Objective: Despite the encouraging pilot results of transcranial direct current stimulation (tDCS) revealing its effectiveness in neuromodulation, there are also studies reporting inconsistent outcomes. Apart from previously studied factors, such as the differences in head model structures, anodal displacements, electrode shape and size, and connector position, the hypothesis that the inevitable spatial mismatch between the electrolyte buffer and electrode might shape current flow in the cerebral cortex was tested in this work, and our results potentially explain some of the reported inconsistent outcomes.
Approach: A finite element head model was built using cylinder electrodes with an arbitrary diameter of 2 cm. Current flow induced by different spatial mismatch types, degrees, and directions was simulated for three montages targeting the left motor cortex. Voxel-level current density differences and Jaccard index values of different percentiles for each mismatched configuration were calculated and compared throughout the cerebral cortex to determine the effect of electrode-electrolyte geometric mismatch.
Main results: Spatial mismatch between the electrolyte buffer and electrode affected the current density distribution in the cerebral cortex to different extents, depending on the position of the return electrode and mismatch type, degree, and direction. Single cortical voxel current-density variance induced by the 50% excess or insufficient mismatch was as high as 14.44% or 38.04%, respectively. Moreover, the distribution of variance changed directionally with the mismatch orientation. Compared with the insufficient mismatch and single-directional mismatch, the excessive and symmetrical mismatch caused a less obvious effect on the current density distribution of tDCS. Specifically, the symmetrical excess electrolyte caused around 2%-4% current density changes for all the montages, with different degrees and directions of mismatch. When the target position was fixed at C3, maximum sensitivity to the electrolyte-electrode mismatch was achieved with Iz as the return electrode, compared with the other two choices. Further, more than 20% voxels with >90% percentile of the peak current density values would shift position if >30% insufficient geometric mismatch occurred for montage C3-Iz.
Significance: Our findings suggest that special attention is required regarding the spatial matching of the electrolyte buffer to the electrode during tDCS to avoid unexpected large changes in current distribution.