Disruptions in global network segregation and integration in adolescents and young adults with fetal alcohol spectrum disorder

Abstract

Background: Fetal alcohol spectrum disorder (FASD) is a significant public health problem that is associated with a broad range of physical, neurocognitive, and behavioral effects resulting from prenatal alcohol exposure (PAE). Magnetic resonance imaging (MRI) has been an important tool for advancing our knowledge of abnormal brain structure and function in individuals with FASD. However, whereas only a small number of studies have applied graph theory-based network analysis to resting-state functional MRI (fMRI) data in individuals with FASD additional research in this area is needed. Methods: Resting-state fMRI data were collected from adolescent and young adult participants (ages 12-22) with fetal alcohol syndrome (FAS) or alcohol-related neurodevelopmental disorder (ARND) and neurotypically developing controls (CNTRL) from previous studies. Group independent components analysis (gICA) was applied to fMRI data to extract components representing functional brain networks. Functional network connectivity (FNC), measured by Pearson correlation of the average independent component (IC) time series, was analyzed under a graph theory framework to compare network modularity, the average clustering coefficient, characteristic path length, and global efficiency between groups. Cognitive intelligence, measured by the Wechsler Abbreviated Scale of Intelligence (WASI), was compared and correlated to global network measures.Results: Group comparisons revealed significant differences in the average clustering coefficient, characteristic path length, and global efficiency. Modularity was not significantly different between groups. The FAS and ARND groups scored significantly lower than the CNTRL group on Full Scale IQ (FS-IQ) and the Vocabulary subtest, but not the Matrix Reasoning subtest. No significant associations between intelligence and graph theory measures were detected. Conclusion:Our results partially agree with previous studies examining global graph theory metrics in children and adolescents with FASD and suggest that the exposure to alcohol during prenatal development leads to disruptions in aspects of functional network segregation and integration.

Disrupted dynamic functional network connectivity in fetal alcohol spectrum disorders

“…In contrast to the many reports of decreased sFNC from fMRI data in various networks in participants with an FASD (Fan et al, 2017; Little et al, 2018; Long et al, 2018, 2020; Santhanam et al, 2011), the dFNC global meta‐state metrics from MEG data reported here indicate increased dynamic fluidity and dynamic range in participants with an FASD. This finding is in line with previous FASD studies which have found increases in global measures of cortical network connectivity such as increased characteristic path length in children with PAE (Wozniak et al, 2013, 2017), and disruptions in global network segregation and integration (i.e., averaging clustering coefficient characteristic path length and global efficiency) in adolescents an FASD (Rodriguez et al, 2021). There are important temporal differences between MEG and fMRI resting‐state data (i.e., MEG data have a higher temporal resolution and may provide connectivity patterns unique from fMRI) and distinctions between global meta‐state information and network correlations (i.e., how participants travel through states versus direct network to network correlations) which should be considered when comparing findings.…”

“…, the dFNC global metastate metrics from MEG data reported here indicate increased dynamic fluidity and dynamic range in participants with an FASD. This finding is in line with previous FASD studies which have found increases in global measures of cortical network connectivity such as increased characteristic path length in children with PAE(Wozniak et al, 2013(Wozniak et al, , 2017, and disruptions in global network segregation and integration (i.e., averaging clustering coefficient characteristic path length and global efficiency) in adolescents an FASD(Rodriguez et al, 2021). There are important temporal differences between MEG and fMRI resting-state data (i.e., MEG data have a higher tem-…”

The resting-state causal human connectome is characterized by hub connectivity of executive and attentional networks

“…The authors then followed this with a meta-analysis demonstrating that the hubs of the human connectome were more likely to contain gray matter lesions than non-hub regions across nine different disorders, including schizophrenia and Alzheimer’s disease. Many psychiatric and neurological disorders are associated with reduced brain global efficiency, including prenatal alcohol exposure ( Wozniak et al, 2013 ) and fetal alcohol syndrome ( Rodriguez et al, 2021 ), schizophrenia ( Hummer et al, 2020 ), ADHD ( Wang et al, 2020 , 2021 ), generalized anxiety disorder ( Guo et al, 2021 ), heavy smoking ( Lin et al, 2015 ), and major depressive disorder (( Meng et al, 2014 ); Wang et al, 2017 ; Zhi et al, 2018 ), among others. Additionally, prefrontal tDCS for alcohol use disorder increased brain network global efficiency ( Holla et al, 2020 ), suggesting that normalizing network global efficiency might contribute to improved treatment outcomes from neuromodulation therapy.…”

The Resting-State Causal Human Connectome is Characterized by Hub Connectivity of Executive and Attentional Networks

“…The authors then followed this with a meta-analysis demonstrating that the hubs of the human connectome were more likely to contain grey matter lesions than non-hub regions across nine different disorders, including schizophrenia and Alzheimer’s disease. Many psychiatric and neurological disorders are associated with reduced brain global efficiency, including prenatal alcohol exposure (Wozniak et al, 2013) and fetal alcohol syndrome (Rodriguez et al, 2021), schizophrenia (Hummer et al, 2020), ADHD (Wang et al, 2020; Wang et al, 2021), generalized anxiety disorder (Guo et al, 2021), heavy smoking (Lin et al, 2015), and major depressive disorder (Meng et al, 2013; Wang et al, 2017; Zhi et al, 2018), among others. Additionally, prefrontal tDCS for alcohol use disorder increased brain network global efficiency (Holla et al, 2020), suggesting that normalizing network global efficiency might contribute to improved treatment outcomes from neuromodulation therapy.…”