Investigating the effects of brain stimulation on the neural substrates of inhibition in patients with OCD: A simultaneous tDCS - fMRI study

Abstract

Inhibition deficits constitute a core characteristic of obsessive-compulsive disorder (OCD). There is evidence in healthy individuals that transcranial direct current stimulation (tDCS) of the pre-supplementary motor area (pre-SMA) leads to a significantly improved inhibition performance. Against this background we investigated the effects of pre-SMA tDCS on inhibition performance and the underlying neural correlates in patients with OCD. Using a double-blind, randomized, sham-controlled, cross-over design (i.e., tDCS sham vs. tDCS stimulation) we investigated the effects of 2 mA anodal tDCS stimulation of the right pre-SMA in a sample of 47 OCD patients. The present study is, to our best knowledge, the first study applying concurrent tDCS-fMRI in patients with OCD. tDCS was applied using the MRI-compatible NeuroConn DC-Stimulator which allowed for a concurrent stimulation, while patients performed an inhibition (i.e., Stroop) task in a 3 T MRI. Imaging data were analysed using a multivariate partial least squares (PLS) approach. tDCS stimulation (vs. sham) was associated with increased activation in a fronto-parieto-cerebellar network comprising, amongst others, the precentral, middle frontal and inferior frontal gyrus, the anterior cingulate and the superior parietal lobe. On the performance level, tDCS stimulation (vs. sham) was linked to an improved inhibition performance in terms of an increased percentage of correct responses in the Stroop task. Present results indicate that tDCS in patients with OCD goes along with an improved inhibition performance as well as activation increases in regions known to be involved in inhibition, motor, and cognitive control. Thus, our findings suggest that tDCS might be a promising method to improve specific impairments in OCD.

Fig. 1. Task Inhibition Measures.

a Schematic illustration of the Stroop task. b, c Inhibition performance outlined as % correct (congruent -incongruent) (b) and response time in millisecond (incongruent - congruent) (c) for sham and stimulation. A non - parametric paired Wilcoxon-signed-rank test was performed to compare performance in the two conditions.

Fig. 2. Electric field Magnitude Distribution.

Mean electric field at 40 different regions of interest (31 from the Desikan-Killiany-Tourville atlas and 9 from the Human Motor Area Template atlas), averaged across the two sessions, and normalised across all subjects for comparison (n = 47).

Fig. 3. Condition Effects, lag 6.

a Clusters where peak voxels have a salience/standard error ratio (i.e. Bootstrap ratio; BSR) of ≥5 or ≤−5 for the latent variable (LV) describing 100% of all the variance for our desired contrast comparing incongruent > congruent. Red clusters indicate increased activation for the incongruent condition, blue clusters illustrate increased activation for the congruent condition. Lag 6 represents the brain activity at 6 TRs post stimulus onset (i.e., 6 s post stimulus given our sequence with a TR of 1 s). B Temporal brain scores plot displaying the mean brain scores and standard deviation across all 47 subjects for each condition during the first 8 TRs post stimulus onset. Brain scores are subject, design-salience, lag and LV contrast specific. Note the positive and negative peaks are at around 5–6 s after stimulus onset.

Fig. 4. Condition-by-timepoint effects, lag 6.

Clusters where peak voxels have a salience/standard error ratio (i.e. Bootstrap ratio; BSR) of ≥3 or ≤−3 (p < 0.001) for the latent variable (LV) describing 100% of all the variance for our desired contrast comparing conditions and timepoints. Yellow, green, red and pink clusters (i.e., positive salience regions) indicate increased activation during tDCS compared to sham for the incongruent compared to the congruent condition. Blue clusters (i.e., negative salience regions) illustrate increased activation for the opposite contrast (i.e., increased activation during sham compared to tDCS for the incongruent compared to the congruent condition). Lag 6 represents the brain activity at 6 TRs post stimulus onset (this sequence had a TR of 1 s). n = 47 subjects. Results based on lags 5, 7, and 8 are reported in the supplementary material (figure S2).

Fig. 5. Mediation Analysis investigating potential effects of electric field strength and right PreSMA thickness (n = 47).

The p-values are depicted on the lines linking the associations and mediations.