Resting-state fMRI links longitudinal brain activity changes to cognitive function in depression

480 

Abstract Submission 

Zi-You Qiu¹, Vincent Chin-Hung Chen^2,3, Yuan-Hsiung Tsai^2,4, Jun-Cheng Weng^1,3

¹Department of Medical Imaging and Radiological Sciences, Chang Gung University, Taoyuan, Taiwan, ²School of Medicine, Chang Gung University, Taoyuan, Taiwan, ³Department of Psychiatry, Chang Gung Memorial Hospital, Chiayi, Taiwan, ⁴Department of Diagnostic Radiology, Chang Gung Memorial Hospital, Chiayi, Taiwan

Zi-You Qiu  
Department of Medical Imaging and Radiological Sciences, Chang Gung University
Taoyuan, Taiwan

Vincent Chin-Hung Chen  
School of Medicine, Chang Gung University|Department of Psychiatry, Chang Gung Memorial Hospital
Taoyuan, Taiwan|Chiayi, Taiwan

Yuan-Hsiung Tsai  
School of Medicine, Chang Gung University|Department of Diagnostic Radiology, Chang Gung Memorial Hospital
Taoyuan, Taiwan|Chiayi, Taiwan

Jun-Cheng Weng  
Department of Medical Imaging and Radiological Sciences, Chang Gung University|Department of Psychiatry, Chang Gung Memorial Hospital
Taoyuan, Taiwan|Chiayi, Taiwan

Depression is a prevalent disorder in the general population. Indeed, 3.8% of the population, including 5% of adults and 5.7% of adults older than 60 years old, have depression [1]. It thus represents a serious public health concern. In addition, depression can be categorized according to severity as mild, moderate, or severe. Individuals with severe depression often experience suicidal ideation (SI), and over 700,000 people die because of suicide every year. In the study, we aimed to investigate whether there was a longitudinal association between brain function alterations and neuropsychological assessment changes in in depressed patients with SI and without SI (NS).

We recruited 68 subjects from Chiayi Chang Gung Memorial Hospital. The subjects were divided into four groups: 27 depressed patients without suicidal ideation (NS), 18 depressed patients suicidal ideation (SI), and 13 depressed patients with SI converted to NS (improved). All subjects underwent resting-state fMRI at baseline (TP1) and one year later (TP2) after receiving therapy. They also underwent four neuropsychological assessments, including the Hamilton Depression Rating Scale (HAM-D)[2], Hospital Anxiety and Depression Scale (HADS) [3], Beck Scale for Suicide Ideation (BSS) [4], and Ruminative Responses Scale (RRS) [5], that yielded scale scores. The mean amplitude of low-frequency fluctuations (mfALFF) and mean regional homogeneity (mReHo) were used to obtain functional indices. Multiple regression analysis was also used to examine the association between brain function alterations and neuropsychological assessment changes in each group.

In multiple regression analysis, differences in mfALFF/mReHo over time were compared with differences in assessment scores. We found several positive and negative correlations between mfALFF/mReHo alterations and assessment score changes after treatment. There was a negative correlation between the ΔmReHo of the precentral gyrus and ΔHAMD scores in the SI group (Fig. 1a). There were negative correlations of ΔmfALFF and ΔmReHo in the precuneus with ΔHAMD scores in the improved group (Fig. 1b, 1c). We found a positive correlation between ΔmfALFF of the caudate and ΔHADS-A scores in the improved group (Fig. 2a). We also found a positive correlation between ΔmReHo of the insula and ΔHADS-A scores in the improved group (Fig. 2b). There was a negative correlation of ΔmfALFF and ΔmReHo of the inferior parietal lobe with ΔHADS-A scores in the improved group (Fig. 1d, 1e). There was a positive correlation between ΔmfALFF of the superior temporal gyrus and ΔBSS scores in the SI group (Fig. 2c). There was a negative correlation between ΔmfALFF of the thalamus and ΔRRS scores in the SI group (Fig. 1f). There were negative correlations of ΔmfALFF of the ACC and insula with ΔRRS scores in the improved group (Fig. 1g, 1h) (corrected p value of all results < 0.05).

This longitudinal study investigated brain function alterations in depressed patients with three trajectories of SI at two time points. In multiple regression analysis, we found several associations between functional mapping changes and neuropsychological assessment changes in each group. Our findings provide more information about potential neural biomarkers of various depressive disorders. In the future, machine learning or deep learning could be included to enhance the ease and accuracy of diagnosis.

Psychiatric (eg. Depression, Anxiety, Schizophrenia) ¹

Executive Function, Cognitive Control and Decision Making

fMRI Connectivity and Network Modeling ²

Anatomy and Functional Systems

BOLD fMRI

Anxiety

Cognition

Data analysis

FUNCTIONAL MRI

MRI

Psychiatric Disorders