4128
Abnormalities of Cortical Morphology and Structural Covariance Network in Patients With Subacute Basal Ganglia Stroke1Radiology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China, 2MR Research, GE Healthcare, Beijing, China
Synopsis
The primary aim of this study was to examine the changes of cortical morphology in patients with unilateral basal ganglia stroke (BS) in subacute phase and evaluate the discrepancies of structural covariance networks (SCNs) between BS and HCs groups by using a seed-based structural covariance approach. The main findings were that BS could cause cortical atrophy of bilateral frontal and temporal lobes and abnormal structural covariance patterns, featured by decreased global efficiency and fragile topological properties in reaction to target attacks. These findings may enable us to better understand the neurobiological mechanisms of behavioral impairment and recovery after BS.
Introduction and Purpose
Lesion location of ischemic stroke is variable and is associated with functional outcome[1].The capsular region is the most common site for subcortical stroke involving the motor pathway. After occurrence of acute subcortical stroke, cortex surrounding the isolated subcortical lesions experienced gray matter atrophy, which is accompanied by changes of structural covariance network (SCN). The pattern of cortical reorganization largely depends on the location of the injury [2]. This research aims to study the pattern of cortical reorganization in the subacute phase of basal ganglia stroke (BS) so as to better understand brain's macrostructure changes after stroke and to provide a more customized diagnosis reference for rehabilitation therapy in the early stage of stroke.Methods
With approval of the Institutional Review Board and signed informed consent of all participants, 26 subacute stroke patients (stroke onset 2-14 days) whose infarcts were restricted to the basal ganglia and/or surrounding areas (involving thalamus, coronal radiata) and 25 age- and gender-matched healthy controls underwent MR examinations. T1-weighted 3D brain volume images were captured using a 3.0 T MR scanner (Discovery MR750, GE Health Care, Waukesha, WI, USA). The Computational Anatomy Toolbox (CAT12) embedded in the Statistical Parametric Mapping software (SPM12) was utilized to analyze the differences in cortical morphology (including cortical thickness, sulcus depth, gyrification index and cortical complexity) between the two groups. We constructed a SCN according to the Desikan-Killiany atlas with 68 parcels bilaterally. The Graph-Theoretical Analysis Toolbox (GAT) was used to perform the graph analysis [3].For the global network index, we analyzed the parameters both at Dmin and across the density range (0.11–0.45) using area under the curve (AUC). To explore the vulnerability of the SCNs to targeted and random attacks, we analyzed the size of remaining giant components of the network in response to the successively random or targeted removals of nodes.Results
Cortical morphological analysis showed that compared with healthy controls (HCs), patients with basal ganglia infarction had cortical thinning in certain areas (p<0.05, FWE corrected), mainly in the bilateral frontal and temporal cortices rather than in the motor cortex. In addition, the gyrification index also decreased in the ipsilateral orbitofrontal gyrus, insula and pars triangularis, overlapping with the thinned cortex. Global network analysis revealed significantly increased parameters of clustering coefficient and transitivity representing network segregation and decreased parameters of global efficiency representing network integration in the SCNs of BS patients. The AUC difference of global network parameter in the two groups was consistent with abovementioned description, meaning the decrease of the global efficiency and the increase of the transitivity and clustering coefficient in the BS group. Also, the SCNs of BS patients have reduced resistance to targeted attacks.Discussions and Conclusions
In this study, we investigated the differences in cortical morphology and SCNs in a cohort of patients with stroke lesions in the basal ganglia compared with healthy controls. In particular, those patients underwent MRI examinations in two weeks after stroke occurred. We found that the cortical thickness of the frontal and temporal cortex decreased in the early stage. The reduced structural measurements reflect secondary degeneration due to direct injury or indirect injury through the damage of corresponding white matter fibers in the cortex[4]. Another longitudinal study[5] showed that the cortical thickness increased from two weeks to six months after basal ganglia stroke, which may reflect the mechanism of motor recovery and motor compensation[6]. These findings indicate that these above mentioned areas may be potential targets for neuromodulation. Neuromodulation reflects structural changes during the course of the basal ganglia stroke and may lead to plasticity changes in some important cortical-striatal pathways, which is conducive to the recovery of motor function.Therefore, consistent with our findings of decreased cortical thickness in basal ganglia and neighboring regions, it is essential to conduct follow-up examinations on our recruited subjects to prove the recovery of motor function.Besides the cross-sectional analysis of abnormal cortical morphology in basal ganglia stroke, the SCN analysis of the global measurements showed that the global efficiency of SCNs in BS patients was significantly reduced, both under the minimum density threshold and the AUC curve across multiple network densities. This reduction suggests that the SCNs in BS patients are sub-optimized, without an optimal balance between segregation and integration[7]. The network resilience analysis also showed significant decreased SCNs in BS patients in response to targeted attacks, implying a relatively fragile topology.
Overall, this study revealed the cortical reorganization pattern and structural covariant network changes in the subacute phase of basal ganglia stroke patients, providing evidence for establishing the relationship between cortical morphology and topology parameters and treatment options.
Acknowledgements
Funding: This project was supported by the National Natural Science Funds of China (Grants No.81730049).References
[1] CHENG B, FORKERT N D, ZAVAGLIA M, et al. Influence of stroke infarct location on functional outcome measured by the modified rankin scale [J]. Stroke, 2014, 45(6): 1695-702.
[2] JIANG L, LIU J, WANG C, et al. Structural Alterations in Chronic Capsular versus Pontine Stroke [J]. Radiology, 2017, 285(1): 214-22.
[3] HOSSEINI S M, HOEFT F, KESLER S R. GAT: a graph-theoretical analysis toolbox for analyzing between-group differences in large-scale structural and functional brain networks [J]. PLoS One, 2012, 7(7): e40709.
[4] DUERING M, RIGHART R, WOLLENWEBER F A, et al. Acute infarcts cause focal thinning in remote cortex via degeneration of connecting fiber tracts [J]. Neurology, 2015, 84(16): 1685-92.
[5] LIU H, PENG X, DAHMANI L, et al. Patterns of motor recovery and structural neuroplasticity after basal ganglia infarcts [J]. Neurology, 2020, 95(9): e1174-e87.
[6] JONES P W, BORICH M R, VAVSOUR I, et al. Cortical thickness and metabolite concentration in chronic stroke and the relationship with motor function [J]. Restor Neurol Neurosci, 2016, 34(5): 733-46.
[7] ALEXANDER-BLOCH A, GIEDD J N, BULLMORE E. Imaging structural co-variance between human brain regions [J]. Nat Rev Neurosci, 2013, 14(5): 322-36.
Figures
Table 1 | Results of group comparison of cortical thickness and gyrification index between BS and HCs groups.
*Based on Desikan-Killiany Atlas with 68 parcels bilaterally. BS, Basal ganglia stroke; HCs, Healthy controls.
