The impact of individual stroke lesions on tDCS current flow compared to neurotypical age-matched controls

Abstract Transcranial direct current stimulation (tDCS) has promise as a tool to aid rehabilitation after stroke by altering cortical excitability to promote neuroplasticity (Kubis, 2016; Ward, 2017). However, tDCS outcomes are subject to high variability, an effect which is amplified in stroke. Current flow models (CFMs) can be used to estimate tDCS electric field (E-field) in the brain, though they do not account for brain lesions (Lee et al., 2021). Previous work investigating the impact of stroke lesions on tDCS current flow is limited by small sample size or simulated lesions which may not reflect the broader stroke population (Piastra et al., 2021; Johnstone et al., 2021; Minjoli et al., 2017). Here we describe the impact of manually segmented individual stroke lesions on tDCS E-field when a commonly used symptom-targeted tDCS protocol is used. We compare our findings to a neurotypical age-matched control group. 123 stroke survivor and 147-age matched control MRI scans from the ENIGMA Stroke Recovery working group database were provided with manually segmented lesion masks. Inclusion criteria for stroke survivor scans were upper limb impairment (<66 FMA >28), unilateral cortical or subcortical stroke, and 1mm3 MRI collected >3 months post-stroke. A classic bipolar electrode montage (C3-Fp1) was simulated, targeting the primary motor cortex (M1). An adapted version of ROAST CFM software (Huang et al., 2019, 2018; Johnstone et al., 2021) was used to account for manually segmented stroke lesions. Lesion conductivity was specified (McCann et al., 2019). E-field intensity in the M1 target is more variable in stroke compared to control scans, even though the average E-field was comparable between groups. Within the stroke survivor group, lesion size, location, and proximity to the M1 ROI impacted current flow in M1. These findings highlight the necessity for individual CFM-informed protocol design, especially for tDCS application in stroke. Research Category and Technology and Methods Clinical Research: 9. Transcranial Direct Current Stimulation (tDCS) Keywords: Computational modelling, tDCS, Stroke, Non-invasive brain stimulation