P 171. Bihemispheric motor cortex stimulation in older adults induces modulations of resting state and task-related activity

Abstract

Bihemispheric transcranial direct current stimulation (“dual” tDCS) of primary motor cortices has been described to enhance motor learning in healthy subjects and to facilitate motor recovery after stroke. In order to investigate the neural correlates of its mode of action, we compared different tDCS montages in a group of healthy older adults in a cross-over design (‘dual’ vs. ‘anodal’ vs. ‘sham’). 20 subjects (mean age 68.7 ± 4.7 years, all right-handed) underwent tDCS and simultaneous MRI at 3T, including resting state fMRI and a choice reaction time task. In the task, subjects were presented with different symbols in a randomized order and required to respond with button presses using either left or right index fingers. In both active stimulation conditions, the anode was placed over left primary motor cortex (M1). The cathode was positioned over right M1 (dual) or the contralateral supraorbital region (anodal). The current was constantly delivered during resting state and task-related fMRI. Task-related fMRI analysis was carried out with SPM8; resting state data were analyzed with Lipsia using low-frequency spectral Eigenvector Centrality Mapping (ECM). Task-specific analyses revealed differential effects of the two active tDCS conditions. Compared with anodal tDCS, dual tDCS yielded stronger activations in bilateral primary motor cortices when either the left or right index fingers were used. In the resting state analysis, ECM values in left prefrontal and cingulate cortices were higher in the dual condition as compared to sham. Anodal yielded higher ECM values in the left prefrontal as well as left ventral premotor cortex when compared with sham; lower values were found in right M1 and the left precuneus. Comparing the two active tDCS conditions, increased connectivity of the cingulate cortex and decreased connectivity in the right cerebellum was found under dual compared to anodal tDCS (all p < .001, Monte Carlo-corrected). In conclusion, tDCS induced specific changes of M1 activation in task-specific activations, dependent on the electrode set-up. As a complement, the resting state analysis demonstrated altered connectivity within a multimodal network including motor cortices as well as prefrontal regions, indicating that the previously documented stronger behavioral effects of dual as compared to anodal tDCS may not be merely mediated by a “simple” add-on effect of cathodal stimulation, but rather due to complex bihemispheric network modulations.