Structural brain network organization in children with prenatal alcohol exposure

Xiaoyun Liang,Claire E. Kelly,Chun-Hung Yeh,Thijs Dhollander,Stephen Hearps,Peter J. Anderson,Deanne K. Thompson

DOI: https://doi.org/10.1016/j.nicl.2024.103690

IF: 4.891

2024-10-25

NeuroImage Clinical

Abstract:Introduction There is growing evidence suggesting that children with prenatal alcohol exposure (PAE) struggle with cognitively demanding tasks, such as learning, attention, and language. Complex structural network analyses can provide insight into the neurobiological underpinnings of these functions, as they may be sensitive for characterizing the effects of PAE on the brain. However, investigations on how PAE affects brain networks are limited. We aim to compare diffusion magnetic resonance imaging (MRI) tractography-based structural networks between children with low-to-moderate PAE in trimester 1 only (T1) or throughout all trimesters (T1-T3) with those without alcohol exposure prenatally. Methods Our cohort included three groups of children aged 6 to 8 years: 1) no PAE (n = 24), 2) low-to-moderate PAE during T1 only (n = 30), 3) low-to-moderate PAE throughout T1-T3 (n = 36). Structural networks were constructed using the multi-shell multi-tissue constrained spherical deconvolution tractography technique. Quantitative group-wise analyses were conducted at three levels: (a) at the whole-brain network level, using both network-based statistical analyses and network centrality; and then using network centrality at (b) the modular level, and (c) per-region level, including the regions identified as brain hubs. Results Compared with the no PAE group, widespread brain network alterations were observed in the PAE T1-T3 group using network-based statistics, but no alterations were observed for the PAE T1 group. Network alterations were also detected at the module level in the PAE T1-T3 compared with the no PAE group, with lower eigenvector centrality in the module that closely represented the right cortico-basal ganglia-thalamo-cortical network. No significant group differences were found in network centrality at the per-region level, including the hub regions. Conclusions This study demonstrated that low-to-moderate PAE throughout pregnancy may alter brain structural connectivity, which may explain the neurodevelopmental deficits associated with PAE. It is possible that timing and duration of alcohol exposure are crucial, as PAE in T1 only did not appear to alter brain structural connectivity.

neuroimaging