AUTHOR=Hu Liyu , Chen Jianxiang , Li Xinbei , Zhang Haoran , Zhang Jinhuan , Lu Yingqi , Lian Jie , Yu Haibo , Yang Nan , Wang Jianjun , Lyu Hanqing , Xu Jinping 

TITLE=Disruptive and complementary effects of depression symptoms on spontaneous brain activity in the subcortical vascular mild cognitive impairment

JOURNAL=Frontiers in Aging Neuroscience

VOLUME=Volume 16 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/aging-neuroscience/articles/10.3389/fnagi.2024.1338179

DOI=10.3389/fnagi.2024.1338179

ISSN=1663-4365

ABSTRACT=Background: Although depression symptoms were widely reported in patients with subcortical vascular mild cognitive impairment (svMCI), whether and how they affected brain functions is largely unknown and their diagnosis is mainly dependent on behavioral analyses. Methods: Herein, resting-state fMRI data were obtained from a cohort of 34 svMCI patients including 18 svMCI with depression symptoms (svMCI+D) and 16 svMCI without depression symptoms (svMCI-D), and 34 normal controls (NC). The fractions of amplitude of low-frequency fluctuations (fALFF), resting-state functional connectivity, correlation analyses, and support vector machine (SVM) were performed. Results: The fALFF of right cerebellum (CERE.R) was altered among svMCI+D, svMCI-D, and NC. The regional mean fALFF of CERE.R was decreased in the svMCI-D compared to NC but increased in the svMCI+D compared to svMCI-D. Moreover, the adjusted fALFF of CERE.R was significantly correlated with Montreal Cognitive Assessment (MOCA) scores in the svMCI-D. The fALFF of right orbital part of superior frontal gyrus was significantly correlated with Hamilton Depression Scale scores in the svMCI+D, whereas the fALFF of right post cingulate cortex (PCC.R) showed significant correlation with MOCA scores in the svMCI-D. Furthermore, RSFC between PCC.R and right precuneus as well as between CERE.R and right lingual gyrus (LING.R) were significantly decreased in the svMCI-D compared to NC. For regional analyses, adjusted RSFC between PCC.R and PreCUN.R as well as between CERE.R and LING.R were decreased in the svMCI-D compared to NC but increased in the svMCI+D compared to svMCI-D. Further SVM analyses achieved good performances with area under curve (AUC) = 0.82 to classify the svMCI+D, svMCI-D and NC, AUC = 0.96 to classify svMCI+D and svMCI-D, AUC = 0.82 to classify svMCI+D and NC, and AUC = 0.92 to classify svMCI-D and NC. Conclusions: Our results showed disruptive effects of cognitive impairment, whereas both disruptive and complementary effects of depression symptoms on the spontaneous brain activity in the svMCI. Moreover, these features might be served as potential biomakers to classify each other among svMCI+D, svMCI-D, and NC, which can further be used to guide clinical treatment of svMCI, such as transcranial magnetic stimulation.