How does transcranial DC stimulation of the primary motor cortex alter regional neuronal activity in the human brain?

Abstract

Transcranial direct current stimulation (tDCS) of the primary motor hand area (M1) can produce lasting polarity-specific effects on corticospinal excitability and motor learning in humans. In 16 healthy volunteers, O positron emission tomography (PET) of regional cerebral blood flow (rCBF) at rest and during finger movements was used to map lasting changes in regional synaptic activity following 10 min of tDCS (+/-1 mA). Bipolar tDCS was given through electrodes placed over the left M1 and right frontopolar cortex. Eight subjects received anodal or cathodal tDCS of the left M1, respectively. When compared to sham tDCS, anodal and cathodal tDCS induced widespread increases and decreases in rCBF in cortical and subcortical areas. These changes in rCBF were of the same magnitude as task-related rCBF changes during finger movements and remained stable throughout the 50-min period of PET scanning. Relative increases in rCBF after real tDCS compared to sham tDCS were found in the left M1, right frontal pole, right primary sensorimotor cortex and posterior brain regions irrespective of polarity. With the exception of some posterior and ventral areas, anodal tDCS increased rCBF in many cortical and subcortical regions compared to cathodal tDCS. Only the left dorsal premotor cortex demonstrated an increase in movement related activity after cathodal tDCS, however, modest compared with the relatively strong movement-independent effects of tDCS. Otherwise, movement related activity was unaffected by tDCS. Our results indicate that tDCS is an effective means of provoking sustained and widespread changes in regional neuronal activity. The extensive spatial and temporal effects of tDCS need to be taken into account when tDCS is used to modify brain function.

Figures

Fig. 1

Panel A illustrates the experimental design. Subjects were divided into two groups of eight and received Real-tDCS (anodal or cathodal) and Sham-tDCS on separate days. Immediately afterwards six sequential H215O-PET scans were acquired at rest (R) or during finger movements (M). The order of intervention (Real-tDCS vs. Sham-tDCS) and experimental conditions (R vs. M) were counterbalanced across subjects. Panel B illustrates the technique used for tDCS. Weak direct current (1 mA) was applied between two large (35 cm2), wet sponge-electrodes placed over left M1 (optimal representation of right FDI as assessed with TMS) and right frontopolar cortex (above the eyebrow). Polarity of tDCS refers to the M1 electrode. Panel C illustrates the motor task performed by the subjects during PET scanning. Subjects were required to freely select from a set of four previously practised movements and execute brisk flexion movements with fingers II–V of their right hand. They were asked to make a fresh choice on each trial, regardless of previous moves. Movements were paced every 2 s to ensure a constant movement rate across scans.

Fig. 2

Changes in movement related activity. Surface rendering of those voxels showing enhanced movement related activity in left dorsal premotor cortex (PMd) after cathodal tDCS (red area; P < 0.001; uncorrected). The yellow areas indicate the motor and frontopolar site of electrode placement during tDCS. The graph (left panel) plots the relative changes in rCBF among the four experimental conditions regarding after effects of cathodal tDCS. Error bars equal SEM.

Fig. 3

Main effect of anodal and cathodal tDCS compared to sham. Surface rendering of brain regions showing a relative increase or decrease rCBF after real-tDCS compared to sham tDCS (P < 0.05, whole-brain corrected). Images show (from the top) right lateral surface, left lateral surface, right medial surface, and left medial surface.

Fig. 4

Differences and commonalities in rCBF changes induced by anodal and cathodal tDCS. Surface rendered statistical parametric maps showing brain regions in which anodal and cathodal tDCS had a differential (left panel) or similar (right panel) effect in terms of lasting changes in rCBF (P < 0.05, whole-brain corrected). Images show (from the top) right lateral surface, left lateral surface, right medial surface, and left medial surface.

Fig. 5

Increases in rCBF in the cortex underlying the electrodes. The left panels highlights those voxels in left SM1 (coronal slice) and right frontopolar cortex (axial slice) showing an increase in rCBF both during anodal and cathodal tDCS (P < 0.05, whole-brain corrected). The graphs (right panels) plot the relative changes in rCBF in the voxel showing peak increases in rCBF during tDCS regardless of polarity. Each column represents mean normalized rCBF for the eight experimental conditions. Error bars equal SEM.