Assessment of Disrupted Brain Structural Connectome in Depressive Patients With Suicidal Ideation Using Generalized Q-Sampling MRI

Vincent Chin-Hung Chen; Chun-Ju Kao; Yuan-Hsiung Tsai; Man Teng Cheok; Roger S McIntyre; Jun-Cheng Weng

doi:10.3389/fnhum.2021.711731

Assessment of Disrupted Brain Structural Connectome in Depressive Patients With Suicidal Ideation Using Generalized Q-Sampling MRI

Front Hum Neurosci. 2021 Aug 27:15:711731. doi: 10.3389/fnhum.2021.711731. eCollection 2021.

Authors

Vincent Chin-Hung Chen^{1

2}, Chun-Ju Kao³, Yuan-Hsiung Tsai^{1

4}, Man Teng Cheok^{3

5}, Roger S McIntyre^{6

7

8}, Jun-Cheng Weng^{2

3

9}

Affiliations

¹ School of Medicine, Chang Gung University, Taoyuan, Taiwan.
² Department of Psychiatry, Chang Gung Memorial Hospital, Chiayi, Taiwan.
³ Department of Medical Imaging and Radiological Sciences, Bachelor Program in Artificial Intelligence, Chang Gung University, Taoyuan, Taiwan.
⁴ Department of Diagnostic Radiology, Chang Gung Memorial Hospital, Chiayi, Taiwan.
⁵ Graduate Institute of Biomedical Electronics and Bioinformatics, National Taiwan University, Taipei, Taiwan.
⁶ Mood Disorder Psychopharmacology Unit, University Health Network, Department of Psychiatry, University of Toronto, Toronto, ON, Canada.
⁷ Institute of Medical Science, University of Toronto, Toronto, ON, Canada.
⁸ Departments of Psychiatry and Pharmacology, University of Toronto, Toronto, ON, Canada.
⁹ Medical Imaging Research Center, Institute for Radiological Research, Chang Gung University and Chang Gung Memorial Hospital at Linkou, Taoyuan, Taiwan.

Abstract

Suicide is one of the leading causes of mortality worldwide. Various factors could lead to suicidal ideation (SI), while depression is the predominant cause among all mental disorders. Studies have shown that alterations in brain structures and networks may be highly associated with suicidality. This study investigated both neurological structural variations and network alterations in depressed patients with suicidal ideation by using generalized q-sampling imaging (GQI) and Graph Theoretical Analysis (GTA). This study recruited 155 participants and divided them into three groups: 44 depressed patients with suicidal ideation (SI+; 20 males and 24 females with mean age = 42, SD = 12), 56 depressed patients without suicidal ideation (Depressed; 24 males and 32 females with mean age = 45, SD = 11) and 55 healthy controls (HC; nine males and 46 females with mean age = 39, SD = 11). Both the generalized fractional anisotropy (GFA) and normalized quantitative anisotropy (NQA) values were evaluated in a voxel-based statistical analysis by GQI. We analyzed different topological parameters in the graph theoretical analysis and the subnetwork interconnections in the Network-based Statistical (NBS) analysis. In the voxel-based statistical analysis, both the GFA and NQA values in the SI+ group were generally lower than those in the Depressed and HC groups in the corpus callosum and cingulate gyrus. Furthermore, we found that the SI+ group demonstrated higher global integration and lower local segregation among the three groups of participants. In the network-based statistical analysis, we discovered that the SI+ group had stronger connections of subnetworks in the frontal lobe than the HC group. We found significant structural differences in depressed patients with suicidal ideation compared to depressed patients without suicidal ideation and healthy controls and we also found several network alterations among these groups of participants, which indicated that white matter integrity and network alterations are associated with patients with depression as well as suicidal ideation.

Keywords: generalized q-sampling imaging; graph theoretical analysis; magnetic resonance imaging; network-based statistical analysis; suicidal ideation.