Atypical effective connectivity from the frontal cortex to striatum in alcohol use disorder
Hongwen Song,Ping Yang,Xinyue Zhang,Rui Tao,Lin Zuo,Weili Liu,Jiaxin Fu,Zhuo Kong,Rui Tang,Siyu Wu,Liangjun Pang,Xiaochu Zhang
DOI: https://doi.org/10.1038/s41398-024-03083-8
2024-09-20
Translational Psychiatry
Abstract:Alcohol use disorder (AUD) is a profound psychiatric condition marked by disrupted connectivity among distributed brain regions, indicating impaired functional integration. Previous connectome studies utilizing functional magnetic resonance imaging (fMRI) have predominantly focused on undirected functional connectivity, while the specific alterations in directed effective connectivity (EC) associated with AUD remain unclear. To address this issue, this study utilized multivariate pattern analysis (MVPA) and spectral dynamic causal modeling (DCM). We recruited 32 abstinent men with AUD and 30 healthy controls (HCs) men, and collected their resting-state fMRI data. A regional homogeneity (ReHo)-based MVPA method was employed to classify AUD and HC groups, as well as predict the severity of addiction in AUD individuals. The most informative brain regions identified by the MVPA were further investigated using spectral DCM. Our results indicated that the ReHo-based support vector classification (SVC) exhibits the highest accuracy in distinguishing individuals with AUD from HCs (classification accuracy: 98.57%). Additionally, our results demonstrated that ReHo-based support vector regression (SVR) could be utilized to predict the addiction severity (alcohol use disorders identification test, AUDIT, R 2 = 0.38; Michigan alcoholism screening test, MAST, R 2 = 0.29) of patients with AUD. The most informative brain regions for the prediction include left pre-SMA, right dACC, right LOFC, right putamen, and right NACC. These findings were validated in an independent data set (35 patients with AUD and 36 HCs, Classification accuracy: 91.67%; AUDIT, R 2 = 0.17; MAST, R 2 = 0.20). The results of spectral DCM analysis indicated that individuals with AUD exhibited decreased EC from the left pre-SMA to the right putamen, from the right dACC to the right putamen, and from the right LOFC to the right NACC compared to HCs. Moreover, the EC strength from the right NACC to left pre-SMA and from the right dACC to right putamen mediated the relationship between addiction severity (MAST scores) and behavioral measures (impulsive and compulsive scores). These findings provide crucial evidence for the underlying mechanism of impaired self-control, risk assessment, and impulsive and compulsive alcohol consumption in individuals with AUD, providing novel causal insights into both diagnosis and treatment.
psychiatry