Alterations of static and dynamic changes in intrinsic brain activity and its relation to behavioral outcomes in subcortical ischemic stroke after one-month intervention

Abstract

Ischemic stroke is a prominent contributor to cognitive dysfunction and disability. Gaining a comprehensive understanding of the neuronal activity and longitudinal changes underlying stroke is crucial for designing effective rehabilitative strategies. However, the neural mechanisms responsible for the longitudinal reorganization of neuronal activity following stroke remain unclear. The objective of this study was to comprehensively investigate potential abnormalities in brain activity among stroke patients before and after one month of intervention (antiplatelet therapy, as well as intravenous citicoline). To achieve this goal, we combined static and dynamic functional imaging indicators for the comprehensive analysis. Twenty ischemic stroke patients at the subacute stage and seventeen age-matched healthy controls were included in the final analysis of this study from one center. Additionally, resting-state functional magnetic imaging scans were conducted on all patients twice with a one-month interval between scans. Four static intrinsic brain activity indicators (static amplitude of low-frequency fluctuation (sALFF), static fractional amplitude of low-frequency fluctuation (sfALFF), static regional homogeneity (sReHo), and static degree centrality (sDC)), along with their corresponding dynamic indicators, were calculated to detect longitudinal alterations in brain activity following stroke onset. Correlation analyses were also performed between these indicators within areas exhibiting group differences as well as clinical scale scores and disease duration. Significant variations in these static and dynamic image indicators were observed among patients with ischemic stroke. There was substantial overlap among the abnormal brain regions detected, primarily including decreased sALFF/sfALFF/dALFF in the bilateral central precuneus, increased sfALFF/sReHo/sDC/dReHo in the left superior precuneus, increased sALFF/sReHo/dfALFF in the left inferior temporal gyrus, decreased sReHo/sDC in the anterior cingulate cortex, increased sReHo/dfALFF in the right inferior parietal lobe, increased sfALFF/sDC in the right fusiform gyrus, as well as decreased sALFF/dALFF and increased sReHo/sDC in the right angular gyrus. Furthermore, these disrupted image indicators in some regions exhibited only partial recovery at the second time point. The percentage changes of these image indicators (sfALFF in the bilateral central precuneus, sDC in the left fusiform and dALFF in the right central precuneus) between the two time points were positively correlated with the percentage changes of clinical scores (FMA and MBI). In combination, this study demonstrates that a comprehensive understanding of abnormal activity and its longitudinal changes in ischemic stroke can be achieved by integrating static and dynamic imaging methods. Regions showing significant overlap among different brain activity indicators and exhibiting consistent image-behavior relationships may have some potential values for predicting clinical outcomes.

Keywords: Brain activity; Dynamic analysis; Longitudinal changes; Resting-state fMRI; Static analysis; Subcortical ischemic stroke.

Fig. 1

Lesion overlap map across stroke patients with left-sided lesions. Lesion maps were normalized to the MNI space and overlapped. The n values of color bar in this map represent the number of patients.

Fig. 2

Brain regions with significant intergroup differences in static image activity indictors between the stroke patients and the normal controls: (A) sALFF, (B) sfALFF, (C) sReHo, (D) sDC. The statistical threshold was set to p < 0.05 corrected with the Gaussian random field approach (the voxel level was p < 0.01). Data are mean ± SD. The color bar indicates the t-values. P_pre, stroke patients’ results at the first time point; P_post, stroke patients’ results at the second time point; HC, healthy control; sALFF, static amplitude of low-frequency fluctuation; sfALFF, static fractional ALFF; sReHo, static regional homogeneity; sDC, static degree centrality.

Fig. 3

Brain regions with significant intergroup differences in dynamic image activity indicators with window sizes of 30TRs between the stroke patients and the normal controls: (A) dALFF, (B) dALFF, (C) dReHo, (D) dDC. The statistical threshold was set to p < 0.05 corrected with the Gaussian random field approach (the voxel level was p < 0.01). Data are mean ± SD. The color bar indicates the t-values. P_pre, stroke patients’ results at the first time point; P_post, stroke patients’ results at the second time point; HC, healthy control; dALFF, dynamic amplitude of low-frequency fluctuation; dfALFF, dynamic fractional ALFF; dReHo, dynamic regional homogeneity; dDC, dynamic degree centrality.

Fig. 4

Scatter diagrams show the significantly correlations between the percentage changes of image indicators and the percentage changes of scale scores in stroke patients with one-month follow-up. (A) The percentage changes of static fALFF and dynamic ALFF in the right precuneus were positively correlated with the percentage changes of FMA and MBI. (B) The percentage changes of static fALFF in the left precuneus were positively correlated with the percentage changes of FMA and MBI. (C) The percentage changes of dynamic fALFF in the left superior precuneus were positively correlated with the percentage changes of FMA. (D-G) Scatter diagrams show the significantly correlations between the percentage changes of image indicators and the disease duration in stroke patients with one-month follow-up. A significance threshold of p < 0.05 was applied.