Both Hypo-Connectivity and Hyper-Connectivity of the Insular Subregions Associated With Severity in Children With Autism Spectrum Disorders

Abstract

Some studies identified hypo-connectivity, while others showed hyper-connectivity of the insula in the autism spectrum disorders (ASD). These contradictory findings leave open the question of whether and to what extent functional connectivity of the insula is altered and how functional connectivity of the insula is associated with the severity of ASD. A newly emerging insular atlas that comprises multiple functionally differentiated subregions provides a new framework to interpret the functional significance of insular findings and uncover the mechanisms underlying the severity of ASD. Using the new insular atlas, the present study aimed to investigate the distinct functional connectivity of the insular subregions and their associations with ASD severity in a cohort of 49 children with ASD and 33 typically developing (TD) subjects. We found that compared with TD group, the ASD group showed different connectivity patterns in the left ventral agranular insula, right ventral dysgranular and granular insula, and dorsal dysgranular insula, characterized by significant hyper-connectivity and/or hypo-connectivity with special brain regions. Furthermore, both the hypo-connectivity and hyper-connectivity patterns of the insular subregions were significantly associated with the severity of ASD symptoms. Our research demonstrated distinct functional connectivity patterns of the insular subregions and emphasized the importance of the subdivisions within the insula to the potential impact of functional difference in children with ASD. Moreover, these results might help us to better understand the mechanisms underlying the symptoms in children with ASD and might elucidate potential biomarkers for clinical applications.

PTSD and its dissociative subtype through the lens of the insula: Anterior and posterior insula resting‐state functional connectivity and its predictive validity using machine learning

“…First, the insula seed regions of interest were taken from the Brainnetome anatomical atlas, thus lacking sensitivity to identify individual anatomical differences given the relative proximity of seed regions. Nonetheless, the insula parcellations derived from this atlas have been used successfully in numerous insula functional connectivity studies (see Xu et al, ; Yu et al, ; Zhang et al, ). Secondly, while our three participant groups were sex matched, we did not examine insula resting‐state connectivity patterns as a function of sex.…”

Specific Functional Connectivity Patterns of Middle Temporal Gyrus Subregions in Children and Adults with Autism Spectrum Disorder

“…The main steps were as follows: slice timing; realign; normalize to Montreal Neurological Institute (MNI) space; regress out head motion effects, as well as the white matter, cerebrospinal fluid, and global signals; linear and quadratic trends; filtering (0.01–0.1 Hz); and smoothed by a kernel of 6 mm. For details, refer to our previous paper [Wang, Wei, et al, , ; Wang, Yu, et al, ; Xu et al, ].…”

“…For the child group, we used the same data that we used in our previous study [Xu et al, ], which included 49 children with ASD (age ranged from 8 to 17) and 33 TD subjects (age ranged from 9 to 17) (Table ) based on criteria from the Quality Assessment Protocol http://preprocessed-connectomes-project.org/quality-assessment-protocol/). Moreover, subjects whose mean framewise displacement (FD) was greater than 1 mm or whose maximum head motion exceeded 3 mm or 3° were excluded.…”

“…Then, two‐sample t ‐tests were performed to investigate group differences in RSFC between children with ASD and the TD group with age, gender, handedness, verbal IQ, performance IQ, and mean FD as covariates, and between adults with ASD and the TD group with age, handedness, verbal IQ, performance IQ, and mean FD as covariates. Given that false discovery rate (FDR) method is too strict for whole brain functional connectivity analyses [Wu et al, ; Xu et al, , ], the significance was determined using the Gaussian random field (GRF) correction with a voxel‐level corrected threshold of P < 0.001 and a cluster‐level corrected threshold of P < 0.05. We also calculated the mean RSFC of the regions which showed significantly altered RSFC with subregions of the MTG in children and adults in the ASD and TD group.…”

Altered Connectivity Between Cerebellum, Visual, and Sensory-Motor Networks in Autism Spectrum Disorder: Results from the EU-AIMS Longitudinal European Autism Project

“…Given that increased connectivity was more frequently reported in childhood ASD and decreased connectivity more frequently in adulthood ASD, these findings were initially ascribed to developmental effects around puberty (15). Yet, this hypothesis does not accommodate more recent studies reporting functional connectivity increases in certain brain regions, but decreases in other areas in both children with ASD (16,17) and adults with ASD (18). These studies indicate that hypotheses of a global increase or decrease in connectivity are likely overly simplistic and that functional connectivity changes in ASD might be network-dependent.…”