Correlating Prenatal Imaging Findings of Fetal Ventriculomegaly with the Need for Surgical Intervention in the First 3 Months after Birth

Machine learning in prenatal MRI predicts postnatal ventricular abnormalities in fetuses with isolated ventriculomegaly

OBJECTIVES

To evaluate the intracranial structures and brain parenchyma radiomics surrounding the occipital horn of the lateral ventricle in normal fetuses (NFs) and fetuses with ventriculomegaly (FVs), as well as to predict postnatally enlarged lateral ventricle alterations in FVs.

METHODS

Between January 2014 and August 2023, 141 NFs and 101 FVs underwent 1.5 T balanced steady-state free precession (BSSFP), including 68 FVs with resolved lateral ventricles (FVM-resolved) and 33 FVs with stable lateral ventricles (FVM-stable). Demographic data and intracranial structures were analyzed. To predict the enlarged ventricle alterations of FVs postnatally, logistic regression models with 5-fold cross-validation were developed based on lateral ventricle morphology, blended-cortical or/and subcortical radiomics characteristics. Validation of the models' performance was conducted using the receiver operating characteristic curve (ROC), calibration curve, and decision curve analysis (DCA).

RESULTS

Significant alterations in cerebral structures were observed between NFs and FVs (p < 0.05), excluding the maximum frontal horn diameter (FD). However, there was no notable distinction between the FVM-resolved and FVM-stable groups (all p > 0.05). Based on subcortical-radiomics on the aberrant sides of FVs, this approach exhibited high efficacy in distinguishing NFs from FVs in the training/validation set, yielding an impressive AUC of 1/0.992. With an AUC value of 0.822/0.743 in the training/validation set, the Subcortical-radiomics model demonstrated its ability to predict lateral ventricle alterations in FVs, which had the greatest predictive advantages indicated by DCA.

CONCLUSIONS

Microstructural alterations in subcortical parenchyma associated with ventriculomegaly can serve as predictive indicators for postnatal lateral ventricle variations in FVs.

CLINICAL RELEVANCE STATEMENT

It is critical to gain pertinent information from a solitary fetal MRI to anticipate postnatal lateral ventricle alterations in fetuses with ventriculomegaly. This approach holds the potential to diminish the necessity for recurrent prenatal ultrasound or MRI examinations.

KEY POINTS

Fetal ventriculomegaly is a dynamic condition that affects postnatal neurodevelopment. Machine learning and subcortical-radiomics can predict postnatal alterations in the lateral ventricle. Machine learning, applied to single-fetal MRI, might reduce required antenatal testing.

Role of fetal MRI to diagnose abnormal cerebral ventricular system and associated fetal brain anomalies

Background

Abnormal cerebral ventricular system is one of the most common clinical indications for fetal MRI, mainly to detect other associated abnormalities that can be occult on prenatal ultrasonography. Although using ultrasound can identify most anomalies, MRI is known to be superior in identifying CNS anomalies as it has higher contrast resolution for brain parenchymal assessment added to the fact of being less affected by fetal positioning, oligohydramnios, maternal obesity, and reverberation artifacts. Fetal brain ventriculomegaly is defined as atrial width of > 10 mm on sonography, measured in the axial plane, at the level of the frontal horns, cavum septi pellucidi, and glomus of the choroid plexus perpendicular to the long axis of the lateral ventricle. One of the most important factors determining the fetal neurological outcome is the presence and severity of additional CNS anomalies that are better clarified by MRI. The aim of this study is to establish the role of fetal MRI in detecting the association between abnormal cerebral ventricular system and other CNS anomalies, correlation with the severity of ventriculomegaly and ventricle asymmetry.

Results

Thirty pregnancies with fetal brain abnormal ventricular system were included in this study, 5 cases with isolated corpus callosum (CC) agenesis (16.666%); 2 cases with cystic lesions [one interhemispheric and the other dorsal] (6.666%) both associated with CC agenesis; 1 case with alobar holoprosencephaly (3.333%) associated with CC agenesis; 2 cases with semi-lobar holoprosencephaly (6.666%) [associated with CC agenesis and one of them is also associated with lissencephaly]; 3 cases with Dandy–Walker Malformation (DWM) (10%) [2 isolated and 1 associated with CC agenesis]; 3 cases with Dandy–Walker Variants (DWV) (10%) [1 isolated and 2 associated with CC agenesis]; 2 cases with Joubert syndrome (6.666%); 1 case isolated lissencephaly (3.333%); 4 cases of obstructive ventriculomegaly (13.333%) (1 of which associated with CC agenesis); 1 case of Arnold Chiari malformation type II (3.333%) associated with CC agenesis; 2 cases with meningoceles (6.666%) (occipital and parieto-occipital); 1 case with Mega cisterna Magna (3.333%); 1 case with anencephaly (3.333%); 1 case with right hemimegalencephaly (3.333%) (associated with frontal meningocele and CC agenesis); and 1 case with grade IV germinal matrix hemorrhage (3.333%). The pregnancies resulted in 20 births (66.66%), 2 died directly after birth (6.66%), 5 terminations (16.66%), and 3 intrauterine fetal deaths (IUFD) (10%). We found that the frequency of associated CNS anomalies was strongly related to the width of the ventricle. The association between CNS findings and ventricle width was particularly evident in sever ventriculomegaly. The greater the width of the ventricular system, the more the risk of associated CNS anomalies. Only one case showed diffusion restriction and was diagnosed to be of hemorrhagic nature. The relation between symmetry and degree of ventriculomegaly was found to be statistically insignificant (P = 0.115). Assessment of different fetal brain congenital anomalies was not significantly affected by the using of DWI sequence as it is a functional modality rather than being a tool for assessment of anatomical gross abnormalities.

Conclusions

Fetal MRI is an important adjuvant to US in cases of ventriculomegaly particularly those associated with complex CNS anomalies. The association between CNS findings and ventricular dilatation was particularly evident in sever ventriculomegaly. The greater the width of the ventricular system, the more the risk of associated CNS anomalies. Fetal MRI may also be helpful in doubtful cases that could be misdiagnosed as ventriculomegaly including holoprosencephaly, hydranencephaly, porencephaly, and various supratentorial cystic lesions. DWI confirms the diagnosis of hemorrhage.

Acceleration and plateau: two patterns and outcomes of isolated severe fetal cerebral ventricular dilation

OBJECTIVES

We sought to characterize patterns of in utero dilation in isolated severe fetal ventriculomegaly (ISVM) and investigate their value in predicting obstetrical and postnatal outcomes.

METHODS

This is a retrospective cohort study. ISVM was defined as a sonographic cerebral ventricle atrial with width ≥15 mm in the absence of additional cerebral or other anatomic anomalies. The aim of this study was to characterize two ISVM groups using a receiver operator curve to evaluate the rate of ventricular progression versus need for ventriculoperitoneal (VP) shunt postnatally. Outcomes were compared between the groups using Pearson's chi-squared test, Student t-test, and descriptive statistics.

RESULTS

Based on the ROC analysis, ventricular growth of ≥3 mm/week versus <3 mm/week distinguished fetuses likely to require a postnatal VP shunt. Fetuses were characterized as accelerators if ventricle growth was ≥3 mm/week at any point and plateaus if <3 mm/week. Accelerators showed a greater average rate of ventricle progression than plateaus (4.1 vs. 1.0 mm/week, respectively, p = .031) and were more likely to be delivered at earlier gestational ages (34.7 vs. 37.1 weeks respectively, p = .02). Ninety percent of accelerators demonstrated a need for shunt placement compared with 18.8% of plateaus (p < .001). Significantly more plateaus (87.5%) underwent a trial of labor while accelerators were more likely to have planned cesareans (70%, p = .009).

CONCLUSIONS

This study characterizes ISVM into two distinct populations based upon the rate of ventricle expansion, differentiated by the need for postnatal shunting. Once a ventricular growth pattern is determined, these distinctions should prove useful in prenatal management and delivery planning.

Use of Fetal Magnetic Resonance Image Analysis and Machine Learning to Predict the Need for Postnatal Cerebrospinal Fluid Diversion in Fetal Ventriculomegaly

IMPORTANCE

Which children with fetal ventriculomegaly, or enlargement of the cerebral ventricles in utero, will develop hydrocephalus requiring treatment after birth is unclear.

OBJECTIVE

To determine whether extraction of multiple imaging features from fetal magnetic resonance imaging (MRI) and integration using machine learning techniques can predict which patients require postnatal cerebrospinal fluid (CSF) diversion after birth.

DESIGN, SETTING, AND PATIENTS

This retrospective case-control study used an institutional database of 253 patients with fetal ventriculomegaly from January 1, 2008, through December 31, 2014, to generate a predictive model. Data were analyzed from January 1, 2008, through December 31, 2015. All 25 patients who required postnatal CSF diversion were selected and matched by gestational age with 25 patients with fetal ventriculomegaly who did not require CSF diversion (discovery cohort). The model was applied to a sample of 24 consecutive patients with fetal ventriculomegaly who underwent evaluation at a separate institution (replication cohort) from January 1, 1998, through December 31, 2007. Data were analyzed from January 1, 1998, through December 31, 2009.

EXPOSURES

To generate the model, linear measurements, area, volume, and morphologic features were extracted from the fetal MRI, and a machine learning algorithm analyzed multiple features simultaneously to find the combination that was most predictive of the need for postnatal CSF diversion.

MAIN OUTCOMES AND MEASURES

Accuracy, sensitivity, and specificity of the model in correctly classifying patients requiring postnatal CSF diversion.

RESULTS

A total of 74 patients (41 girls [55%] and 33 boys [45%]; mean [SD] gestational age, 27.0 [5.6] months) were included from both cohorts. In the discovery cohort, median time to CSF diversion was 6 days (interquartile range [IQR], 2-51 days), and patients with fetal ventriculomegaly who did not develop symptoms were followed up for a median of 29 months (IQR, 9-46 months). The model correctly classified patients who required CSF diversion with 82% accuracy, 80% sensitivity, and 84% specificity. In the replication cohort, the model achieved 91% accuracy, 75% sensitivity, and 95% specificity.

CONCLUSION AND RELEVANCE

Image analysis and machine learning can be applied to fetal MRI findings to predict the need for postnatal CSF diversion. The model provides prognostic information that may guide clinical management and select candidates for potential fetal surgical intervention.