Frequency-dependent functional connectivity of the nucleus accumbens during continuous transcutaneous vagus nerve stimulation in major depressive disorder

Abstract

Transcutaneous vagus nerve stimulation (tVNS) may be a promising treatment for major depressive disorder (MDD). In this exploratory study, fMRI scans were acquired during continuous real or sham tVNS from 41 MDD patients. Then, all patients received real or sham tVNS treatment for four weeks. We investigated the functional connectivity (FC) of the nucleus accumbens (NAc) at different frequency bands during real and sham tVNS and explored their associations with depressive symptom changes after one month of treatment. The results revealed: 1) significant positive FCs between the NAc and surrounding areas including the putamen, caudate, and distinct areas of the medial prefrontal cortex (MPFC) and the anterior cingulate cortex (ACC) during continuous real and sham tVNS; 2) compared with sham tVNS, real tVNS increased the FC between the left NAc and bilateral MPFC/rACC in the slow-5 band (0.008-0.027) and between the right NAc and left insula, occipital gyrus, and right lingual/fusiform gyrum in the typical low band (0.008-0.09); and 3) the FC of the NAc-MPFC/rACC during real tVNS showed a negative association with Hamilton Depression Rating Scale (HAMD) score changes in the real tVNS group after one month of treatment, but not in the sham group. Our findings demonstrate that tVNS can modulate low frequency intrinsic FC among key brain regions involved in reward and motivation processing and provide insights into the brain mechanism underlying tVNS treatment of MDD.

Keywords: Functional connectivity; Major depressive disorder; Neuromodulation; Nucleus accumbens; Slow 5 frequency band; Transcutaneous vagus nerve stimulation.

Conflict of interest statement

Conflict of interest

J.K has a disclosure to report (holding equity in a startup company (MNT)), but declare no conflict of interest.

Copyright © 2017. Published by Elsevier Ltd.

Figures

Figure 1

Experimental design

Figure 2

Real tVNS was applied on the concha area where there is a rich vagus nerve distribution. Sham tVNS was applied at the superior scapha (outer ear margin midpoint) where there is no vagus nerve distribution.

Figure 3

Functional connectivity results using a one-sample t-test of real tVNS and sham tVNS across different frequency bands. Slow-5 band, 0.008~0.027 Hz; slow 4 band, 0.027~0.073 Hz; typical band, 0.008~0.09 Hz. The threshold was set to voxelwise p 50. NAcL, left nucleus accumbens; NAcR, right nucleus accumbens.

Figure 4

The standard change (post minus pre-treatment divided by pre-treatment) in HAMD scores was negatively associated with the FC between the overlapped area of the left NAc (blue seed) and MPFC/rACC. (A) Left NAc seed. (B) The significantly increased FC between the left NAc and bilateral MPFC /rACC during real tVNS stimulation compared with the resting state condition. (C) Continuous stimulation of the real tVNS treatment group showed increased FC between the left NAc and bilateral MPFC/rACC compared with the sham tVNS group in the slow-5 frequency band (red). Yellow is the association map between the left NAc and bilateral OFC/MPFC/sgACC/rACC at the slow-5 frequency band. (D) Scatter plot between stimulated FC increases of the NAc-MPFC/rACC and HAMD score changes across the real tVNS group. (E) Scatter plot showed no significant relationship between the stimulated FC of the NAc-MPFC/rACC and HAMD score changes across the sham tVNS group.