Insula functional connectivity in obsessive-compulsive disorder: subregions and gradients
Poster No:
645
Submission Type:
Abstract Submission
Authors:
Lingxiao Cao1, Hailong Li1, Jiaxin Jiang2, Bin Li2, Shuangwei Chai1, Huan Zhou1, Qiyong Gong1, Xiaoqi Huang1
Institutions:
1Department of Radiology and Huaxi MR Research Center, West China Hospital, Sichuan University, Chengdu, China, 2Mental Health Center, West China Hospital of Sichuan University, Chengdu, China
First Author:
Lingxiao Cao
Department of Radiology and Huaxi MR Research Center, West China Hospital, Sichuan University
Chengdu, China
Department of Radiology and Huaxi MR Research Center, West China Hospital, Sichuan University
Chengdu, China
Co-Author(s):
Hailong Li
Department of Radiology and Huaxi MR Research Center, West China Hospital, Sichuan University
Chengdu, China
Department of Radiology and Huaxi MR Research Center, West China Hospital, Sichuan University
Chengdu, China
Shuangwei Chai
Department of Radiology and Huaxi MR Research Center, West China Hospital, Sichuan University
Chengdu, China
Department of Radiology and Huaxi MR Research Center, West China Hospital, Sichuan University
Chengdu, China
Huan Zhou
Department of Radiology and Huaxi MR Research Center, West China Hospital, Sichuan University
Chengdu, China
Department of Radiology and Huaxi MR Research Center, West China Hospital, Sichuan University
Chengdu, China
Qiyong Gong
Department of Radiology and Huaxi MR Research Center, West China Hospital, Sichuan University
Chengdu, China
Department of Radiology and Huaxi MR Research Center, West China Hospital, Sichuan University
Chengdu, China
Xiaoqi Huang
Department of Radiology and Huaxi MR Research Center, West China Hospital, Sichuan University
Chengdu, China
Department of Radiology and Huaxi MR Research Center, West China Hospital, Sichuan University
Chengdu, China
Introduction:
The insula is gaining increasing attention in neurocircuitry models of obsessive-compulsive disorder (OCD) (Shephard E, 2021). Most previous neuroimaging studies investigating insula functional connectivity (FC) in OCD have treated insula as a single, homogeneous region (Zhou Z, 2022), and thus any potential variation in connectivity across the topography of insula may be overlooked. The connectional variation can be characterized using two models: 1) discrete clusters (subregions) and 2) continua of variation (gradients), which have been applied to examine insula FC architecture in other psychiatric disorders (Tian Y, 2019), with few investigations in OCD yet. This study aimed to examine insula FC architecture in OCD with a novel clustering-based technique and gradient-based approach.
Methods:
A total of 92 medication-free OCD patients and 90 age- and sex-matched healthy controls (HC) (Table 1) were scanned using 3T GE MRI.
Subregions were generated using CBPtools (Reuter N, 2020). Specifically, k-means clustering was used to each subject's insula-to-whole-brain FC matrix to assign each insula voxel to a cluster, effectively grouping similar voxels based on their FC patterns. Then, the individual clusterings were relabeled and the mode of each insula voxel was computed, serving as the group-level clustering results. Diagnosis-by-subregion interaction effects were examined, with age, sex and mean FD as covariates. The FC values were extracted from the clusters showing significance for further post hoc analysis using simple effects test.
Gradient mapping was generated using BrainSpace (Vos de Wael R, 2020). Specifically, only the top 10% connections of each insula voxel were used to calculate the cosine similarity. The similarity matrix was scaled into a normalized angle matrix and further fed into the diffusion map embedding algorithm. The global-level gradient distribution was compared using Mann-Whitney U test. Three global metrics, including gradient explanation ratio, range, and variation, were calculated, and compared by a GLM model with age, sex, and mean FD as covariates.
Subregions were generated using CBPtools (Reuter N, 2020). Specifically, k-means clustering was used to each subject's insula-to-whole-brain FC matrix to assign each insula voxel to a cluster, effectively grouping similar voxels based on their FC patterns. Then, the individual clusterings were relabeled and the mode of each insula voxel was computed, serving as the group-level clustering results. Diagnosis-by-subregion interaction effects were examined, with age, sex and mean FD as covariates. The FC values were extracted from the clusters showing significance for further post hoc analysis using simple effects test.
Gradient mapping was generated using BrainSpace (Vos de Wael R, 2020). Specifically, only the top 10% connections of each insula voxel were used to calculate the cosine similarity. The similarity matrix was scaled into a normalized angle matrix and further fed into the diffusion map embedding algorithm. The global-level gradient distribution was compared using Mann-Whitney U test. Three global metrics, including gradient explanation ratio, range, and variation, were calculated, and compared by a GLM model with age, sex, and mean FD as covariates.
Results:
Compared with HC, OCD patients showed enhanced insula FC with supplementary motor area (SMA), precuneus, cerebellum and inferior parietal lobule (Figure 1A). The insula was parcellated into two subregions corresponding to anterior insula and posterior insula (Figure 1B). We found significant diagnosis-by-subregion interaction in SMA, precentral gyrus (PCG), middle frontal gyrus (MFG) and dorsomedial prefrontal cortex (dmPFC) (Figure 1C). Relative to HC, the anterior subregion in OCD related to increased connectivity strength with the SMA, PCG and MFG, whereas the posterior subregion showed decreased connectivity strength with the dmPFC (Figure 1D).
The spatial patterns of the group-averaged principle gradient maps of the insula were organized along a gradual anterior/posterior axis. A visual inspection of the histogram demonstrated that the extremes of the gradient were contracted in OCD relative to HC (Figure 2A), though the between-group comparisons in the cumulative distribution were not significant. The gradient of the left insula in OCD explained less variance than that in HC (P = 0.026, Figure 2B). Moreover, the OCD patients showed a narrower range of scores (P = 0.040, Figure 2C) and less spatial variation (P = 0.042, Figure 2D) in the left insula.
The spatial patterns of the group-averaged principle gradient maps of the insula were organized along a gradual anterior/posterior axis. A visual inspection of the histogram demonstrated that the extremes of the gradient were contracted in OCD relative to HC (Figure 2A), though the between-group comparisons in the cumulative distribution were not significant. The gradient of the left insula in OCD explained less variance than that in HC (P = 0.026, Figure 2B). Moreover, the OCD patients showed a narrower range of scores (P = 0.040, Figure 2C) and less spatial variation (P = 0.042, Figure 2D) in the left insula.
Conclusions:
We comprehensively characterized insula FC architecture using discrete subregions as well as continual gradients in OCD. We found unbalanced insula subregional FC alterations in OCD patients with increased FC strength in the anterior subregion while decreased FC strength in the posterior subregion. We also found OCD patients exhibited global topographic alterations in the principal anterior-posterior gradient of the left insula. These results highlight the disrupted FC architecture of the insula in OCD, providing insights into the neurobiological underpinnings of OCD.
Disorders of the Nervous System:
Psychiatric (eg. Depression, Anxiety, Schizophrenia) 1
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural) 2
fMRI Connectivity and Network Modeling
Segmentation and Parcellation
Keywords:
FUNCTIONAL MRI
Obessive Compulsive Disorder
Segmentation
Other - Gradient; Insula
1|2Indicates the priority used for review
Provide references using author date format
Shephard, E. (2021), 'Toward a neurocircuit-based taxonomy to guide treatment of obsessive-compulsive disorder', Molecular psychiatry, vol. 26, no. 9, pp. 4583-4604.
Tian, Y. (2019), 'Insula Functional Connectivity in Schizophrenia: Subregions, Gradients, and Symptoms', Biological psychiatry. Cognitive neuroscience and neuroimaging, vol. 4, no. 4, pp. 399-408.
Vos de Wael, R. (2020), ' BrainSpace: a toolbox for the analysis of macroscale gradients in neuroimaging and connectomics datasets ', Communications biology, vol. 3, no. 1, pp. 103.
Zhou, Z. (2022), 'Abnormal resting-state functional connectivity of the insula in medication-free patients with obsessive-compulsive disorder', BMC Psychiatry, vol. 22, no. 1, pp. 742.