Alterations of White Matter Connectivity in Preschool Children with Autism Spectrum Disorder

Purpose: To investigate the topologic architecture of white matter connectivity networks in preschool-aged children with a diagnosis of autism spectrum disorder (ASD) versus typical development (TD). Materials and Methods: Forty-two participants were enrolled, including 21 preschool children with ASD (14 male children and seven female children; mean age, 4.56 years +/- 0.97 [standard deviation]) and 21 children with TD (11 males and 10 females; mean age, 5.13 years +/- 0.82). The diagnosis of ASD was determined according to the Diagnostic and Statistical Manual of Mental Disorders Global Assessment of Functioning scores (mean score, 8.00 +/- 0.50). All participants underwent diffusion-tensor imaging (DTI) and T2-weighted imaging on a 3-T magnetic resonance system. A graph theoretical analysis was applied to investigate the topologic organization of the brain network including global and local topologic parameters. Statistical analysis was then performed for the comparison between the groups. Results: Compared with the TD group, children with ASD demonstrated shortened characteristic path length (t(1) = 0.536, t(2) = 0.534, t(3) = 0.523, t(4) = 0.510, and t(5) = 0.501; P<.05) and increased global efficiency (t(1) = 0.499, t(2) = 0.497, t(3) = 0.486, t(4) = 0.473, and t(5) = 0.465; P<.05) and clustering coefficient (t(1) = 0.673, t(2) = 0.750, t(3) = 0.757, t(4) = 0.738, and t(5) = 0.741; P<.05). Significant increases in nodal efficiency were mainly found in left pallidum (0.037 vs 0.032, respectively; P<.01) and right caudate nucleus (0.037 vs 0.032, respectively; P<.01) of the basal ganglia network. Conclusion: Significantly altered patterns of global and local brain network topography may underlie the abnormal brain development in preschool children with ASD compared with those who have TD. The identification of altered structural connectivity in basal ganglia and paralimbic-limbic networks may point toward potential imaging biomarkers for preschool-age patients with ASD. (c) RSNA, 2018