670 - Term-equivalent age Global Brain Abnormality Score, but not Diffuse White Matter Abnormality, is associated with cortical and white matter dysmaturation in individuals born preterm at 5-years of age

Publication Number: 2653.670

Mitchell A. Batschelett, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States; Armin Allahverdy, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States; Hailong Li, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States; Lili He, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States; Nehal A. Parikh, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States

Background: Abnormal term-equivalent age (TEA) MRI findings, including diffuse white matter abnormality (DWMA) and elevated Kidokoro global brain abnormality (GBA) scores, are common in preterm infants and predict later functional outcomes. However, their association with cortical and white matter maturation at early school age remains poorly understood.

Objective: We aimed to quantify relationships between DWMA and GBA at TEA and cortical and white matter maturation at 5-years corrected-age (CA).

Design/Methods: This study includes individuals born at or before 32 weeks' gestational age from the multisite Cincinnati Infant Neurodevelopmental Early Prediction Study cohort. Participants underwent MRI exams at TEA (40-44 weeks' post-menstrual age) and 5-years CA. DWMA was quantified using our published algorithm and GBA by a neonatal neuroradiologist using high-resolution T2-weighted sequences at TEA. We used MRtrix3 to estimate fiber density (FD), cross-section (FC), and combined density-cross-section (FDC) throughout white matter from 5-year diffusion MRI sequences and employed fixel-based analysis (n=182) to test brain-wise associations between TEA DWMA/GBA and white matter metrics. We used FreeSurfer to extract intracranial volume-corrected, whole-brain measures of white matter volume (WMV), cortical surface area (CSA), cortical thickness (CT), cerebrospinal fluid volume (CSF), and mean cortical curvature (MCC) using 5-year T1 images and applied robust regression (n=180) to test associations between TEA DWMA/GBA and cortical maturation markers. Analyses were corrected for confounders (post-menstrual age at TEA MRI, gestational age at birth, sex, bronchopulmonary dysplasia, retinopathy of prematurity, maternal milk at NICU discharge, and high social risk).

Results: TEA GBA demonstrated significant negative associations with FD in the rostrum of the corpus callosum and with FC/FDC across widespread white matter regions (Figure 1). It was also significantly negatively associated with WMV (beta=-0.187, p=0.014) and positively associated with CSF volume (beta=0.216, p=.041; Table 1). TEA DWMA showed no significant associations with any white matter or cortical outcomes (Table 2).

Conclusion(s): GBA, but not DWMA, is associated with white matter pathology, especially reduced cross-section, and cortical dysmaturation at 5-years CA. These findings suggest different underlying pathophysiologic mechanisms, and clinicians may benefit from considering both MRI findings in concert during early risk stratification.

Figure 1. Results from whole-brain diffusion MRI fixel-based analysis, testing for associations between GBA and FD, FC, and FDC.
A) Representative axial slice of significant white matter fixels overlaid onto population template white matter image. Increasing GBA score at TEA showed significant negative associations with 5-year FD in the rostrum of the corpus callosum. B) Representative coronal slice of significant white matter fixels overlaid onto population template white matter image. Increasing GBA score at TEA showed significant negative associations with 5-year FC in widespread white matter. C) Representative coronal slice (same slice as panel B) of significant white matter fixels overlaid onto population template white matter image. Increasing GBA score at TEA showed significant negative associations with 5-year FDC in widespread, yet more restricted than FC, white matter. Abbreviations: GBA=Kidokoro global brain abnormality score; FD=fiber density; FC=fiber cross-section; FDC=fiber density-cross-section; TEA=term-equivalent age.

Table 1. Results from robust regression, testing for associations between GBA and whole-brain cortical maturation metrics.
Regressions were run with n=180 unique subjects with high quality T1 imaging at 5-years corrected-age. GBA, PMA, and GA were continuous variables, and Sex, BPD, ROP, Mother's Milk, and Social Risk were binary variables. Beta coefficients and p-values for Sex represent the effect of male sex on the outcome variable. Abbreviations: GBA=Kidokoro global brain abnormality score; PMA=post-menstrual age; GA=gestational age; BPD=bronchopulmonary dysplasia; ROP=retinopathy of prematurity; WMV=white matter volume; CSF=cerebrospinal fluid volume; CSA=cortical surface area; CT=cortical thickness; MCC=mean cortical curvature.

Table 2. Results from robust regression, testing for associations between DWMA and whole-brain cortical maturation metrics.
Regressions were run with n=180 unique subjects with high quality T1 imaging at 5-years corrected-age. DWMA, PMA, and GA were continuous variables, and Sex, BPD, ROP, Mother's Milk, and Social Risk were binary variables. Beta coefficients and p-values for Sex represent the effect of male sex on the outcome variable. Abbreviations: DWMA=diffuse white matter abnormality; PMA=post-menstrual age; GA=gestational age; BPD=bronchopulmonary dysplasia; ROP=retinopathy of prematurity; WMV=white matter volume; CSF=cerebrospinal fluid volume; CSA=cortical surface area; CT=cortical thickness; MCC=mean cortical curvature.