Disruptive and complementary effects of depression symptoms on spontaneous brain activity in the subcortical vascular mild cognitive impairment
Liyu Hu,Jianxiang Chen,Xinbei Li,Haoran Zhang,Jinhuan Zhang,Yingqi Lu,Jie Lian,Haibo Yu,Nan Yang,Jianjun Wang,Hanqing Lyu,Jinping Xu
DOI: https://doi.org/10.3389/fnagi.2024.1338179
2024-09-17
Abstract:Background: Although depression symptoms are commonly reported in patients with subcortical vascular mild cognitive impairment (svMCI), their impact on brain functions remains largely unknown, with diagnoses mainly dependent on behavioral assessments. Methods: In this study, we analyzed resting-state fMRI data from a cohort of 34 svMCI patients, comprising 18 patients with depression symptoms (svMCI+D) and 16 patients without (svMCI-D), along with 34 normal controls (NC). The study used the fraction of the amplitude of low-frequency fluctuations (fALFF), resting-state functional connectivity, correlation analyses, and support vector machine (SVM) techniques. Results: The fALFF of the right cerebellum (CERE.R) differed among the svMCI+D, svMCI-D, and NC groups. Specifically, the regional mean fALFF of CERE. R was lower in svMCI-D patients compared to NC but higher in svMCI+D patients compared to svMCI-D patients. Moreover, the adjusted fALFF of CERE. R showed a significant correlation with Montreal Cognitive Assessment (MOCA) scores in svMCI-D patients. The fALFF of the right orbital part of the superior frontal gyrus was significantly correlated with Hamilton Depression Scale scores in svMCI+D patients, whereas the fALFF of the right postcingulate cortex (PCC.R) showed a significant correlation with MOCA scores in svMCI-D patients. Furthermore, RSFC between PCC. R and right precuneus, as well as between CERE. R and the right lingual gyrus (LING.R), was significantly reduced in svMCI-D patients compared to NC. In regional analyses, the adjusted RSFC between PCC. R and PreCUN. R, as well as between CERE. R and LING. R, was decreased in svMCI-D patients compared to NC but increased in svMCI+D patients compared to svMCI-D. Further SVM analyses achieved good performances, with an area under the curve (AUC) of 0.82 for classifying svMCI+D, svMCI-D, and NC; 0.96 for classifying svMCI+D and svMCI-D; 0.82 for classifying svMCI+D and NC; and 0.92 for classifying svMCI-D and NC. Conclusion: The study revealed disruptive effects of cognitive impairment, along with both disruptive and complementary effects of depression symptoms on spontaneous brain activity in svMCI. Moreover, these findings suggest that the identified features might serve as potential biomarkers for distinguishing between svMCI+D, svMCI-D, and NC, thereby guiding clinical treatments such as transcranial magnetic stimulation for svMCI.