Assessment of Brain Development in Children With Congenital Diaphragmatic Hernia - an Automated Brain Segmentation Approach

Abstract

Background/aim: Congenital diaphragmatic hernia (CDH) is a critical condition affecting newborns, which often results in long-term morbidities, including neurodevelopmental delays, which affect cognitive, motor, and behavioral functions. These delays are believed to stem from prenatal and postnatal factors, such as impaired lung development and chronic hypoxia, which disrupt normal brain growth. Understanding the underlying mechanisms of these neurodevelopmental impairments is crucial for improving prognosis and patient outcomes, particularly as advances in treatments like ECMO have increased survival rates but also pose additional risks for neurodevelopment. This study aimed to evaluate brain development in 2-year-old children who underwent CDH repair, with and without ECMO, compared to healthy controls using an MRI-based automated segmentation approach.

Patients and methods: The study included 31 children with CDH, of which 10 received ECMO therapy, and a control group of 31 healthy children. MRI-examinations were performed using a 3-T system. MRI data were processed using the CerebroMatic toolbox and SPM12 software to measure cerebrospinal fluid (CSF), gray matter (GM), white matter (WM), and cortical thickness (CT).

Results: Patients with CDH showed significantly increased volumes of CSF (p=0.009), GM (p=0.02), and total intracranial volume (TIV) (p=0.01), compared to healthy controls. ECMO-treated patients had significantly increased GM (p=0.01) and CSF (p=0.005) volumes in comparison to healthy controls. CT was significantly higher in CDH patients regardless of ECMO therapy, indicating potential maturational deficits.

Conclusion: The study reveals neurodevelopmental differences in children with CDH, particularly in those requiring ECMO therapy. Increased CT, GM, and CSF volumes suggest complex neurodevelopmental challenges.

Keywords: Congenital diaphragmatic hernia (CDH); brain segmentation; cerebrospinal fluid (CSF); cortical thickness; gray matter (GM); white matter (WM).

Figure 1

Flowchart of the study population. The population initially included 47 consecutive children diagnosed with congenital diaphragmatic hernia (CDH) underwent CDH repair postnatally and had a follow-up MRI examination at the age of 2 years and 31 healthy children as the control group. However, 16 children with inadequate MRI image quality were excluded. The final study population consisted of 62 children, including 31 children with CDH: 15 boys and 16 girls, mean age 25±2.7 months, and 31 healthy control children: 15 boys and 16 girls, mean age 25±3.1.

Figure 2

Imaging postprocessing pipeline. The figure illustrates the steps involved in the MRI data processing for both patient and control groups, including coregistration to a customized DARTEL template, automated segmentation of cerebrospinal fluid (CSF), Gray matter (GM), and white matter (WM), and measurement of cortical thickness (CT) and total intracranial volume (TIV) using surface-based and voxel-based morphometry algorithms within CAT 12 toolbox.

Figure 3

Comparison of cerebrospinal fluid (CSF), gray matter (GM), and white matter (WM) between healthy controls and patients with congenital diaphragmatic hernia (CDH). The figure shows statistically significant increases in CSF and GM in CDH patients compared to healthy controls. WM did not show significant differences. Error bars represent standard deviations.

Figure 4

Mean cortical thickness (CT) in healthy controls, congenital diaphragmatic hernia (CDH) patients, and CDH patients with and without extracorporeal membrane oxygenation (ECMO) therapy. CDH patients show a statistically significant increase in CT compared to healthy controls, with no significant difference between CDH patients with and without ECMO therapy. Error bars represent standard deviations.