Determinants of the electric field during transcranial direct current stimulation

Highlights

• Evaluation of determining factors of the electric field distribution during tDCS

• Anatomy and electrode position explain up to 50% of the spatial variation.

• Anatomical factors such as a thin CSF layer can lead to stimulation hotspots.

• Useful to predict some of the field characteristics expected in practice

Abstract

Transcranial direct current stimulation (tDCS) causes a complex spatial distribution of the electric current flow in the head which hampers the accurate localization of the stimulated brain areas. In this study we show how various anatomical features systematically shape the electric field distribution in the brain during tDCS. We constructed anatomically realistic finite element (FEM) models of two individual heads including conductivity anisotropy and different skull layers. We simulated a widely employed electrode montage to induce motor cortex plasticity and moved the stimulating electrode over the motor cortex in small steps to examine the resulting changes of the electric field distribution in the underlying cortex. We examined the effect of skull thickness and composition on the passing currents showing that thinner skull regions lead to higher electric field strengths. This effect is counteracted by a larger proportion of higher conducting spongy bone in thicker regions leading to a more homogenous current over the skull. Using a multiple regression model we could identify key factors that determine the field distribution to a significant extent, namely the thicknesses of the cerebrospinal fluid and the skull, the gyral depth and the distance to the anode and cathode. These factors account for up to 50% of the spatial variation of the electric field strength. Further, we demonstrate that individual anatomical factors can lead to stimulation “hotspots” which are partly resistant to electrode positioning. Our results give valuable novel insights in the biophysical foundation of tDCS and highlight the importance to account for individual anatomical factors when choosing an electrode montage.

Introduction

Transcranial direct current stimulation (tDCS) is a widely used brain stimulation technique with various applications in different areas of neuroscience and clinical research. It is typically employed by attaching two large pad electrodes (with a few centimeters edge length) to the head and passing a weak electric current in the range of a few mA through them. It has been shown that tDCS can induce changes on motor cortex excitability (Nitsche and Paulus, 2000) that are dependent on stimulation strength and duration (Nitsche and Paulus, 2001). In slice preparations it has been demonstrated that weak DC fields can shift thresholds for action potential generation and exert an influence on spike timing (Bikson et al., 2004, Radman et al., 2007). While the exact mechanisms of action are still under discussion it is generally agreed that the local electric field and its orientation with respect to neuronal structures are main determinants of the stimulation effects that differ between various stimulation (Miranda et al., 2013, Radman et al., 2009). As direct measurements of these electric fields are difficult to implement various efforts have been made to estimate the electric field distribution by means of computational modeling. Simulation approaches range from spherical models (Datta et al., 2008, Miranda et al., 2006) to more realistic MRI derived head models (Datta et al., 2009, Sadleir et al., 2010) in order to demonstrate the effects of electrode shape, conductivity anisotropy (Suh et al., 2012), different skull layers (Neuling et al., 2012) or artificial skull openings (Datta et al., 2010) on the injected electric field.

However, few studies have tried to quantify how much the spatial distribution of the electric field on the cortical surface is pre-determined by individual anatomical features as well as electrode position. That is, to which extent does individual anatomy in addition (or opposing) to the electrode placement dictate the stimulated brain areas?

In the current study we explore the impact of several anatomical factors such as skull thickness and sulcal depth in combination with effects of electrode position on the electric field pattern in the brain. We employ anatomically realistic FEM models that are based on MR images and accurately capture the gyrification of gray matter (GM), the conductivity anisotropy of white matter (WM), the thickness of the cerebrospinal fluid (CSF) layer, the different skull layers, the eye regions and the nasal cavities. We focus on simulating the electric field distribution that might occur during a standard tDCS experiment aimed at inducing motor cortex plasticity. By investigating the effects of systematic displacements of the electrode position on the electric field distribution, we characterize how anatomical constraints interact with the electrode placement. In particular, we assess how stable the stimulated areas in the brain are when varying electrode positions. We demonstrate that our results are consistent across FEM models of two individual heads and are robust across a wide range of simulated electrode thicknesses and conductivities.