Repetitive transcranial magnetic stimulation regulates effective connectivity patterns of brain networks in the spectrum of preclinical Alzheimer's disease

Abstract

Objectives: Subjective cognitive decline (SCD) and amnestic mild cognitive impairment (aMCI) are considered as the spectrum of preclinical Alzheimer's disease (AD), with abnormal brain network connectivity as the main neuroimaging feature. Repetitive transcranial magnetic stimulation (rTMS) has been proven to be an effective non-invasive technique for addressing neuropsychiatric disorders. This study aims to explore the potential of targeted rTMS to regulate effective connectivity within the default mode network (DMN) and the executive control network (CEN), thereby improving cognitive function.

Methods: This study included 86 healthy controls (HCs), 72 SCDs, and 86 aMCIs. Among them, 10 SCDs and 11 aMCIs received a 2-week rTMS course of 5-day, once-daily. Cross-sectional analysis with the spectral dynamic causal model (spDCM) was used to analyze the DMN and CEN effective connectivity patterns of the three groups. Afterwards, longitudinal analysis was conducted on the changes in effective connectivity patterns and cognitive function before and after rTMS for SCD and aMCI, and the correlation between them was analyzed.

Results: Cross-sectional analysis showed different effective connectivity patterns in the DMN and CEN among the three groups. Longitudinal analysis showed that the effective connectivity pattern of the SCD had changed, accompanied by improvements in episodic memory. Correlation analysis indicated a negative relationship between effective connectivity from the left angular gyrus (ANG) to the anterior cingulate gyrus and the ANG.R to the right middle frontal gyrus, with visuospatial and executive function, respectively. In patients with aMCI, episodic memory and executive function improved, while the effective connectivity pattern remained unchanged.

Conclusion: This study demonstrates that PCUN-targeted rTMS in SCD regulates the abnormal effective connectivity patterns in DMN and CEN, thereby improving cognition function. Conversely, in aMCI, the mechanism of improvement may differ. Our findings further suggest that rTMS is more effective in preventing or delaying disease progression in the earlier stages of the AD spectrum.

Clinical trial registration: http://www.chictr.org.cn, ChiCTR2000034533.

Keywords: amnestic mild cognitive impairment; dynamic causal model; effective connectivity; repetitive transcranial magnetic stimulation; subjective cognitive decline.

Figure 1

The spatial pattern of the (A) DMN and (B) CEN. (A) The spatial pattern of the DMN. (B) The spatial pattern of the CEN. DMN, default mode network; CEN, central executive network.

Figure 2

Spatial location of ROIs for DCM analyses. (A) DMN ROIs: PCUN, bilateral ANG, ACG, MFG.L, MTG.L, bilateral IOG. (B) CEN ROIs: ANG.R, IPL.L, MTG.L, bilateral MFG. ROI, region of interest; DCM, dynamic causal model; DMN, default mode network; CEN, central executive network; PCUN, precuneus; ANG, angular gyrus; ACG, anterior cingulate gyrus; MFG, middle frontal gyrus; MTG, middle temporal gyrus; IOG, inferior occipital gyrus; ANG, angular gyrus; IPL, inferior parietal lobule; L, left; R, right.

Figure 3

Mean values in effective connectivity for HC, patients with SCD and aMCI. (A) DMN effective connectivity patterns of HCs, SCD and aMCI. (B) CEN effective connectivity patterns of HCs, SCD and aMCI. HC, healthy controls; SCD, subjective cognitive decline; aMCI, amnestic mild cognitive impairment; DMN, default mode network; CEN, central executive network; PCUN, precuneus; ANG, angular gyrus; ACG, anterior cingulate gyrus; MFG, middle frontal gyrus; MTG, middle temporal gyrus; IOG, inferior occipital gyrus; IPL, inferior parietal lobule; L, left; R, right.

Figure 4

Group differences in effective connectivity HC, patients with SCD, and aMCI in DMN. (A) Differences in effective connectivity in patients with SCD compared to HC. A bar chart indicating the quantitative comparison of effective connectivity between these regions. (B) Differences in effective connectivity in patients with aMCI compared to HC. A bar chart indicating the quantitative comparison of effective connectivity between these regions. (C) Differences in effective connectivity in patients with aMCI compared to SCD. A bar chart indicating the quantitative comparison of effective connectivity between these regions. *Significant different (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001); error bar, standard error of the mean (SEM). HC, healthy controls; SCD, subjective cognitive decline; aMCI, amnestic mild cognitive impairment; DMN, default mode network; PCUN, precuneus; ANG, angular gyrus; ACG, anterior cingulate gyrus; MFG, middle frontal gyrus; MTG, middle temporal gyrus; L, left; R, right.

Figure 5

Group differences in effective connectivity HC, patients with SCD, and aMCI in CEN. (A) Differences in effective connectivity in patients with SCD compared to HC. A bar chart indicating the quantitative comparison of effective connectivity between these regions. (B) Differences in effective connectivity in patients with aMCI compared to HC. A bar chart indicating the quantitative comparison of effective connectivity between these regions. (C) Differences in effective connectivity in patients with aMCI compared to SCD. A bar chart indicating the quantitative comparison of effective connectivity between these regions. *Significant different (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001); error bar, standard error of the mean (SEM). HC, healthy controls; SCD, subjective cognitive decline; aMCI, amnestic mild cognitive impairment; CEN, central executive network; IPL, inferior parietal lobule; ANG, angular gyrus; MFG, middle frontal gyrus; MTG, middle temporal gyrus; L, left; R, right.

Figure 6

Effective connectivity changes in DMN and CEN before and after rTMS in patients with SCD. (A) Effective connectivity changes in DMN. A bar chart indicating the quantitative comparison of effective connectivity between these regions. (B) Effective connectivity changes in CEN. A bar chart indicating the quantitative comparison of effective connectivity between these regions. *Significant different (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001); error bar, standard error of the mean (SEM). DMN, default mode network; CEN, central executive network; SCD, subjective cognitive decline; rTMS, repetitive transcranial magnetic stimulation; ANG, angular gyrus; PCUN, precuneus; ACG, anterior cingulate gyrus; MFG, middle frontal gyrus; IOG, inferior occipital gyrus; L, left; R, right.

Figure 7

Relationship between altered effective connectivity and cognitive function in HC, SCD, aMCI, before rTMS and after rTMS. (A) Relationship between effective connectivity of the MFG.L to MTG.L and episodic memory in patients with SCD and aMCI in the DMN. (B) Relationship between effective connectivity of the IPL.L to ANG.R and visuospatial function in patients with HC and SCD in the CEN. (C) Relationship between effective connectivity of the ANG.L to ACG and visuospatial function in patients with the SCD before and after rTMS in the DMN. (D) Relationship between effective connectivity of the ANG.R to MFG.R and executive function in patients with the SCD before and after rTMS in the CEN. HC, healthy controls; SCD, subjective cognitive decline; aMCI, amnestic mild cognitive impairment; rTMS, repetitive transcranial magnetic stimulation; DMN, default mode network; CEN, central executive network; EM, episodic memory; VF, visuospatial function; EF, executive function. MFG, middle frontal gyrus; MTG, middle temporal gyrus; IPL, inferior parietal lobule; ANG, angular gyrus; ACG, anterior cingulate gyrus; L, left; R, right.