P 18. Transcranial direct current stimulation reverses stroke-induced network alterations in mice

Abstract

Introduction. Transcranial direct current stimulation (tDCS) non-invasively modulates cortical function. We and others reported that tDCS could promote recovery after stroke ( Braun et al., Exp. Neurol., 2016; Hummel et al., Brain, 2005 ). As a stroke affects entire networks, an aspect long neglected in translational research, we recently established a resting-state functional magnetic resonance imaging (rs-fMRI) protocol to detect similar network alterations in both rodent and human stroke ( Blaschke et al., under review ). We here investigated whether rs-fMRI can validate the efficacy of tDCS in altering functional connectivity after experimental stroke. Methods. Focal cortical stroke was induced via photothrombosis in male C57Bl/6J mice. Three days thereafter, cathodal tDCS over the lesioned sensory-motor cortex was applied for 15 minutes each day for ten days (charge density of 39600 C/m 2 ). Longitudinal assessment of functional connectivity as measured by rs-fMRI was performed for up to 28 days post-stroke using a 9.4 T Bruker small animal scanner. Images were registered to the Allen Mouse Brain Atlas to derive seed regions for graph-based analysis and global graph parameters of network integration (characteristic path length, CPL) and segregation (clustering coefficient, CC) were computed over a sparse density range (10 – 30 % density), mimicking the brains“ functional structure. Additionally, a multiple linear regression was fitted to predict motor recovery based on initial network alteration, initial motor deficit, and tDCS treatment on motor recovery. Results. Ischemia induced an increase in connectivity. Already detectable three days after stroke, effects peaked at day 14, and spontaneously declined by day 28. Moreover, there was a significant reduction in characteristic path length (CPL) 14 days after stroke (p < 0.01). Intriguingly, ten days of cathodal tDCS treatment reversed the ischemia-induced hyperconnectivity, with reversal of ischemia-induced reduction in CPL ( p < 0.05). Networks effect correlated to tDCS accelerated motor recovery within the observation period. On the individual level, multiple linear regression model (F (3,22) = 11.1, p < 0.001, adjusted R 2 = .54) revealed, that - besides the initial motor deficit ( p < 0.01) - functional network alteration ( p < 0.01) and the treatment with tDCS ( p < 0.01) significantly improved prediction of motor recovery after stroke. Conclusion. Stroke induces characteristic network changes that can be detected by rs-fMRI. Importantly, these network changes can, at least in part, be reversed by tDCS. Besides, the data suggest that early markers of network impairment improve the prediction of motor recovery. Overall, the data suggest that the characterization of functional connectivity by rs-fMRI is a powerful tool to analyze global changes after experimental stroke and to assess the therapeutic effects of non-invasive brain stimulation in preclinical studies.