Poster No:
235
Submission Type:
Abstract Submission
Authors:
Elisabeth Dirren1, Julian Klug1, Cecilia Jarne2, Emmanuel Carrera1
Institutions:
1University Hospital Geneva, Geneva, Switzerland, 2Aarhus University CFIN /Universidad Nacional de Quilmes/ CONICET, Aarhus, Aarhus C
First Author:
Co-Author(s):
Julian Klug
University Hospital Geneva
Geneva, Switzerland
Cecilia Jarne, PhD
Aarhus University CFIN /Universidad Nacional de Quilmes/ CONICET
Aarhus, Aarhus C
Introduction:
Recurrent strokes are frequent, occurring in up to 10 % of patients within 3 months of the initial event. Whether and how the brain reorganizes to limit the consequences of a second event is largely unknown (van Assche et al, 2022). In fact, most studies investigating the physiological changes that occur after stroke have focused on the neural correlates of recovery, disregarding in turn, the processes that may increase brain resilience to further attacks.
Here we used a large dataset of first-time stroke patients with resting-state connectivity assessed at three time-points within 1 year of stroke to determine how brain networks reconfigure to prevent the consequences of new lesions.
Methods:
75 first-time stroke patients and 18 healthy controls were included from a large dataset of stroke patients (Corbetta et al, 2015). Gradient echo EPI resting-state functional images and T1 structural images were obtained in healthy subjects, and at three time-points in patients: within 1-2 weeks (TP1), at three months (TP2) and at one year (TP3). After atlasing brain images using the Brainnetome atlas (Fan et al, 2016), connectivity matrices were built for each control, patient and time-point by computing Pearson correlations.
We investigated resilience to recurrent strokes by evaluating changes in two graph metrics that capture network integration (global efficiency) and specialization (modularity)(Rubinov et al, 2010). Virtual lesions were applied to patients and controls' connectivity matrices by removing Brainnetome regions that had at least 50% overlap with lesion masks from 122 stroke patients taken from the present cohort and an additional in-house cohort of stroke patients (Klug et al, 2021). Global efficiency and modularity were recalculated following node deletion. We defined resilience (R) as the difference between pre- and post-virtual lesion measures. R was computed and normalized to the controls' mean R, to yield Rnorm values for each metric. Mixed linear models were built to statistically compare controls and patients at all three time-points. FDR correction was applied for multiple comparisons.
Lesion and patient-specific modulators of brain resilience after virtual strokes were evaluated by building a mixed linear model with lesion size, site, side, patient age handedness, gender and acute NIHSS (a clinical stroke scale) as fixed factors and either Rnorm(global efficiency) or Rnorm(modularity) as dependent variable.
Results:
We observed increased resilience in brain networks of stroke patients, with a lower impact of virtual lesions on global efficiency and modularity. Rnorm(global efficiency) was significantly higher in patients at TP1 (0.133, p=0.04) and TP2 (0.135, p=0.04) but not TP3 (0.023, p= 0.854), compared to controls (0.000). Similarly, Rnorm(modularity) was higher in patients at TP1 (0.661, p<0.001), TP2 (0.316, p=0.073) and TP3 (0.456, p=0.007), compared to controls (0.000). Lesion side, lesion site, patient age, acute NIHSS, gender and handedness modulated resilience to recurrent virtual strokes, but not lesion size.
Conclusions:
Network reorganization after stroke strengthens resilience to recurrent lesions. More specifically, this reconfiguration limits the impact of recurrent virtual lesions on integration and specialization of brain networks. Both lesion and patient-specific characteristics modulated resilience. These results suggest that specific reorganization features in brain network architecture after stroke are not only associated with clinical improvement but also with reinforcement of resilience of brain networks to future lesions.
Disorders of the Nervous System:
Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) 1
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural)
fMRI Connectivity and Network Modeling 2
Keywords:
Cerebrovascular Disease
FUNCTIONAL MRI
Other - Resilience
1|2Indicates the priority used for review
Provide references using author date format
1. van Assche M, Klug J, Dirren E, Richiardi J, Carrera E. Preparing for a Second Attack: A Lesion Simulation Study on Network Resilience After Stroke. Stroke 2022;53:2038-2047.
2. Corbetta M, Ramsey L, Callejas A, et al. Common behavioral clusters and subcortical anatomy in stroke. Neuron 2015;85:927-941.
3. Fan L, Li H, Zhuo J, et al. The Human Brainnetome Atlas: A New Brain Atlas Based on Connectional Architecture. Cereb Cortex 2016;26:3508-3526.
4. Rubinov M, Sporns O. Complex network measures of brain connectivity: uses and interpretations. Neuroimage 2010;52:1059-1069.
5. Klug J, Dirren E, Preti MG, et al. Integrating regional perfusion CT information to improve prediction of infarction after stroke. J Cereb Blood Flow Metab 2021;41:502-510.