positioned perpendicular to the ventricular cavity and placed inside the echoes generated by the lateral walls. For cases of fetal VM, diagnosis was confirmed under double-blind conditions by at least two physicians, including at least one associate chief physician or above, with final diagnosis requiring consensus.

Fetuses with confirmed genetic abnormalities identified via conventional karyotyping and/or SNP array analysis, as well as those with congenital cytomegalovirus (CMV) infection, were excluded from the study.

As a result, a total of 720 pregnancies were enrolled. The cohort was grouped as follows:

1. Isolated ventriculomegaly (IVM) and Non-isolated ventriculomegaly (NIVM):

Ventriculomegaly without any associated intracranial or extracranial abnormalities was classified as isolated ventriculomegaly (IVM). Cases with additional abnormalities were categorized as non-isolated ventriculomegaly (NIVM). The associated abnormalities in NIVM were further classified into three categories: central nervous system (CNS) abnormalities, extracranial abnormalities, and soft markers.Soft markers included: choroid plexus cysts, widened posterior fossa cistern, mild tricuspid or mitral regurgitation, long bones shorter than gestational age, hyperechoic bowel, renal pelvis separation, intracardiac echogenic focus, thickened nuchal fold (NF), single umbilical artery, mild pericardial effusion, ascites, umbilical cord cyst, and nasal bone hypoplasia.

2. Unilateral ventriculomegaly (UVM) and bilateral ventriculomegaly (BVM) refer to VM involving enlargement of the lateral ventricles on single and both sides respectively.

3. Degree of VM: Based on the lateral ventricular at initial scan, VM was graded as mild (10–11.9 mm), moderate (12–14.9 mm), and severe (≥15 mm). Intrauterine progression was defined based on the comparison of ventricular width at the initial diagnosis with that in the last prenatal ultrasound found: regressive (remised ≥2mm), stable (changing <2mm), and progressive (increased ≥2mm).

Follow-Up Assessments

After the initial detection of VM, MRI was recommended for all affected fetuses. Serial ultrasound scanning was performed at interval of 2–4 weeks until delivery to monitor the progression of ventricular dilation.

Postnatal assessments focusing on motor, language, and intellectual development were collected via medical records or telephone interviews with the parents. The follow-up time varied from 2 to 8 years after birth. Adverse postnatal outcomes were defined as the presence of developmental retardation, including cerebral palsy, intellectual retardation, motor dysfunction, speech difficulties, neonatal death, or infant death.

Postnatal cranial ultrasound examinations were performed. Abnormal findings included ventriculomegaly, agenesis of the corpus callosum, encephalomalacia, leukodystrophy, intracranial hemorrhage, microcephaly, and cerebellar atrophy.

The study was approved by the Ethics Committee of Fujian Provincial Maternity and Child Hospital. Written informed consent was obtained from each patient to participate in the study.

Statistical Analysis

Statistical analysis was performed using SPSS Statistics software v26.0 (IBM SPSS, Armonk, NY, USA). Comparisons between groups were conducted using the chi-square test or Fisher’s exact test, Wilcoxon test, and logistic regression. A p-value of <0.05 was considered statistically significant.

Results

Descriptive Results

The cohort comprised 451 cases of IVM and 269 cases of NIVM. The initial diagnosis was made after 24 gestational weeks in 78.9% (n=568/720) of cases. The frequency of NIVM in the severe, moderate, and mild groups were 49.3% (38/77), 44.3% (39/88), and 34.5% (192/555), respectively. NIVM was found to have a statistically significant positive correlation with increasing ventricular width (p = 0.005).

Compared to UVM, BVM was more frequently diagnosed before 24 gestational weeks (χ²=5.208, p<0.05). BVM was also significantly associated with moderate and severe VM (χ²=76.687, p<0.0001), and more often presented as NIVM (χ²=31.150, p<0.001). The details of the characteristics are summarized in Table 1.

Table 1 Details of the Pregnancy Characteristics

A total of 405 cases underwent fetal cranial MRI examinations, which revealed additional abnormal cranial findings in 5 cases (1.23%). These included pachygyria (n = 3), gray matter heterotopia (n = 1), and Dandy–Walker syndrome (n = 1).

Prognostic Analysis and Risk Factors Associated with IVM

Pregnancy outcomes were available in 370 (82.0%, 370/451) IVM fetuses. Excluding one case of intrauterine fetal demise, 33 (8.9%) cases were terminated due to severe VM (n=27) and maternal preeclampsia (n=6).

The natural intrauterine course of IVM differed among subgroups. Among mild cases, the proportions of intrauterine regression, stability and progression were 69.5% (205/295), 28.55% (84/295) and 2.0% (6/295), respectively. For moderate cases, the proportions were 48.6% (18/37), 45.9% (17/37), and 5.5% (2/37), respectively. In severe cases, 0% (0/4) regressed, 75% (3/4) remained stable, and 25.5% (1/4) progressed. Wilcoxon test showed that the intrauterine course of IVM was significantly correlated with ventricular width (p=0.0003), rather than the laterality of UVM or BVM (p=0.5498).

A total of 336 IVM fetuses resulted in livebirths, with 21 cases (6.25%) developing adverse postnatal outcomes. Among them, 20 cases presented as developmental retardation, and the rest one with progressive mild VM ended with neonatal death due to respiratory failure. Postnatal cranial imaging abnormalities were documented in at least 57.1% (12/21) of the unfavorable outcome cases. The intrauterine evolution of IVM are presented in Table 2. Details of the adverse outcomes in IVM cases are presented in Table 3.

Table 2 The Intrauterine Evolution of IVM

Table 3 Details of the Adverse Outcomes of Pregnancies with IVM

Both univariate and multivariate logistic regression was performed to assess risk factors associated with adverse outcomes in IVM, including ventricular width, unilateral (UVM) or bilateral ventriculomegaly (BVM), and intrauterine progression. The results showed no statistically significant difference in risk between UVM and BVM (OR 2.462, 95% CI 0.978–6.196). Severe IVM was significantly associated with a higher risk of adverse outcomes compared to mild IVM (OR 60.214, 95% CI 5.882–616.37). Moderate IVM did not show a statistically higher risk of adverse outcomes compared to mild (OR 2.433, 95% CI 0.756–7.825). In terms of intrauterine course, fetuses with stable or progressing IVM increased the risk of adverse outcomes compared to those with intrauterine regression (OR 5.687, 95% CI: 1.948–16.602). The risk was even higher for fetuses with progressive ventricular dilation (OR 34.88, 95% CI: 7.143–170.316). Detailed natural in utero history and pregnancy outcomes of IVM are presented in Table 4.

Table 4 Logistic Regression Analysis for IVM Ultrasound Results and Pregnancy Outcomes

Additional Imaging Findings and Outcomes of NIVM

In our cohort of 269 pregnancies with non-isolated ventriculomegaly (NIVM), additional soft markers were the most prevalent associated feature, observed in 47.2% of cases (127/269). CNS malformations were observed in 64 cases (23.8%), primarily involving corpus callosum dysplasia, intracranial hemorrhage, cerebellar dysplasia, Dandy-Walker malformation, and spina bifida.

Pregnancy outcomes were available for 222 out of 269 NIVM fetuses (82.5%), representing 82.5% of the cohort. Among these, 40.1% (89/222) of pregnancies were electively terminated, a rate significantly higher than that observed in the isolated ventriculomegaly (IVM) group (8.9%, 33/370; χ² = 78.829, p < 0.05). The highest rate of termination of pregnancy (TOP) was observed in NIVM with CNS abnormalities (82.8%, 48/58), followed by those with extracranial malformations (61.1%, 33/54). In contrast, the TOP rate in cases with soft markers alone was only 7.2% (8/110).

Among the 133 NIVM cases with complete follow-up (birth cohort), the natural intrauterine course varied significantly across severity subgroups (p < 0.001, χ²-test). In mild cases (n = 112), VM regressed in 67.9% (76), remained stable in 26.8% (30), and progressed in 5.3% (6). In moderate cases (n = 18), regression occurred in 33.3% (6), stability in 55.6% (10), and progression in 11.1% (2). All severe cases (n = 3) remained stable throughout gestation. Overall, regression was most common in mild cases, while stability was the predominant course in moderate and severe cases.

In the 133 ongoing pregnancies, 84.2% (112/133) showed a favorable prognosis after birth, while 15.8% (21/133) developed adverse postnatal outcomes, higher than the 6.25% (21/336) observed in the IVM group. The concurrent imaging features significantly impacted the prognosis: a favorable prognosis was observed in 92.2% (94/102) of cases accompanied by soft markers, 61.9% (13/21) of cases with other extracranial aberrations, and only 50% (5/10) of cases with co-occurring CNS abnormalities.

Among the 21 cases with poor postnatal outcomes, one-third (n = 7) resulted in premature death, and two-thirds (n = 14) exhibited developmental delays. Postnatal cranial ultrasonography revealed persistent abnormal intracranial images in 66.7% (14/21) of these cases. The most common findings were VM (n=10), followed by agenesis of the corpus callosum (n=2), encephalomalacia (n= 2), microcephaly, and cerebellar atrophy. Detailed information on adverse outcomes associated with NIVM is presented in Table 5.

Table 5 Details of the Adverse Outcomes of Pregnancies with NIVM

Discussion

Ventriculomegaly (VM) is a common finding during prenatal ultrasound examination. The clinical outcomes of VM are highly dependent on the various potential pathological factors. Even in cases of IVM, the prognosis remains contentious. Understanding the course of VM is essential to support informed pregnancy decision-making. This retrospective study enrolled 720 fetuses diagnosed with VM over five years. A detailed analysis was performed, considering different types of VM, the degree of ventricular dilation, intrauterine progression, pregnancy outcomes, and postnatal growth.

In accordance with a previous report,^9,16 mild VM was more frequently observed in IVM fetuses. In our study, the majority of IVM cases (80.5%) manifested mild ventricular enlargement, while a small proportion (8.6%) exhibited severe enlargement. The incidence of neurodevelopmental delay in mild and moderate IVM was similar to that of the general population.¹³ Most parents opted to continue their pregnancies in cases of mild and moderate IVM. In our study, only 7.3% of IVM cases resulted in pregnancy termination, mainly due to progressive severe VM.

Among the 336 cases with documented in utero courses, Wilcoxon analysis showed that the natural progression of IVM was significantly associated with ventricular width, rather than the laterality of ventricular involvement. Most mild IVM cases demonstrated intrauterine regression, whereas moderate and severe cases tended to exhibit stable or progressive courses. Current evidence on the intrauterine outcomes of moderate IVM remains limited due to small sample sizes in existing studies. In our cohort, 51% (19/37) of moderate IVM cases remained stable or progressed, which is slightly higher than the rates reported in previous studies: 41% (14/34) in a prospective study¹⁷ and 45% (33/51) in a multicenter study.¹⁸ However, a prospective study by Wei-Xi Hu et al’s¹⁹ reported a 100% rate (29/29) of stability or progression in cases of IVM. Notably, all available studies, including our own, are constrained by small sample sizes, underscoring the need for larger, prospective investigations to better characterize the natural history of moderate IVM.

It is well recognized that the severity of prenatal IVM can influence postnatal outcomes. In our study, 95.3% of mild, 89.2% of moderate, and 25% of severe IVM cases had normal development between the age of 2 to 6 years, largely consistent with the data reported by The Society for Maternal-Fetal Medicine (SMFM).²⁰ However, the prognostic evaluation value of intrauterine progression and the distinction between unilateral (IUVM) and bilateral IVM (IBVM) remains unclear.

Previous studies have found that progression and bilateral VM are associated with unfavorable outcomes,^11,17,21 while Falip et al reported that postnatal clinical outcomes were independent of whether the VM was unilateral or bilateral and found no difference between stable, regressive, and resolved IVM.²² In our study, further logistic regression analysis indicated that postnatal outcomes were independent of whether the VM was unilateral or bilateral. Severe VM did increase the risk of unfavorable outcomes, but there was no significant difference between mild and moderate IVM groups. Notably, our cohort included only 49 moderate (10.9%) and 39 severe (8.6%) VM cases. As extensively discussed, this limited representation of higher-grade VM necessitates validation through large-scale multicenter studies to establish reliable prognostic benchmarks highlighting the need for expanded cohorts to strengthen prognostic conclusions.

When compared with utero resolution during follow-up, both intrauterine stable and progressive were associated with a higher risk of adverse outcomes,with odds ratios of 34.88 and 5.69, respectively, which aligns with findings from a published systematic review.²¹ Our study indicates that severe IVM and intrauterine stability or progression are significant risk factors for unfavorable outcomes. Therefore, continuous intrauterine monitoring is required for accurate prognosis evaluation.

NIVM was observed in approximately 37.4% (269/720) of the cohort, similar to previous studies.^9,16 As indicated by earlier data,^9,23 the occurrence of NIVM increased significantly with the width of the ventricle. Furthermore, our study revealed that additional ultrasound findings were more likely to be observed in BVM cases. Therefore, detailed monitoring should be undertaken for BVM and severe VM cases. The presence of additional imaging abnormalities influenced pregnancy decision-making. In the NIVM group, 39.5% of pregnancies were terminated, a significantly higher rate than in the IVM group (χ² = 78.829, p < 0.05). Central nervous system (CNS) abnormalities involving agenesis of the corpus callosum, intracranial hemorrhage, cerebellar dysplasia, Dandy-Walker malformation, and spina bifida, were the second most common finding in NIVM cases, consistent with previous studies.^9,24,25 Due to concerns about co-existent dysfunction, NIVM accompanied by CNS abnormalities led to the highest proportion of terminations of pregnancy (TOP), up to 75% (48/64), followed by pregnancies with extracranial abnormalities (58.9%, 33/56), for cases with soft markers alone, was 7.2% (8/110).

Features of coexistent imaging abnormalities also significantly increased the risk of adverse postnatal outcomes, such as mortality and developmental delay.^26,27 In the present study, 15.6% of the NIVM cases developed adverse outcomes, significantly higher 6.25% observed in the IVM group. Cases with additional CNS abnormalities had the highest risk of adverse outcomes, followed by those with other imaging abnormalities and soft markers. As stated by Beeghly M., the prognosis primarily depends on the associated abnormalities rather than the degree of ventricular dilation.²⁸ A recent retrospective cohort study revealed that the ganglionic eminence (GE) area, measured via MRI, was significantly larger in fetuses with IVM compared to healthy controls. This finding suggests that GE enlargement may serve as a potential imaging biomarker for predicting postnatal neurodevelopmental impairments, including speech delay/difficulty and attention-deficit/hyperactivity disorder (ADHD).²⁹ Therefore, when VM is observed, careful prenatal imaging is of great importance for evaluating the prognosis.

The genetic background of fetuses with VM should be carefully considered. In this study, all cases underwent conventional karyotyping and/or SNP-array analysis, and major chromosomal abnormalities were excluded. However, smaller copy number variants and single-gene disorders may still have been missed. These undetected genetic factors could contribute to the development of fetal VM and may also affect postnatal neurodevelopmental outcomes. Previous studies have suggested that monogenic disorders account for a notable proportion of cases.³⁰ Therefore, in selected cases, more comprehensive genetic testing, such as exome sequencing, may be helpful for clarifying the underlying etiology and improving the accuracy of prognostic counseling.

It is well recognized that individuals with postnatal cranial abnormalities often develop poor prognoses. In the present study, at least 66.7% of NIVM cases and 57.1% of IVM cases with adverse outcomes had sustained intracranial imaging anomalies. Additionally, Lee²⁷ reported that more than half of cases with VM and developmental delays or functional disorders had additional abnormalities that were not detected prenatally. In this study, 42 cases had adverse postnatal outcomes, with 11 cases showing newly discovered abnormalities after birth. Among these, seven were intracranial abnormalities, all presenting as developmental retardation. These included encephalomalacia (n=3), leukodystrophy (n=1), intracranial hemorrhage (n=1), cerebellar atrophy (n=1), and microcephaly (n=1). There were also two cases of neonatal death due to respiratory failure. One case was diagnosed with esophageal atresia after birth; although surgical intervention was provided in time, the child still presented with developmental retardation and cerebral palsy. This emphasizes that VM may be a marker of brain dysplasia or brain lesions, underscoring the importance of postnatal examinations for fetuses with VM. These examinations aim to detect late-onset manifestations and provide timely medical care to improve prognosis.

This study has several limitations. First, its retrospective design limits control over data completeness and consistency. Although referrals to tertiary fetal medicine centers have increased annually, most pregnancies in our cohort underwent routine prenatal ultrasound monitoring alone. Only 56.3% (405/720) received detailed fetal CNS MRI scans, primarily due to financial constraints or advanced gestational age limitations, which may have resulted in undetected subtle cerebral abnormalities. Nevertheless, our analysis of 405 fetuses with VM provides large-scale empirical evidence for this clinical domain. Second, due to limitations of prenatal imaging and the progressive nature of certain abnormalities, such as intracranial hemorrhage, encephalomalacia, and cerebral atrophy, it is difficult to confirm truly IVM before birth. Third, postnatal outcomes were obtained from medical records or telephone interviews, with 17.9% of cases lost to follow-up. Furthermore, adverse outcomes may have been underreported due to social stigma or parental unawareness of mild neurological issues, potentially underestimating the true incidence.

Conclusion

Severe VM and intrauterine progression or stability are risk factors for unfavorable outcomes. Features of coexistent imaging abnormalities greatly influence decision-making and pregnancy outcomes. Therefore, consecutive imaging scans should be provided to fetuses with VM, both prenatally and postnatally.