Connectivity Correlates of Ataxia After Stroke
Poster No:
1806
Submission Type:
Abstract Submission
Authors:
ASLI AKDENIZ-KARATAY1, Ana Sofía Ríos2, Uchralt Temuulen3, Ahmed Khalil2, Anna Kufner2
Institutions:
1Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany, 2Center for Stroke Research Berlin (CSB), Berlin, Germany, 3Freie Universität Berlin, Berlin, Germany
First Author:
Co-Author(s):
Introduction:
Ataxia is a motor coordination impairment primarily related to cerebellar dysfunction.1 About 50% to 65% of individuals with cerebellar stroke develop ataxia;2 however ataxia can originate from non-cerebellar injury as well.3 This could be explained by diaschisis, a term that describes functional impairment beyond focal brain damage in distant but connected brain areas. The study of diaschisis has broadened the understanding of the intricate dynamics involved in neurological and psychiatric symptoms. Lesion Network Mapping (LNM) is a method for utilizing resting-state functional connectivity measurements to unveil brain circuits underlying symptoms that emerge after injury to the brain. Integrating normative connectome data with symptom mapping after stroke allows LNM to provide insights into the relationship between lesion-inclusive networks and symptoms, as has been shown in conditions like Parkinsonism and tics,4,5 which can inform targets for neuromodulation therapies.6 This could be useful for post-stroke ataxia since no treatments for this condition have been approved by the FDA.6
This study aims to identify a network of functionally interconnected brain regions associated with post-stroke ataxia.
This study aims to identify a network of functionally interconnected brain regions associated with post-stroke ataxia.
Methods:
Here, we retrospectively analyzed a cohort of 380 patients with acute ischemic stroke (140 female, mean age: 65.8 ± 13.4 years) who underwent neurological examination through the National Institutes of Health Stroke Scale (NIHSS).7 Limb ataxia, specifically assessed using item 7 of the NIHSS scale, was evaluated on both sides through the finger-nose-finger and heel-shin tests, scoring dependent on its manifestation to a degree disproportionate to weakness.
For LNM, the stroke lesions were manually delineated from DWI images of the subjects and registered to common space according to Montreal Neurological Institute (MNI152 atlas, 2 × 2 × 2 mm) using the FSL FLIRT function. The resulting lesion masks were used as seeds to calculate functional connectivity from the lesion to every other voxel in the brain, based on a normative connectome from rs-fMRI data from 1,000 healthy subjects,8 using the lead-mapper tool from lead-dbs.org.9 The significance of the connections was validated by a non-parametric approach, using randomise from the FSL software.10 Ataxia scores (NIHSS, item 7) were transformed into binary variables to accommodate cases with or without ataxia and used as the independent variable, while connectivity profiles were used as the dependent variable. Two contrasts were established to compare connectivity to lesions causing ataxia(i) and to lesions not causing ataxia (ii), n=67 and n=313, respectively. We applied the family-wise error rate (FWE) to correct for multiple comparisons. To identify significant connections for each group, regions with p<0.05 were considered.
For LNM, the stroke lesions were manually delineated from DWI images of the subjects and registered to common space according to Montreal Neurological Institute (MNI152 atlas, 2 × 2 × 2 mm) using the FSL FLIRT function. The resulting lesion masks were used as seeds to calculate functional connectivity from the lesion to every other voxel in the brain, based on a normative connectome from rs-fMRI data from 1,000 healthy subjects,8 using the lead-mapper tool from lead-dbs.org.9 The significance of the connections was validated by a non-parametric approach, using randomise from the FSL software.10 Ataxia scores (NIHSS, item 7) were transformed into binary variables to accommodate cases with or without ataxia and used as the independent variable, while connectivity profiles were used as the dependent variable. Two contrasts were established to compare connectivity to lesions causing ataxia(i) and to lesions not causing ataxia (ii), n=67 and n=313, respectively. We applied the family-wise error rate (FWE) to correct for multiple comparisons. To identify significant connections for each group, regions with p<0.05 were considered.
Results:
The connections to lesions of patients with ataxia were distributed across all vascular territories of the brain. A map of connectivity to each group was obtained; after thresholding to pFWE<0.05 to isolate the significant voxels connected to each group, we found that lesions associated with ataxia were mainly connected to the cerebellum, specifically to Crus I and II, as well as lobule VI in both hemispheres. The group without ataxia lesions showed connectivity to the primary motor cortex, premotor areas, and somatosensory cortex.
·Strongest connections to lesions associated with ataxia after stroke
Conclusions:
In conclusion, this study reveals that brain lesions in variable vascular territories of the brain are functionally connected to cerebellar lobules, underlying the importance of this brain structure in ataxic symptoms, providing insights for potential neuromodulatory therapies by understanding the impact of stroke on cortico-cerebellar networks. The study also found that the commonly used NIHSS can detect indirect cerebellar damage effectively.
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural) 2
fMRI Connectivity and Network Modeling 1
Novel Imaging Acquisition Methods:
BOLD fMRI
Keywords:
Cerebrovascular Disease
FUNCTIONAL MRI
Other - stroke
1|2Indicates the priority used for review
Provide references using author date format
Go S. (2015). Posterior Circulation Ischemic Stroke. 112(3):192-6.
Fukutake T. et al. (2023). 63(4):201-208
Joutsa J, Horn A, Hsu J, Fox MD. (2018) Brain J Neurol. 141(8):2445-245
Ganos C, Al-Fatly B, Fischer JF, et al. (2020) Brain J Neurol. 145(12):4385-4397.
Grimaldi G et al. (2014).Cerebellum.13(1):121-38.
Malsch C et al. (2018) Outcome after stroke attributable to baseline factors—The PROSpective Cohort with Incident Stroke (PROSCIS). PLOS ONE. e0204285.
Holmes, A., Hollinshead, M., O’Keefe, T. et al. (2015). Sci Data 2, 150031.
Neudorfer C, Butenko K, Oxenford S, Rajamani N, Achtzehn J, Goede L, Hollunder B, Ríos AS, Hart L, Tasserie J, Fernando KB, Nguyen TAK, Al-Fatly B, Vissani M, Fox M, Richardson RM, van Rienen U, Kühn AA, Husch AD, Opri E, Dembek T, Li N, Horn A. (2023). Neuroimage 268:119862.
Eklund A, Nichols TE, Knutsson H. (2016). Proc Natl Acad Sci. 113(28):7900-7905.